THE THEORY AND TREATMENT OF DEPRESSION

TOWARDS A DYNAMIC INTERACTIONISM MODEL

FIGURES OF THE UNCONSCIOUS 5

Editorial Board

J. CORVELEYN, P. MOYAERT, PH. VAN HAUTE,

W. VER EECKE, R. BERNET
 The Theory and Treatment of Depression:
Towards a Dynamic Interactionism Model

Edited by

Jozef Corveleyn
Patrick Luyten
Sidney J. Blatt

Published jointly by

LEUVEN UNIVERSITY PRESS LAWRENCE ERLBAUM ASSOCIATES

BELGIUM MAHWAH, NEW JERSEY, LONDON
Copyright © 2005 by

Leuven University Press

Lawrence Erlbaum Associates, Inc.

All rights reserved. No part of this book may be reproduced in any form, by photostat,
microform, retrieval system, or any other means, without the prior written permission of the
publishers.

Published 2005 by

Leuven University Press

Europahuis
Blijde-Inkomststraat 5
B 3000 Leuven

ISBN 90 5867 425 8

0/2005/1869/5
NUR: 777

and

Lawrence Erlbaum Associates, Inc.

10 Industrial Avenue
Mahwah, New Jersey 0 7430

Library of Congress Cataloging-in-Publication Data

The theory and treatment of depression: towards a dynamic interactionism model 1 edited by
Jozef Corveleyn, Patrick Luyten, Sidney J. Blatt.
p.cm.
Includes bibliographical references and index.
ISBN 0-8058-5669-2 (alk. paper)
1. Depression, Mental. 2. Depression, Mental-Treatment. 1. Corveleyn, Jozef. II. Luyten,
Patrick. III. Blatt, Sidney J. (Sidney Jules), 1928-

RC537.T4782005
616.85'2706-dc22 2005040632

Design Cover: Lejon Tits

Printed in Belgium
 About the editors

Jozef Corveleyn, PhD, is Professor of Psychology, Department of Psychology,

Faculty of Psychology and Educational Sciences at the Catholic University of
Leuven (Belgium) and part-time professor, Faculty of Psychology and Educational
Sciences at the Free University of Amsterdam (The Netherlands), psychoanalyst
(Belgian School of Psychoanalysis), and clinical psychologist. He teaches courses
on psychoanalysis, psychodynamic psychology, qualitative research methodology
and clinical psychology of religion. Professor Corveleyn is also Profesor Honorario
of the Universidad de Lima (Peru) and of the Universidad Nacional Mayor de San
Marcos (Lima, Peru), and Honorary member of the Center for Psychoanalytic
Psychotherapy of Lima, Peru. His main research interests concern psychodynamic
approaches of psychosis, moral emotions and religious psychopathology.

Patrick Luyten, PhD, is Postdoctoral Fellow of the Fund for Scientific Research-
Flanders (FWO) (Belgium) at the Center for Research in Psychoanalysis and
Psychodynamic Psychology (Department of Psychology, University of Leuven,
Belgium). He is particularly interested in empirical research of psychodynamic
concepts and theories and in the interface between psychodynamic, cognitive-
behavioral, and neurobiological research. Currently, his main research interest
focuses on the relationship between personality, life stress, and depression. Other
research interests include psychotherapy research, emotion research, Chronic
Fatigue Syndrome, eating disorders, and the (clinical) psychology of religion.

Sidney J. Blatt, PhD, is Professor of Psychiatry and Psychology at Yale University

and Chief of the Psychology Section in the Department of Psychiatry at the Yale
University School of Medicine. He is also a member of the faculty of the Western
New England Institute of Psychoanalysis in New Haven, Connecticut (USA). His
primary interests are in the psychological development of mental representations,
or cognitive-affective schemas, of self and significant others; the differential
impairments of these representations in various forms of psychopathology (espe-
cially schizophrenia and depression); and the differential change in the content and
cognitive organization of these representations in the therapeutic process. In addi-
tion to his interest in mental representation in individual psychological develop-
ment, Professor Blatt is also interested in the cultural development of mental repre-
sentation, particularly the development in Western Civilization of the capacity to
represent a three-dimensional reality on a two-dimensional surface in the develop-
ment of art. Dr. Blatt is the recipient of awards for Distinguished Scientific
Contributions from The Association of Medical School Professors of Psychology
(1995); and from two divisions of the American Psychological Association -
Division of Psychoanalysis (2000) and the Division of Clinical Psychology (2004).

v
 List of contributors

Sidney J. Blatt, PhD, Departments of Psychiatry and Psychology, Yale University,

New Haven, Connecticut (USA)
Stephan .T. Claes, MD, Department of Molecular Genetics, Flanders
Interuniversity Institute for Biotechnology (VIB); Department of Psychiatry,
University of Antwerp, Antwerpen (Belgium)
Gaston Cluckers, PhD, Department of Psychology, University of Leuven, Leuven
(Belgium)
Jozef Corveleyn, PhD, Department of Psychology, University of Leuven, Leuven
(Belgium)
Jurgen De Fruyt, MD, Department of Psychiatry, St.-Jan Hospital, Brugge
(Belgium)
Koen Demyttenaere, MD, Department of Psychiatry, University Hospital
Gasthuisberg, Leuven (Belgium)
Paul Eelen, PhD, Department of Psychology, University of Leuven, Leuven
(Belgium)
Dirk Hermans, PhD, Department of Psychology, University of Leuven, Leuven
(Belgium)
Patrick Luyten, PhD, Postdoctoral Fellow of the Fund for Scientific Research-
Flanders (FWO) (Belgium), Department of Psychology, University of Leuven,
Leuven (Belgium)
Patrick Meurs, PhD, Department of Psychology, University of Leuven, Leuven
(Belgium)
Charles B. Nemeroff, MD, PhD, Department of Psychiatry, Emory University,
Atlanta, Georgia (USA)
Filip Raes, Research Assistant of the Fund for Scientific Research-Flanders (FWO)
(Belgium), Department of Psychology, University of Leuven, Leuven (Belgium)
Golan Shahar, PhD, Department of Psychiatry, Yale University School of
Medicine, New Haven, Connecticut (USA); and Department of Behavioral
Sciences, Ben Gurion University of the Negev, Israel

Lukas Van Oudenhove, MD, Research Assistant of the Fund for Scientific
Research-Flanders (FWO) (Belgium), Department of Psychiatry, University
Hospital Gasthuisberg, Leuven (Belgium)

Nicole Vliegen, Department of Psychology, University of Leuven, Leuven

(Belgium)

VII
 Contents

Foreword (Robert S. Wallerstein) 5

Introduction (Patrick Luyten, Sidney J. Blatt, & Jozef Corveleyn) 5
Chapter 1 17
The life cycle of depression (Koen Demyttenaere, Lukas Van Oudenhove,
& Jiirgen De Fruyt)
Chapter 2 43
Mood and memory: A cognitive psychology perspective on maintenance of
depressed mood and vulnerability for relapse (Dirk Hermans, Filip Raes,
& Paul Eelen)

Chapter 3 67
The convergence among psychodynamic and cognitive-behavioral theories
of depression: Theoretical overview (Patrick Luyten, Sidney J. Blatt, & Jozef
Corveleyn)
Chapter 4 95
The convergence among psychodynamic and cognitive-behavioral theories
of depression: A critical review of empirical research
(Patrick Luyten, Jozef Corveleyn, & Sidney J. Blatt)

Chapter 5 137
A dialectic model of personality development and psychopathology:
Recent contributions to understanding and treating depression
(Sidney J. Blatt and Golan Shahar)
Chapter 6 163
'Closed doors and landscapes in the mist' I.
Childhood and adolescent depression in developmental psychopathology
(Nicole Vliegen, Patrick Meurs, & Gaston Cluckers)
Chapter 7 189
'Closed doors and landscapes in the mist' 2. Depression in psychodynamic
developmental psychopathology: From single track models to complex
developmental pathways
(Patrick Meurs, Nicole Vliegen, & Gaston Cluckers)
Chapter 8 227
Corticotropin releasing factor (CRF) and major depression: Towards an
integration of psychology and neurobiology in depression research
(Stephan J. Claes and Charles B. Nemeroff)

IX
Epilogue 253
Towards integration in the theory and treatment of depression?
The time is now (Patrick Luyten, Sidney J. Blatt, & Jozef Corveleyn)

Author I Subject index 285

x
 Foreword

Robert S. Wallerstein, MD

Depression, in its various manifestations, is the chief symptom, and complaint,

that brings its sufferers into the mental health care system. In this it is the
counterpart of physical pain, which plays the same central role in propelling
individuals into the general medical care system. But what is made very clear in this
impressively comprehensive volume about the totality of the depressive phenomena
and the depressive experience, from the most subtle subclinical dispositions to the
most profoundly crippling and even life-threatening illness pictures, is that we have
been living in a world of vastly insufficient and even misleading understandings of
this major symptom and character complex, as well as of insufficient awareness of
the conceptual, diagnostic, and therapeutic implications of the chronic, recurrent,
and often progressive, nature of so much depressive illness.
An implicit message of this volume is that the chief basis for these
misunderstandings and skewed conceptualizations has been the dominance over
these past several decades of the a-theoretical and a-etiological framework imposed
upon diagnosis and therapy recommendations within the realm of mental and
emotional disorders by the American Psychiatric Association's Diagnostic and
Statistical Manual of Mental Disorders (DSM). Valuable as that compendium has
been for categorization for statistical (epidemiological) purposes, and for reliability
and comparability in empirical research study, DSM has, by its insistence on
precision in compartmentalized categorization of distinct entities, as created by
number counts and intensities of symptom clusters, but devoid of meanings or
conditions of onset and development in the individual patients, has led, by this
insistence, to grossly inadequate understandings of depression, and in many ways
has rendered more difficult the kind of study and research needed to truly enhance
our knowledge of the conditions of onset and development, and the most effective
treatment of depression, in all its variant forms.
What this volume so convincingly demonstrates by its multi-pronged and
dynamic interactional consideration of epidemiological, psychodynamic (psycho-
analytic), cognitive-behavioral, developmental, psychopathological and neurobio-
logical factors, is that depression, in its predispositions and in its various overt
expressions, is profoundly dimensional, not categorical, in nature. Its various
features and manifestations are intertwined and mutually interactive, not
orthogonally separated as distinctive entities. And it is above all the outcome of a
long developmental process, in which in-built (genetic) vulnerabilities and quality
of life circumstance, most profoundly in the earliest mother-infant caregiver
relationship, but also throughout the lifetime developmental trajectory, further
accentuated by the traumata, losses and disappointments that can impact any life,
all lead together in dialectically and recursively interacting ways to the significant
distortions and deviations in the normal developmental process. These distortions
can be expressed - even early in life, if severe enough - as a depressive personality
disposition, or as mild or transient depressive symptomatic episodes in response to
readily identifiable life crisis and distress (such as loss, illness, work failures or
disappoint-ments, etc.). Alternatively, these distortions can also be expressed as
more severe, and at times, recurrent, chronic, and progressive severe-enough
depressive illness outbreaks, often of increasing intensity and diminishing interval
states, or all the way to what have been classically designated as major unipolar or
bipolar de-pressions. In later years, involutional melancholias may result, with
profound disruptions in life functioning, necessary hospitalizations, and marked
suicidal propensities that need to be vigorously guarded against.
A major theoretical organizing framework for this volume has been provided by
the work of Sidney Blatt of Yale University, one of the co-editors, as well as one of
the co-authors of chapters in this volume, and has been further developed and then
brought together in all its ramifying and mutually interacting expressions by a
gifted and dedicated group of clinicians and researchers headed by Professor Jozef
Corveleyn of the University of Leuven in Belgium, chief editor and also co-author
of chapters in this volume, with most but not all contributing authors members, of
his strongly psychoanalytically-informed research group at the Department of
Psychology, including his chief collaborator in this overall enterprise, Patrick
Luyten, the third co-editor and co-author. The very persuasive conceptual frame-
work, elaborated and brought together by this whole cooperating group, is based on
that developed originally by Blatt and his many collaborators at Yale, over an entire
professional lifetime, and recently brought together in comprehensive and compel-
ling exposition in Blatt's own 2004 magisterial volume, Experiences ofDepression:
Theoretical, Clinical, and Research Perspectives.
What is the originating essence of this elaborated organizing framework? It is
simply and fundamentally that clinical depression and/or the depressive character
formation do not represent a unitary, coherent, illness and character pattern. Rather,
both out of his clinical psychoanalytic experience, and his thirty years of empirical
research study (along with an array of gifted colleagues), Blatt separated out
two significantly different depressive diatheses, designated by him anaclitic and
introjective, differing in their character attributes and in their presenting
symptomatology, differing in the formative early life experiences that create their
predispositions, and differing in the implications for prognosis and for psychothera-
peutic course.

2
 The anaclitic (or dependent) depression is marked by interpersonal issues of care
and connectedness and relatedness. A central fear is of abandonment and loss, and
of a state of helplessness as a consequence. There is a desperate need to maintain
supporting and nurturant links, and to ward off feelings of being unwanted and
unloved. At its life threatening extremes in infancy and toddlerhood, it can be
manifested as hospitalism and marasmus, as has been chronicled by Spitz and the
Robertsons. And it occurs more frequently in females, who are declared to be more
oriented to issues of relatedness than are males. Blatt locates the infantile proto-
types of this disorder in the deformative experiences of the earliest, oral, develop-
mental stage.
Contrariwise, the introjective (self-critical) depression is marked by issues of
self-definition, separateness, and autonomy, with often a coercive characterological
perfectionism. The central concerns are with self-criticism, self-worth, and self-
doubt. There is a profound fear of failure to measure up to one's own exalted
standards (internally felt, or presumably externally imposed), with consequent guilt
and either self blame or blame aggressively projected onto one's most important
objects. In the Engel-Schmale formulations, this engenders a feeling of hopeless-
ness. This species of depression occurs more frequently in males, who are declared
to be more oriented to issues of self-definition, ambition, and achievement. And it
is located by Blatt at a higher developmental level, built around the deviant
experiences of Mahler's separation-individuation phase, and then the superego
issues of guilt and blame.
Given this originating theoretical frame, the overall thrust of this volume by
Corveleyn and his many colleagues is not to explore further the many similarities
and differences between these two variant forms of depression, nor the linkages to
intermediate states which partake of both and border on both, but rather to explore
them as alternative deviations from the normal developmental path with, in each
instance, an abnormal preoccupation with the issues and conflicts of one or the
other side of this dichotomy (and possibly a concomitant excessive avoidance of, or
defense against, the issues and conflicts of the other side), and then to set all of
these etiological and developmental considerations that can give rise to the variant
expressions of the depressive experience and illness, conceptualized thus
psychoanalytically, into conjunction with, and in dialectical interaction with, all the
other realms of accruing knowledge of the depressive condition already mentioned,
the epidemiological, the cognitive-behavioral, the developmental, the psychopatho-
logical, and the neurobiological - each of these developed very comprehensively
and convincingly, in each instance by a chapter co-authored by experts in that arena,
mostly Belgian co-authors, but also with two cooperating American workers
beyond Blatt, Golan Shahar at Yale (now at Ben Gurion University in Israel), and
Charles Nemeroff at Emory University in Atlanta.
The treat awaiting the reader of this volume is to witness the unfolding interplay
of these multiple considerations of the many-sided complexity of the totality of the
depressive experience, as illuminated from these many disparate vantage points,
variously clinical or empirical, psychosocial or neurobiological, nomothetic or
idiosyncratic, etc., as they are all mutually interacting, and at each moment exist at
various points of convergence or divergence. Examples abound and recur, through-

3
out the text, for example, the various convergences seen among psychological
theoretical perspectives grounded in seemingly antithetical assumptions about the
nature of mental activity, like the quite classically psychoanalytic perspectives of
Blatt, the behavioral theory perspectives of Beck, the attachment formulations of
Bowlby, and the interpersonal propositions of Arieti and Bemporad; or like the
inner-outer interplay of a depressive character disposition (itself an outcome of the
interplay of genetic vulnerabilities and adverse developmental life experience) and
its "depressogenic" impact in shaping continuing environmental contingency in
ways that reinforce the depressive outlook and confirm the depressive character
expectation.
Perhaps the best conclusion to this foretelling to the readers is to quote from the
very last paragraph of this volume. Overall, "an etiologically-based, dynamic
interactionism view of depression, emphasizing interactions among genetics, early
adversity, current life stress, and relatively stable cognitive-affective schernas or
personality dimensions, emerges as a model that may facilitate the integration of
various theoretical, methodological, and clinical approaches to depression. At the
same time, much work remains to be done." The entire volume propounds very
convincingly all the evidence that buttresses this conclusion, all the work done by
so many that has brought us to this point, and that points enticingly to all the many
opened avenues of continuing clinical and research endeavor awaiting our
collective attention. As such, it is a tribute to this group of primarily Belgian
clinicians and researchers who have collaboratively written it. Reading it should be
a journey of pleasure and of widening perspective. Bon voyage!

4
 Introduction

Patrick Luyten, Sidney l. Blatt, & Jozef Corveleyn

Depression: Changing Views

Depression, with a lifetime prevalence of approximately 15% (Blazer, Kessler,

McGonagle, & Swartz, 1994), is one of the most prevalent disorders worldwide
(NIMH, 2001). By the year 2020, depression is expected to be the second most
serious disorder with respect to global disease burden (Murray & Lopez, 1996). The
personal and social costs associated with this disorder are immense. Depression
seriously affects both intrapersonal and interpersonal functioning (Beach, 2000,
not only for those suffering from depression, but also for those in their immediate
environment. Moreover, these negative effects are not limited to an episode of
depression, but may extend over years, as is for instance shown by the fact that
children of depressed parents are at elevated risk for psychopathology in later life
(Blatt & Homann, 1992; Goodman & Gotlib, 2002). In addition, depression has
also a serious economic cost. For instance, the National Institute of Mental Health
estimated the total cost of depression in the U.S.A. to be $30-44 billion annually
(NIMH, 1999).
Until recently, it was believed that depression, though prevalent, was a relatively
"benign" disorder, because it was thought to be associated with a good prognosis,
even when untreated. Depression therefore was often called the "common cold" of
psychopathology. However, over the past two decades, our view on the natural
course and the treatment of depression has dramatically changed (Costello et aI.,
2002; Hollon et aI., 2002). Research concerning the natural course of depression
has made increasingly clear that depression is a recurrent and for a considerable
number of patients even a chronic disorder (Frank et aI., 2002; Segal, Pearson, &
Thase, 2003). Recent estimates suggest relapse rates after a first episode of 20-30%
within 3 years, and 70-80% within the same period in subjects who have had three
depressive episodes or more (Judd, 1997; Segal et al., 2003; Solomon et al., 20(0).
The risk for at least one other episode after a first episode is estimated at almost
90% (Kupfer & Frank, 2001), and the average depressed patient, during their life
time, will experience four episodes, each of about 20 weeks duration (Judd, 1997).
Also, research has shown that with each new episode, the time between two next
episodes shortens (Solomon et aI., 2000). Moreover, in about 10% to 30% of the
cases, depression becomes chronic (Verheul, 2003). Finally, adding to the growing
realization that depression is not a relatively benign and isolated disorder, research
indicates that comorbidity of depression with other Axis I and especially Axis II
disorders is the rule rather than the exception. Estimates of comorbitity between
depression and personality disorders, for instance, vary in psychiatric populations
between 50-60% (Klein & Hayden, 2000; Mulder, 2002). And it is now well
established that comorbid Axis I and Axis II disorders have a negative impact on
the prognosis and treatment of depression (Mulder, 2002; Westen, Novotny, &
Thompson-Brenner, 2004).

This leads us to a second domain in which our view of depression has

dramatically changed. The treatment of depression by both psychotherapeutic and
pharmacological means has proved much more difficult and less successful then
once was hoped for. In the 1980's, many researchers and clinicians believed that
depression could be effectively treated with short-term (standardiz.ed)
psychotherapeutic treatments, such as Cognitive-Behavioral Therapy (CBT) or
Interpersonal Therapy (IPT), and/or pharmacological treatment (e.g., see Dobson,
1989; APA, 2000).
This optimism has been seriously tempered. A growing body of research has
shown important limitations of these brief treatments for many depressed patients
(Hollon et al., 2002; Luyten, Lowyck, & Corveleyn, 2003; Parker, Roy, & Eyers,
2003; Rush & Thase, 2002; Westen & Morrison, 2001; Westen et aI., 2004). For
instance, in what is probably the most prestigious and extensive Randomised
Clinical Trial (RCT) to date, the National Institute of Mental Health (NIMH)-
sponsored Treatment of Depression Collaborative Research Program (TDCRP;
Elkin, Parloff, Hadley, & Autry, 1985), only about half of the patients that received
16 weeks of treatment with either Cognitive-Behavioral Therapy (CBT),
Interpersonal Therapy (lPT), or antidepressant medication (Imipramine plus
Clinical Management; IMI-CM), met criteria for remission at termination (Elkin et
aI., 1989) and only about 20% of the patients were considered as fully recovered
(remission without relapse) at a follow-up assessment conducted 18-months after
the termination of treatment (Shea et aI., 1992).
Westen and Morrison (2001) estimated from well-designed randomized clinical
trials (RCTs) of so-called Empirically Supported Treatments (ESTs) of depression,
such as Cognitive-Behavioral Therapy (CBT) and Interpersonal Therapy (lPT), that
only about 50% of depressed patients who completed such treatments showed
improvement. For the Intent-To-Treat (ITT) sample, this figure was only 37%.
Moreover, because of high exclusion rates in outcome studies, patients in routine
practice might show even less response to these treatments. Congruent with this

6
hypothesis, Westen and Morrison (200 I) found that there was a high correlation
between outcome and exclusion criteria (r=.4I), indicating that the more stringent
the exclusion criteria (the greater the exclusion rate), the more likely it was for the
study to have a greater number of patients who showed improvement. In addition,
it is important to note that the average patient who shows improvement after these
brief treatments, remains symptomatic and thus does not show full recovery
(Westen & Morrison, 200 I; see also Elkin et al., 1989; Shea et al., 1992).
Besides these modest effects of short-term treatments at termination, there is also
a lack of controlled studies on the long-term effects of ESTs of depression. Westen
and Morrison (2001), for instance, could locate only 4 studies published in the
1990's in which the long-term effects (from 12 to 18 months) of ESTs of depression
were investigated. The average percentage of patients showing no relapse after
12-18 months was 36.6%. For instance, in the NIMH TDCRP-study, at follow-up
after 18 months, only 20% of the patients were fully recovered (Blatt, Zuroff,
Bondi, & Sanislow, 2000; Shea et al., 1992). Westen and Morrison (2001) could
only locate one study with follow-up data of 2 years or more. The results of this
study demonstrated that as little as 27% of the patients who were in treatment and
8% of all patients initially screened, showed no relapse. Westen and Morrison
(2001, p. 886) rightfully conclude from these and similar findings: "By any
standards, it is difficult to construe these data as evidence for the hypothesis that
these treatments show genuine efficacy for the treatment of depressive disorders".
Similar response rates of about 50% (compared to 30-35% placebo response)
have been reported for the pharmacotherapeutic treatment of depression (e.g.,
Williams et al., 2002). More critical reviews, which also include non-published
negative studies, have even claimed that drug-placebo differences are minimal
and "of questionable clinical significance" (Kirsch & Sapirstein, 1998; Kirsch,
Moore, Scoboria, & Nicholls, 2002; see also Khan & Khan, 2003). In addition,
a recent meta-analysis has shown that approximately 40 years of research on
drug treatment of depression has only led to progression concerning the side effects
of medication, but not concerning its efficacy (Barbui & Hotopf, 200 I). Thus,
so-called "older" antidepressants are at least as effective as "newer" ones, but
they show more side effects. And although research has shown that the long-
term treatment of depression with antidepressants leads to a significant reduction
in relapse rates (e.g., see Kupfer & Frank, 2001), there is no relationship be-
tween the duration of drug treatment and the probability of relapse after
discontinuation of the drug (Viguera, Baldessarini, & Friedberg, 1998; see also
Fava, 2002).

The Search for More Effective Treatments

and the Need for Integration in Research on Depression

The findings just reviewed clearly suggest that further research on depression, its
origins, the factors influencing its course, and particularly its treatment, are

7
important future tasks from a scientific, clinical, as well as from a national mental
health perspective. We clearly have an important responsibility to find better
treatments for this disabling disorder, and also to develop primary and secondary
prevention strategies. Moreover, these prevention and treatment strategies need to
be made more available to the public because research has shown that very few
depressed persons - estimates vary between 19-37% - ever seek treatment (Grote
& Frank, 2003). In addition, the modal patient who does seek treatment, either is
not treated or inadequately treated or drops out early from treatment (Grote &
Frank,2003).
Not surprisingly, these findings concerning the recurrent and often chronic
course of depression, the high rates of comorbidity of depression with personality
disorders, and the growing awareness of the limits of brief treatments of depression,
have led to a renewal of interest in theories and techniques that are based on long-
term treatments of depression, such as psychodynamic theories (Jones & Pulos,
1993; Kwon, 1999; Shapiro et al., 1994), and especially in theoretical models that
focus on the relationship between personality and depression (Gunderson,
Triebwasser, Philips, & Sullivan, 1999; Klein, Kupfer, & Shea, 1993; Kupfer &
Frank, 2001; Verheul, 2003).
The convergence among psychodynamic and more recent cognitive-behavioral
theories of depression plays an important role in this evolution (Blatt, 2004; Blatt
& Maroudas, 1992; Robins, 1993). Most research in this domain has concentrated
on the psychodynamic conceptions of Sidney J. Blatt (Yale University) and the
cognitive-behavioral formulations of Aaron T. Beck (University of Pennsylvania).
Both Blatt (e.g., Blatt, 1974, 1998,2004; Blatt, D'Afflitti, & Quinlan, 1976; Blatt,
Quinlan, Chevron, McDonald, & Zuroff, 1982) and Beck (e.g., Beck, 1983, 1999)
have proposed that two personality dimensions, i.e., interpersonal dependency or
sociotropy and self-critical perfectionism or autonomy, are vulnerability factors for
clinical and nonclinical forms of depression. According to Blatt and Beck, these
personality dimensions are associated with different personality structures, a
different relational and attachment style, a vulnerability for specific life events (loss
vs. failure), a different clinical presentation, and a different response to psycho-
therapy and pharmacotherapy. Moreover, Blatt has proposed that these personality
dimensions are also associated with different developmental factors and biological
vulnerabilities (Blatt & Homann, 1992; Blatt, Cornell, & Eshkol, 1993). Support
for these formulations has come from three decades of empirical research (Blatt,
2004; Blatt & Zuroff, 1992; Clark & Beck, 1999). Yet, this research is often little
known, even among experts in research on depression.
This is only one example that shows that the field of research on depression is
still relatively fragmented. Different theoretical and research traditions concerning
depression have often developed separately from each other. The time seems
overdue to bring these different views together with the aim of developing a more
comprehensive theoretical framework for depression. It is our firm conviction that
only then more effective treatments of this serious and disabling disorder can be
developed and implemented. This has also been recognized recently by the National
Institute of Mental Health Strategic Plan for Mood Disorders Research (NIMH,
2003). This Strategic Plan, which will guide the NIMH's research initiatives in the

8
coming years, identified several routes and barriers towards integration among
approaches towards depression with the aim of enhancing both theoretical insights
in depression and treatment strategies. Although this initiative from the NIMH
deserves utmost praise, in our view it misses out on some important developments
in mood disorders research (Luyten et aI., 2003). This is in part due to the
fragmented state of research in this field, but also to different theoretical and
research traditions within the U.S.A., and between the U.S.A. and other parts of the
world, especially in Europe.

This led us to the idea of organizing a conference on depression, inviting several

scholars from a wide variety of different fields in depression research. Professor
Sidney J. Blatt (Yale University, New Haven, USA), who was, at that time, Visiting
Professor at the University of Leuven (Belgium), was invited as the main speaker.
The aim of this conference was not only to present a "state of the art" of depression
research, but also, and perhaps more importantly, to formulate possible avenues for
integration between different approaches of depression. In addition, it seemed
crucial to identify important barriers towards integration among various
psychological and biological approaches of depression. Subsequently, on March
14, 2003, a conference on the integration of psychological and biological
approaches towards depression was held at the St. Jozef University Psychiatric
Clinic in Kortenberg (Leuven), Belgium.
The chapters in this book are based on the presentations at this conference.
However, most chapters have been rewritten based on the interactions and
discussion among the participants. In preparing their chapter for this volume, we
asked the authors (I) to review the current state of research in their area and (2) to
also identify possible routes and barriers towards the integration among different
approaches in depression research. In the epilogue, we have made an attempt at an
"integration of the integration", summarizing possibilities and barriers towards
integration, spelling out future lines of investigation, and discussing clinical
implications of the recent research on depression from the various perspectives
presented in this volume.

Overview of the Book

The first chapter by Demyttenaere, Van Oudenhove and De Fruyt sets the stage
for this volume by summarizing and discussing recent epidemiological studies,
including the recent European Study on Mental Disorders (ESEMeD), which have
dramatically changed our view of depression. Among the many interesting findings
they discuss, two seem particularly important. First, recent epidemiological studies
do not support the categorical view of depression promulgated by the Diagnostic

9
and Statistic Manual for Mental Disorders (DSM; APA, 1994). Instead, a
dimensional view appears to more adequately fit to the data. Second, these
epidemiological studies also clearly show that depression is not an isolated, state-
like disorder, but a recurrent disorder that becomes chronic in a considerable
amount of patients, and one that can lead to considerable and lasting changes in
functioning, even on the neurobiological level.
In the second chapter, Hermans, Raes and Eelen review recent developments in
cognitive behavioral theory and research on depression. Their contribution is
noteworthy in many respects, including their emphasis on the need for more
experimental research to inform research on depression. In this context, they
emphasize the relevance of cognitive psychological research on memory in general
and on autobiographical memory in particular for understanding depression. In
addition, this chapter is also an excellent example of how fundamental experimental
research can inform clinical practice. Hermans and colleagues, for example,
demonstrate how research on mood congruent encoding and recall might help
understand the vicious cycles of negative thoughts typical of depressed patients.
Equally important is their emphasis on relapse prevention. They note that the fact
that depression tends to be a recurrent disorder should lead to a focus in both
research and treatment on the prevention of future relapses. In this context,
Hermans and colleagues discuss the integration between cognitive research on
(autobiographical) memory and more recent cognitive-behavioral treatments for
depression such as Mindfulness-Based Cognitive Therapy (MBCT; Segal,
Williams, & Teasdale, 2002).
The next three chapters are devoted to the convergence among psychodynamic
and cognitive-behavioral theories of depression. In Chapter 3, Luyten, Blatt and
Corveleyn provide an overview of the convergence in psychodynamic and
cognitive-behavioral theorizing concerning depression. In particular, they focus on
the conceptualizations of Sidney J. Blatt and Aaron T. Beck concerning
Dependency/Sociotropy and Self-Critical Perfectionism/Autonomy as primary
vulnerability factors for depression. As this overview shows, there are many
similarities and thus much common ground between psychodynamic and cognitive-
behavioral theories of depression. However, at the same time, Luyten and
colleagues note that there are also some important barriers towards further
integration because of major differences in the underlying view of human nature in
these two models.
In Chapter 4, Luyten, Corveleyn and Blatt review empirical research concerning
the three most central assumptions of the theories of Blatt and Beck as well as
research concerning the clinical implications of these theories. The review in this
chapter demonstrates that the central assumptions of Blatt and of Beck have
received considerable empirical support. However, Luyten and colleagues also
identify several areas for further research. Their review also shows that Blatt's and
Beck's views have not only led to considerable integration between psychodynamic
and cognitive-behavioral formulations, but also between psychodynamic and
cognitive-behavioral theories and research in the fields of personality, social
psychology, developmental psychology, and neurobiology, clearly illustrating the
vitality of these theoretical models in generating integrative research.

10
 Subsequently, in Chapter 5, Blatt and Shahar address recent developments in
research on Dependency/Sociotropy and Self-Critical Perfectionism/Autonomy. In
particular, Blatt and Shahar argue that depression should be situated within a
general theory of personality development. Depression should not be viewed
categorically as a disease, but as a distortion of normal personality development. In
addition, a dynamic interactionism model is proposed that involves reciprocal
interactions between adaptive and maladaptive dimensions of self-definition
(identity) and interpersonal relatedness, life stress, social support, and depression.
According to this dynamic interactionism model, individuals are not just the passive
recipients of their stressful environment, but actively, though often unwittingly,
interpret and generate, in part, their own (social) environment. Finally, Blatt and
Shahar also discuss recent research exploring the dynamics of patients showing
mixed dependent and self-critical characteristics.
Whereas until recently many even doubted whether children could experience
depression, today the question is rather how many adults have their first onset of
depression in childhood (Costello et aI., 2002). Developmental research on
depression is therefore a most important task for the future. In Chapter 6, Vliegen,
Meurs and Cluckers review research in developmental psychopathology on
depression in childhood and adolescence. This research has not only shown that
depression in childhood and adolescence is not uncommon, but that developmental
factors appear to play an important role in the etiology and pathogenesis of
depression. Moreover, Vliegen, Meurs and Cluckers also stress the need for broad,
comprehensive, developmentally-based theories of depression in childhood and
adolescence. Such comprehensive theories are needed to develop prevention
programs and treatments strategies that are developmentally sensitive. The next
chapter, Chapter 7, provides the outlines for such an overarching, psychodyna-
mically inspired developmental psychopathological framework to understand the
origins of depression. In particular, in this chapter, Meurs, Vliegen and Cluckers
delineate several developmental tasks and stages that might be involved in the
development of depression. Their views are illustrated by a detailed discussion of
excerpts of the treatment of a depressed boy.
Chapter 8 by Claes and Nemeroff addresses what can be considered to be the
greatest future challenge for research on depression and perhaps psychopathology
in general, namely the integration between biological and psychological
approaches. This integration is of high relevance for both researchers and clinicians
alike. From a research perspective, with each new study, it becomes increasingly
clear that genes, neurobiological processes, psychological, and environmental
factors constantly interact, and that any theory that assumes a neat distinction
between "nature" and "nurture" is incapable of capturing these intrinsic and
recursive interactions between biological, psychological, and social factors.
Clinicians, in turn, are faced with difficult questions such as the decision on
whether and how they should combine pharmaceutical and psychotherapeutic
treatments of depression. At present, no theoretical rationale that is firmly grounded
in empirical research exists for such decisions. Hence, the need for integrative
research and theories is clear and urgent. According to Claes and Nemeroff, such
integrative research has already begun in the domain of neurobiological research on

11
stress. In particular, they argue that this area of research could provide an important
framework for bridging the gap between psychological and biological theories of
depression. However, they also point to the tremendous problems associated with
this research. Not only is such research very costly and time-consuming, it also
assumes the existence of interdisciplinary research teams, consisting of geneticists,
neurobiologists, psychiatrists, and psychologists.
This volume concludes with an epilogue in which we have tried to answer three
basic sets of questions. Our first set of questions concerns identifying the current
possibilities and barriers to the integration between psychological and biological
approaches of depression. Is there sufficient "common ground" between these
approaches? If so, can we determine more precisely what these approaches have in
common? And what are important barriers towards such integration? Our second
set of questions is even more challenging. Taking into account existing possibilities
as well as barriers to integration among various approaches of depression, can we
delineate important research tasks for the future that could foster further dialogue
and enhance integration between psychological and biological approaches of
depression? With the third and final set of questions, we returned to our starting
point. How can "old" and "new" knowledge that has accumulated over the years
concerning depression be used to improve our treatment of patients suffering from
this disorder? In other words: what are the clinical implications? Our answers to
these three sets of questions led us to propose an etiologically-based,
transdiagnostic dynamic interactionism model as an alternative for the DSM view
of depression.

When we embarked on the adventure of writing this book, we were already

convinced that we had much to learn about depression. Yet, while working on this
book, we came to realize that our knowledge of depression and its treatment is even
more limited than we initially assumed. It has therefore not only been a pleasure,
but also a privilege to work with all of the authors that have contributed to this book,
and with all those who participated in the discussions on the conference in
Kortenberg (Leuven, Belgium) that gave the impetus for this volume. Their
expertise and openness towards perspectives often very different from their own are
exemplary. In addition, although they sometimes must have tired of our continuous
questions, they never failed to provide us with what we asked from them, often
within very short time spans.
It is also a pleasure to thank those people who have contributed to our
preparation of this book. First of all, we want to thank all our colleagues - too many
to list individually - that have either directly or indirectly contributed to this
volume. We are indebted to Benedicte Lowyck, Catherine Estas, Bart Soenens (all
from the Department of Psychology, University of Leuven), and Raymond
Paloutzian (Westmont College, California), who read portions of the manuscript for
this book and provided us with many insightful suggestions. We are also indebted
to Marleen Devijver for her dedication in preparing this manuscript, to Hilde Lens,
our publisher at Leuven University Press, and to Susan Milmoe and Kristen Depken

12
at Lawrence Erlbaum Associates, for their expertise and patience. In addition, we
wish to thank Larry Erlbaum, for his interest, support, and generosity, and Barbara
Lavrysen for compiling the author and subject index.
This volume was made possible, in part, by a grant to Patrick Luyten and Jozef
Corveleyn from the Research Advisory Board (RAB) of the International
Psychoanalytic Association (lPA). We gratefully acknowledge their support. We
also want to express our gratitude to Peter Fonagy, Drew Westen, John Clarkin,
Linda Mayes, Manfred Beutel, Marianne Leuzinger-Bohleber, George Gergely, and
Rudi Vermote for their support and encouragement. Special thanks to Bernard
Sabbe, Sieglinde Meganck, Bart Jansen, Carmen De Grave, Marek Swinnen, and
the whole team from the Psychiatric Center St. Norbertushuis in Duffel (Belgium),
for their collaboration and friendship over the years. We also thank the Director of
the St. Jozef University Psychiatric Clinic in Kortenberg (Leuven), Jozef Peuskens,
for hosting the conference on depression that led to this volume. And, finally, we
want to thank our patients, because they constantly remind us that we have still
much to learn about the nature of depression.

References

American Psychiatric Association (1994). Diagnostic and statistical manual of mental disorders (4th ed.).
Washington. DC: Author.
American Psychiatric Association (2000). Practice guidelines for the treatment of patients with major
depression. Retrieved September 3, 2004, from https://www.psych.org/psych_practitreatg/pglDepres
sion2e.book.cfm
Barbui, C; & Hotopf, M. (200 I). Amitriptyline v. the rest: Still the leading antidepressant after 40 years of
randomised controlled trials. British Journal of Psychiatry, 178, 129-144.
Beach. S. R. H. (Ed.). (2001). Marital and family processes in depression: A scientific foundation for
clinical practice. Washington, DC: American Psychological Association.
Beck. A. T. (1983). Cognitive therapy of depression: New perspectives. In P. 1. Clayton & J. E. Barrett
(Eds.), Treatment (!( depression: Old controversies and new approaches (pp. 265-290). New York:
Raven Press.
Beck. A. T. (1999). Cognitive aspects of personality disorders and their relation to syndromal disorders: A
psychoevolutionary approach. In C. R. Cloninger (Ed.), Personality and psychopathology (pp. 411-
429). Washington, DC/London: American Psychiatric Press.
Blatt. S. J. (1974). Levels of object representation in anaclitic and introjective depression. The Psycho-
analytic Study of the Child. 29. 107-157.
Blatt. S. J. (1998). Contributions of psychoanalysis to the understanding and treatment of depression.
Journal of the American Psychoanalytic Association, 46, 722-752.
Blatt, S. J. (2004). Experiences of depression: Theoretical, clinical and research perspectives. Washington,
DC: American Psychological Association.
Blatt. S. J., Cornell, C. E., & Eshkol, E. (1993). Personality style, differential vulnerability and
clinical course in immunological and cardiovascular disease. Clinical Psychology Review, fJ,
421-450.
Blatt. S. J., D'Afflitti, J. P., & Quinlan. D. M. (1976). Experiences of depression in normal young adults.
Journal ofAbnormal Psychology, 85, 383-389.
Blatt. S. J., & Homann, E. (1992). Parent-child interaction in the etiology of dependent and self-critical
depression. Clinical Psychology Review, /2,47-91.
Blatt, S. J., & Maroudas, C. (1992). Convergence among psychoanalytic and cognitive-behavioral theories
of depression. Psychoanalytic Psychology, 9, 157-190.

13
Blatt, S. J., Quinlan, D. M., Chevron, E. S., McDonald, c.. & Zuroff, D. C. (1982). Dependency and self-
criticism: Psychological dimensions of depression. Journal of Consulting and Clinical Psychology, 63,
125-132.
Blatt, S. J., & Zuroff, D. C. (1992). Interpersonal relatedness and self-definition: Two prototypes for
depression. Clinical Psychology Review, 12,527-562.
Blatt, S. 1., Zuroff, D. C; Bondi, C. M., & Sanislow III, C. A. (2000). Short- and long-term effects of
medication and psychotherapy in the brief treatment of depression: Further analyses of data from the
NIMH TDCRP. Psychotherapy Research, 10, 215-234.
Blazer, D. G., Kessler, R. c, McGonagle, K. A., & Swartz, M. S. (1994). The prevalence and distribution
of major depression in a national community sample: The National Comorbidity Survey. American
Journal of Psychiatry, 151, 979-986.
Clark, D. A., & Beck, A. T (1999). Scientific foundations of cognitive theory and therapy of depression.
New York: John Wiley & Sons.
Costello, E. 1., Pine, D. S., Hammen, c, March, 1. S., Plotsky, P. M., Weissman, M. M., Biederman, J.,
Goldsmith, H. H., Kaufman, J.. Lewinsohn, P. M., Hellander, M., Hoagwood, K., Koretz, D. S., Ne.son,
C. A., & Leckman, J. F. (2002). Development and natural history of mood disorders. Biological
Psychiatry, 52, 529-542.
Dobson, K. S. (1989). A meta-analysis of the efficacy of cognitive therapy for depression. Journal of
Consulting and Clinical Psychology, 57. 414-419.
Elkin, I., Parloff, M. 8., Hadley, S. w.. & Autry, 1. H. (1985). NIMH Treatment of Depression
Collaborative Research Program. Archives of General Psychiatry, 42,305-316.
Elkin, I., Shea, M. T, Watkins, J. T, Imber, S. D., Sotsky, S. M., Collins, J. E, Glass, D. R., Pilkonis, P. A.,
Leber, W. R., Docherty, J. P., Fiester, S. J., & Parloff, M. B. (1989). NIMH treatment of depression
collaborative research program: General effectiveness of treatments. Archives of General Psychiatry,
46,971-983.
Fava, G. A. (2002). Long-term treatment with antidepressant drugs: The spectacular achievements of
propaganda. Psychotherapy and Psychosomatics, 71, 127-132.
Frank, E., Rush, A. J., Blehar. M., Essock, S., Hargreaves, W., Hogan, M., Jarrett, R., Johnson, R. L.,
Katon, W. J., Lavori, P., McNulty. J. P., Niederehe, G., Ryan, N., Stuart, G., Thomas, S. B., Tollefson,
G. D., & Vitiello, B. (2002). Skating to where the puck is going to be: A plan for clinical trials and
translation research in mood disorders. Biological Psychiatry, 52, 631-654.
Goodman, S., & Gotlib, I. (2002). Children of depressed parents. Mechanisms of risk and implication 5 [or
treatment. Washington, DC: American Psychological Association.
Grote, N. K., & Frank, E. (2003). Difficult-to-treat depression: The role of contexts and comorbidities.
Biological Psychiatry. 53. 660-670.
Gunderson, J. G., Triebwasser, J., Philips. K. A., & Sullivan, C. N. (1999). Personality and vulnerability to
affective disorders. In C. R. Cloninger (Ed.), Personality and psychopathology (pp, 3-32). Washington.
DClLondon: American Psychiatric Press.
Hollon, S. D., Munoz, R. F.. Barlow, D. H., Beardslee, W. R., Bell, C. c.. Bernal, G., Clarke, G. N..
Franciosi, L. P., Kazdin, A. E., Kohn, L.. Linehan, M. M., Markowitz, J. C, Miklowitz, D. J., Persons,
J. 8., Niederehe, G., & Sommers, D. (2002). Psychosocial intervention development for the prevention
and treatment of depression: Promoting innovation and increasing access. Biological Psychiatry. 52,
610-630.
Jones, E. E., & Pulos, S. M. (1993). Comparing the process of psychodynamic and cognitive-behavoral
therapies. Journal of Consulting and Clinical Psychology, 61, 306-316.
Judd, L. J. (1997). The clinical course of unipolar major depressive disorders. Archives of General
Psychiatry, 54. 989-991.
Khan, A., & Khan, S. (2003). Placebo in mood disorders: The tail that wags the dog. Current Opinion in
Psychiatry, 16, 35-39.
Kirsch, I., Moore, T 1., Scoboria, A., & Nicholls, S. S. (2002). The emperor's new drugs: An analysis of
antidepressant medication data submitted to the U.S. Food and Drug Administration. Prevention &
Treatment, 5, Article 23. Retrieved September 3, 2004, from https://journals.apa.org/prevemion/
volume5/pre0050023a.html
Kirsch, I., & Sapirstein, G. (1998). Listening to Prozac but hearing placebo: A meta-analysis of
antidepressant medication. Prevention & Treatment, 1, Article oo02a. Retrieved September 3, 2D04,
from https://journals.apa.org/prevention/volumel/preooloo02a.html
Klein, D. N., & Hayden, E. P. (2000). Dysthymic disorder: Current status and future directions. Current
Opinion in Psychiatry, 13,171-177.

14
Klein, M. H., Kupfer, D. 1., & Shea, M. T. (1993). Personality and depression. A current view. New
YorkfLondon: The Guilford Press.
Kupfer, D. 1., & Frank, E. (2001). The interaction of drug- and psychotherapy in the long-term treatment
of depression. Journal ofAffective Disorders, 62, 131-137.
Kwon, P. (1999). Attributional style and psychodynamic defense mechanisms: Toward an integrative model
of depression. Journal of Personality, 67, 645-658.
Luyten, P., Lowyck, B., & Corveleyn, J. (2003). Teoria y tratamiento de la depression: Hacia su
integracion? [Theory and treatment of depression: Towards integration?). Persona, 6, 81-97.
Mulder, R. T. (2002). Personality pathology and treatment outcome in major depression: A review.
American Journal of Psychiatry, 159, 359-371.
Murray, C. J. L., & Lopez, A. D. (Eds.) (1996). The global burden ofdisease: A comprehensive assessment
of mortality and disability from diseases, injuries and risk factors in 1990 and projected to 2020.
Cambridge, MA: Harvard University Press.
National Institute of Mental Health (1999). The effects of depression in the workplace. Retrieved
September 3, 2004, from https://www.nimh.nih.gov/publicatlworkplace.cfm
National Institute of Mental Health (2001). The numbers count. Mental disorders in America. Retrieved
September 3, 2004, from https://www.nimh.nih.gov/publicatlnumbers.pdf
National Institute of Mental Health (2003). Breaking ground, breaking through: The Strategic Plan for
Mood Disorders Research. Retrieved September 3, 2004, from https://www.nimh.nih.gov/strategic/
mooddisorders.pdf
Parker, G., Roy, K., & Eyers, K. (2003). Cognitive behavior therapy for depression? Choose horses for
courses. American Journal of Psychiatry, 160, 825-834.
Robins, C. J. (1993). Implications of research in the psychopathology of depression for psychotherapy
integration. Journal of Psychotherapy 1ntegration, 3, 313-330.
Rush, A. J., & Thase, M. E. (2002). Psychotherapies for depressive disorders: A review. In M. Maj & N.
Sartorius (Eds.), Depressive disorders (2nd ed.) (pp, 161-206). Chichester: Wiley.
Segal, Z. V., Pearson, J. L., & Thase, M. E. (2003). Challenges in preventing relapse in major depression.
Report of a National Institute of Mental Health Workshop on state of the science of relapse prevention
in major depression. Journal ofAffective Disorders, 77,97-108.
Segal, Z. v., Williams, J. M.G., & Teasdale, J. D. (2002). Mindfulness-based Cognitive Therapy for
depression. A new approach to preventing relapse. New York: Guilford Press.
Shapiro, D. A., Barkham, M., Rees, A., Hardy, G. E., Reynolds, S., & Startup. M. (1994). Effects of
treatment duration and severity of depression on the effectiveness of cognitive-behavioral and
psychodynamic interpersonal psychotherapy. Journal of Consulting and Clinical Psychology. 62, 522-
534.
Shea, T., Elkin, I., Imber, S. D., Sotsky, S. M., Watkins, 1. T., Collins, 1. E. et al. (1992). Course of
depressive symptoms over follow-up: Findings from the National Institute of Mental Health Treatment
of Depression Collaborative Research Program. Archives of General Psychiatry, 49, 782-787.
Solomon. D. A., Keller, M. B., Leon, A. c.. Mueller, T. I.. Lavori, P. w., Shea, M. T., Coryell, w., Warshaw,
M.. Turvey, c.. Maser. J. D., & Endicott, J. (2000). Multiple recurrences of major depressive disorder.
American Journal of Psychiatry, 157, 229-233.
Verheul, R. (2003). Chronische depressie en persoonlijkheidsstoornis. Overeenkomsten, onderscheid en
consequenties [Chronic depression and personality disorder. Similarities, distinctions and
consequences]. In M. B. J. Blom, J. Spijker & R. van Dyck (Eds.), Behandelingsstrategieen bij
chronische depressie en dysthymie [Treatment strategies for chronic depression and dysthymic
disorder) (pp, 27-44). HoutenlMechelen: Bohn Stat1eu Van Loghum.
Viguera, A. c.. Baldessarini, R. 1., & Friedberg, J. (1998). Discontinuing antidepressant treatment in major
depression. Harvard Review of Psychiatry, 5, 293-306.
Westen, D., & Morrison, K. (200 I). A multidimensional meta-analysis of treatments for depression, panic.
and generalized anxiety disorder: An empirical examination of the status of empirically supported
therapies. Journal of Consulting and Clinical Psychology, 69, 875-899.
Westen, D., Novotny, C. M., & Thompson-Brenner, H. (2004). The empirical status of empirically
supported psychotherapies: Assumptions, findings, and reporting in controlled clinical trials.
Psychological Bulletin, 130, 631-663.
Williams, J. w., Jr., Mulrow, C. D., Chiquette, E., Hitchcock Noel, P., Aguilar, C., & Cornell, J. (2000). A
systematic review of newer pharmacotherapies for depression in adults: Evidence report summary.
Annals of Internal Medicine, 132,749-756.

15
 Chapter 1

The Lifecycle of Depression

Koen Demyttenaere, Lukas Van Oudenhove, & Jiirgen De Fruyt

Mood disorders in general and major depressive disorder (MOD) in particular

are common, disabling psychiatric disorders with an estimated lifetime prevalence
for MOD of around 17% in the community (Angst, 1997). The Belgian ESEMeD
(European Study of Epidemiology of Mental Disorders) results confirm this
estimate (lifetime prevalence of MOD 10.4% and 18.9% in males and females,
respectively) (Alonso, Angermeyer, Bernert, Bruffaerts, et aI., 2004).
Furthermore, the prevalence rates of MOD seem to have increased considerably
over the past decades. A substantial increase in the rates in cohorts born after World
War II and a decrease in the age of onset have been found in several large
epidemiologic and family studies (Klerman & Weissman, 1989). There is some
evidence that these changes are at least partly artifactual, though. Identification
phenomena, memory effects and differential mortality, among others, can explain
why these findings may be to some extent due to artifacts (Klerman & Weissman,
1989).
Over the past fifteen years, there has been increasing interest in the longitudinal
course of MOD in the scientific literature. The discovery of psychotropic drugs
in general and antidepressants in particular certainly made it possible to treat
the acute phase of MOD more appropriately, but did this influence the long-
term naturalistic outcome of the disorder? Nowadays, it is becoming more and
more clear that MOD should be seen as a disorder that is chronic, progressive
and recurrent in nature (Angst, 1997; Mueller et aI., 1999; Solomon et aI.,
2000; Kessing, Andersen, Mortensen, & Bolwig, 1998). In this review, we'll try to
clarify some important aspects concerning the longitudinal course of depressive
illness, in particular the recent evidence on the differences between first and
recurrent episodes.
Finally, there has been a growing amount of research on the validity of the
concept of MDD as defined by DSM-IV diagnostic criteria. Until a few years ago,
little was known about the significance of subthreshold depressive symptoms or
"subsyndromal depression". Some recent research provides support for the
hypothesis that the "cut-off' of five symptoms or two weeks duration as required
by DSM-IV is rather arbitrary: conventional limits put on a continuum of
depressive symptoms of varying severity and duration. In other words, the key
question is: should the diagnosis of MDD be seen in a categorical or a dimensional
perspective (Judd et al., 1998a, 1998b; Kendler & Gardner, 1998)?

Increasing Rates of Depression: Fact or Fiction?

In 1989, Klerman and Weissman (1989) commented on the results of several

large epidemiologic and family studies, suggesting important temporal changes in
the rate of major depression. In the cohorts born after World War II, an increasing
rate is found, as well as a decrease in the age of onset (with an increase in the late
teenage and early adult years), an increase between 1960 and 1975 in the rates of
depression for all ages, a persistent gender effect (with the risk of depression
consistently two to three times higher among women than men across all adult
ages) and finally a persistent family effect, with the risk about two to three times
higher in first-degree relatives as compared with controls.
Undoubtedly, the use of standardized diagnostical instruments in large
community-based studies and the application of modem statistical techniques to the
large data sets from these studies have made more reliable research on the temporal
changes in rates of mental disorders possible. The results from this kind of studies,
especially those with a longitudinal design, can provide some important
information regarding syndromal validity, natural course and, more indirectly,
etiology, pathogenesis and finally treatment of mental disorders. For instance, the
combination of familial aggregation and temporal changes suggests that gene-
environment interaction plays an important role in the pathogenesis of MDD
(Klerman & Weissman, 1989). It is important, though, to mention that the reported
studies have a cross-sectional instead of a longitudinal design, which plays a
considerable role in some of the issues discussed below.
Some of the findings mentioned above, like the higher rate among women and
first-degree relatives of depressed patients, are consistently reported in the literature
(Klerman & Weissman, 1989). Others, especially the increasing rate in recent birth
cohorts are more controversial and merit some further clarification. First, the cohort
effect in MDD is not universally reported. For example, community-based

18
epidemiological studies in Puerto Rican, Mexican-American and Korean
populations do not report these temporal changes, whereas comparable Ameri-
can, German and Canadian studies consistently find a cohort effect (Klerman
& Weissman, 1989). Furthermore, there are some possible explanations why
these findings of temporal changes may be artifactual rather than reflecting real
changes in the rate of MOD. We will discuss three of the most interesting ones in
detail.

Identification Phenomenon

An identification phenomenon can be an important confounding factor, as the

studies mentioned above relied mainly on subjects' retrospective reports. Hasin and
Link (1988) set up a study to test the hypothesis that there is an interaction between
age and recognition. In other words, they expected that older individuals are less
likely to recognize depression as a mental disorder, compared to younger adults,
who might have a more "psychological" view to their experiences. The subjects in
the study, 152 randomly selected citizens living in a suburban area of New York,
were asked to imagine themselves in the situation described by the following
vignette, describing a OSM-III major depressive episode with a duration of a
month. A statement reflecting impairment in functioning was inserted after the
vignette. Furthermore, subjects were requested to fill in a questionnaire. The
questions asked whether the subjects regarded the situation described by the
vignette as emotional/psychological in nature, and whether they considered the
situation to be a problem or not:
"During the last month, even though nothing has gone wrong that you can think
of, you have been feeling depressed. YcJU have not been able to get enough sleep at
night, you haven't had much appetite, and you have lost overfive pounds. You have
felt tired much of the time, and haven't been able to concentrate as well as usual.
You have also noticed that you aren't enjoying things the way you would
normally..;" (Hasin & Link, /988).

The results after analysis of the answers were quite clear. Age had a highly
significant effect that did not change after controlling for vignette types, education
and gender. The chance of recognizing the vignette as major depression was 3.78
times higher for a 25-year old than for a 60-year old. Thus, older people are less
likely to remember depressive episodes or report them when asked for in a
questionnaire regarding mental health. The authors conclude that there is a cohort
effect in the perception rather than in the real rate of depression (Hasin & Link,
1988).

19
Memory Effect: Diminished Recall

Memory effects can provide a second possible explanation for a cohort effect in
rates of major depression as found in cross-sectional studies: older adults are more
likely to forget previous depressive episodes compared to younger ones. Giuffra
and Risch (1994) set up a simulation study to examine the effect of forgetting on
differences between cohorts . The results suggest that small, constant annual rates of
forgetting applied to successive birth cohorts generate cohort-like effects. These
findings confirm the hypothesis that diminished recall in older cohorts is one of the
possible confounding factors involved in the reported cohort effect in major
depression rates. More generally, we have to take into account the somewhat
limited value of cross-sectional studies when doing research on temporal changes
in prevalence.
Moreover, work by Rice et al. indicates that a lifetime diagnosis of major de-
pression is not stable over time, although the error decreases as severity increases.
They assessed a population of 2,226 first-degree relatives of 612 probands
who participated in another study at two time points, with an interval of 6 years,
A substantial proportion of the assessed subjects were considered as having a
lifetime diagnosis of major depression at one of the two time points, but not at
the other. This could be due to suboptimal repeatability of the diagnostic instru-
ment (reliability), diminished recall or limited validity of the underlying diag-
nostic construct, however (Rice, Rochberg, Endicott, Lavori, & Miller, 1992).
Anyway, these findings provide some further evidence for the role of memo-
ry effects as a confounding factor in cohort effects found in cross-sectional
studies.

Differential Mortality

Finally, the findings of temporal changes could be at least partly explained by a

difference in mortality between healthy subjects and depressive subjects. If
depression is associated with a higher mortality, the older birth cohorts would have
lost more depressed subjects by death compared to younger birth cohorts. This
could decrease the rate of depression in interviewed older populations (Klerman &
Weissman, 1989).
The relationship between MDD and mortality is complex and there have been
some conflicting studies concerning this issue. In a recent systematic review
regarding this topic, Wulsin, Vaillant and Wells (1999) found 57 studies. Twenty-
nine of these (51%) found a positive association, 13 (23%) were negative and the
results of the remaining IS studies (26%) were mixed. In general, most of the
studies on the link between depression and early death varied widely in metho-

20
dology (sample selection, comparison groups, control for confounding variables,
etc.), making comparisons between the different studies rather difficult. Further-
more, publication bias certainly is a problem. Some of the mentioned studies were
well designed however. For these reasons, "the evidence that depression increases
mortality may not be strong enough to answer the question definitively" (Wulsin et
aI., 1999).
However, some conclusions can be drawn from this review. 16% to 19% of the
mortality in psychiatric samples of depression is due to suicide, which is consistent
with the often reported suicide rate of 15% in MDD. By contrast, in most studies
based on depressive samples recruited from the community or medical settings, no
more than I% of deaths are caused by suicide. Apart from suicide and other non-
natural causes of death, a quite strong link between depression and cardiovascular
mortality was found (Wulsin et al., 1999).
Although the important and interesting discussion on possible mechanisms
linking depression and mortality lies beyond the scope of this review, we will
mention two hypotheses briefly. First, depression may cause early death in an
indirect way, for example due to poor self-care (including unhealthy food habits but
also compliance with medical treatment). Second, there may be a more direct,
pathophysiological link between depression and mortality. For example, there has
been considerable evidence that depression is associated with important changes in
endocrinological and immune systems. These changes make depressive individuals
more vulnerable to cardiovascular and other diseases (Whooley & Browner, 1998).
There is some evidence from well-designed studies that even subsyndromal
depressive symptoms are associated with both higher mortality and immunological
activation, which suggests a possible link (Glaser, Robles, Sheridan, Malarkey, &
Kiecolt-Glaser, 2003; Whooley & Browner, 1998). We will discuss the importance
of subsyndromal depressive symptoms more thoroughly further on. Dysfunction of
the autonomous nervous system or abnormal platelet aggregation could also cause
higher cardiovascular mortality among depressed individuals (Whooley &
Browner, 1998; Osby, Brandt, Correia, Ekbom, & Sparen, 2001).
As already mentioned above, there have been some studies on the asso-
ciation between depression and mortality in all kinds of populations and samples
and it is important to take this into account when looking at the result of those
studies.
Bruce, Leaf, Rozal, Florio and Hoff (1994) examined the relationship between
psychiatric illness and mortality over 9 years in the community sample from the
New Haven Epidemiologic Catchment Area Study, consisting of 3,560 subjects.
Subjects who reported a recent episode of MDD at the first interview had a 9-year
relative mortality risk of 2.0 I. There was a significant interaction with gender:
recently depressed men were 4.22 times more likely to die during the follow-up
period compared to men without recent depressive episode; for women, the relative
mortality risk was 1.65 (Bruce et aI., 1994).
In a recent study by Osby et aI. (2001), the cause of death of all Swedish
inpatients with a diagnosis of bipolar (N=15,386) or unipolar mood disorder
(N=39,182) between 1973 and 1995 was determined. All patients were followed
from their first hospital admission, using the Swedish psychiatric inpatient register.

21
The date and the cause of death were found in the national cause-of-death register.
The standardized mortality ratios (SMRs) for suicide in unipolar disorder were 20.9
and 27.0 for males and females, respectively. These rates are particularly high,
which can be explained by the population studied, consisting of severely ill
inpatients. The risk of suicide was particularly high for younger patients during the
first years after the first diagnosis, findings that are important when looking for
prevention strategies. The SMRs for all natural causes of death in unipolar disorder
were 1.5 and 1.6 for males and females, respectively (Osby et al., 2001). In a
longitudinal study of psychiatric inpatients, Angst et al. found higher suicide rates
compared to a control population, as well as a higher cardiovascular mortality rate.
Moreover, pharmacological treatment lowered the suicide rate (Angst, Stassen,
Clayton, & Angst, 2002).
Bingefors, Isacson, Knorring, Smedby and Wicknertz (1996) conducted a
community-based study in a primary and psychiatric ambulatory care setting. They
identified all first-incidence antidepressant users in ambulatory care and analyzed
their mortality during a nine year follow-up period. They found that in patients aged
65 and older, antidepressant treatment at index was a significant predictor of higher
9-year mortality. A hazard ratio of 1.52 was found, controlling for potential
confounding variables like chronic medical disease at baseline. Regarding
cardiovascular death, antidepressant treatment was not an independent risk factor,
but patients with both ischemic heart disease and antidepressant treatment at
baseline had a significantly higher risk of death compared to individuals with
ischemic heart disease or antidepressant treatment alone. The implications of
this study are somewhat limited, though. First, not all individuals taking
antidepressants suffer from MOD. Second, the higher mortality rate among patients
suffering from ischemic heart disease and taking antidepressants at baseline could
be due to the widespread use of tricyclic antidepressants at the time this study was
conducted, as those agents are known to have adverse cardiac effects (Bingefors et
aI., 1996).
As a part of the recent Dutch Amsterdam Study of the Elderly (AMSTEL),
Schoevers et a1. (2000) examined the influence of major depression, but also minor
depressive symptoms, on risk of death. The study is longitudinal in design and the
large sample consists of non-institutionalized, older persons who are living in the
community and consult their GP. The follow-up period was 6 years, with an average
of 55.5 months. The overall prevalence of depression was 12.9%, with 6.9% in men
and 16.5% in women. These findings are in line with other studies on the
prevalence of depression in community-living elderly people. Severe depression
with psychotic features was significantly associated with higher mortality in both
men and women, with unadjusted relative risks of 3.77 and 2.43, respectively. After
adjustment for potential other explanatory variables, the relative risks were 1.64
and 1.66 respectively, which was still statistically significant. However, at the end
of the follow up period 75% of the men suffering from psychotic depression had
died, as compared to 41.4% of the women. For less serious "neurotic depression",
the relative mortality risk reached statistical significance only in men, before
and after controlling for potential confounding factors (2.67 and 1.90, respecti-
vely). In "neurotically depressed" women, an unadjusted relative risk of 1.14

22
was found. After adjustment, the RR was 1.02. It is noteworthy that not only MDD
but also "milder" depression (not meeting the full DSM-IV diagnostic criteria)
has an impact on mortality in old men. As a consequence, more attention should
be paid to the diagnosis and treatment of this kind of depression in elderly
men. Some possible explanations for the differential mortality between men and
women are provided by Schoevers et al. (2000) but will not be discussed further
here.
Finally, the association between depression and mortality in a nursing home
population was investigated by Rovner et al. (1991). Newly admitted patients were
followed prospectively for one year. First, the rates of depression in nursing homes
were higher than those reported in the general population. A lot of factors, both
psychosocial and medical, contribute to these higher rates. However, this will not
be addressed further here. Second, the majority of cases of MDD were not recog-
nized nor treated by nursing home physicians. Finally, MDD but not subsyndromal
depressive symptoms, was a significant predictor of mortality, independent of other
risk factors like medical disease, with a relative risk of 1.59 (Rovner et aI., 1991).

Conclusion

In summary, the cohort effect of MDD reported in several, but not all, large
cross-sectional epidemiologic studies is at least partly due to artifacts. We discussed
some potential explanations, including identification phenomena, memory effects
and differential mortality.

Long Term Course of MDD: First Versus Recurrent Episode

"The attacks begin not infrequently after the illness

or death of near relatives...
We must regard all alleged (psychic) injuries as possible sparks
for the discharge of individual attacks, but the real cause of the malady
must be sought in permanent internal changes ...
fn spite of the removal of the discharging cause,
the attack follows its independent development...
But finally, the appearance of wholly similar attacks on
wholly dissimilar occasions or quite without external
occasion shows that even there where there has been
external influence, it must not be regarded as a necessary
presupposition for the appearance of the attack"
E. Kraepelin (/921)

23
 "Single episodes are extremely rare if the period of observation
is significantly extended"

J. Angst (1973)

These two quotes illustrate that there has been interest in the long-term course of
affective disorders for many years, from the beginning of the 20 th century onward.
However, the amount of research on this topic has considerably increased over the
past two decades. As it is now generally accepted that mood disorders are often
chronic, recurrent and progressive in nature, attention has partly shifted from
treatment of the acute phase to prevention of relapse/recurrence. In other words, the
high rates of relapse, recurrence and chronicity reported in the literature made clear
that the search for effective and efficacious strategies of continuation or
maintenance treatment for affective disorders is one of the most important
challenges of contemporary psychiatry.
When comparing different studies on the longitudinal course of MDD, the terms
"response", "recovery", "relapse" and "remission" are not always defined in the
same way, which may make comparisons between studies difficult. However, these
terms are conceptualized by Frank et al. (1991). The term "response" indicates a
decrease in depressive symptoms without returning to baseline, as opposed to
"remission". The difference between "remission" and "recovery" is based on a
rather arbitrarily chosen time criterion (6 months) and not on a pathophysiological
basis. This is also true for the distinction between "relapse" and "recurrence".
Moreover, not all studies use the same time criterion.
We will discuss some of the most relevant topics regarding the long-term course
of MOD in the following paragraphs.

High Risk of Recurrence ill MDD

There is a growing consensus that the risk of relapse and/or recurrence in

MDD is generally high. Lavori, Dawson and Mueller (1994) followed depressive
patients longitudinally after recovery from their first lifetime episode. The
recurrence rate was 13% after six months, 28% after one year, 62% after five years,
75% after ten years and finally 87% after fifteen years. These findings clearly
provide further evidence for the 30-year-old quote by Angst (1973), cited above.
In the NIMH Collaborative Program on the Psychobiology of Depression -
Clinical Studies, a cohort of 380 depressed patients from five university psychia.ric
clinics has been prospectively followed after recovery from the index episode over
a period of fifteen years. MUlier et al. (1999) report on the results of this unique
longitudinal, naturalistic study. The cumulative proportion of recurrence (by

24
Kaplan-Meier estimate) at the end of the follow-up period was 85%, with a median
time to recurrence (well interval) of 132 weeks. The group who experienced a
recurrence and the group who remained well received similar low levels of
antidepressant treatment during both the index episode and the subsequent well
interval. During this period of remission/recovery the amount and extent of
antidepressant treatment decreased further. In addition, the authors focused on the
subgroup of patients who remained well for five years after the index episode
(n= I05). In this subgroup, 58% suffered from a recurrent episode of MDD during
the subsequent 10 years of follow-up. This indicates that the risk of recurrence
decreases with the length of the well interval after an index episode, but there
remains a considerable risk of relapse even after lengthy well intervals. It is
important to take this finding into account when making clinical decisions on
maintenance treatment (Mueller et aI., 1999).

Predictors of Early Relapse/Recurrence

First of all, having experienced more prior episodes is strongly associated with a
higher risk of relapse/recurrence. It is consistently reported in the literature that risk
of recurrence increases by some 15% with each successive recurrence (Mueller et
aI., 1999; Solomon et aI., 2000; Kessing & Andersen, 1999), although this is not
found in all studies (Kessing, Andersen, & Mortensen, 1998).
Sociodemographic variables like gender, age at onset and marital status predict
recurrence only in the earlier episodes (Kessing et aI., 1998; Post, 1992). This
seems to be the case for comorbid alcoholism too (Kessing, 1999). We will return
to this issue later on.
Moreover, the risk of recurrence decreases progressively as the "well interval"
(duration of recovery) increases (Mueller et aI., 1999; Solomon et aI., 20(0).
However, even after a well interval of five years, recurrence rates as high as 58%
over the subsequent 10 years are reported in the above-mentioned well-designed
naturalistic NIMH study (Mueller et aI., 1999). Therefore, long enough courses of
maintenance treatment with antidepressants (and/or psychotherapy) are strongly
recommended, especially in patients with a history of highly recurrent depression.
The optimum duration of maintenance treatment is less clear, though, and remains
often a clinical rather than a really evidence-based decision. Durations varying
from two to five years or even lifelong are suggested in the literature (Angst, 1997).
Despite these recommendations, in a naturalistic setting, many patients do not
receive adequate antidepressant maintenance therapy. As already mentioned above,
in the NIMH study doses were low and duration was short (Mueller et aI., 1999).
Finally, incomplete remission/recovery (the presence of residual subsyndromal
depressive symptoms) is strongly associated with early relapse or recurrence. This
is an important finding as it questions the validity of the boundaries of MDD as
defined by DSM-IV diagnostic criteria. Patients suffering from residual

25
subsyndromal depression (SSD) relapsed three times faster to a major depressive
episode compared to patients who fully recovered after the index episode. This
association is stronger than the association between the number of prior episodes
and time to relapse or recurrence (Judd et al., 1998a, 1998b). Of course, these
findings have important implications regarding treatment too. Residual SSD will be
discussed more extensively later on in this review.

Duration of Episodes

In a longitudinal, prospective, naturalistic study conducted in tertiary care

centers, Solomon et al. followed a cohort of 258 subjects suffering from unipolar
major depressive disorder over 10 years from the index episode on. The medi an
duration of illness was around 20 weeks and this was highly consistent across
multiple recurrent mood episodes (Solomon et al., 1997). Although consistent with
several other studies in the literature (Angst, 1986; Kessing & Mortensen, 1999),
this finding does not support the often quoted clinical impression that duration
increases with each recurrent episode. That impression may be at least partly due
to decreasing well intervals between each subsequent episode (however see below).
Moreover, the results described are applicable to the whole cohort, whereas in
individual patients consistency in the duration of illness across multiple episodes
was low to moderate (Solomon et al., 1997). No sociodemographic or clinical
factors were associated with the duration of episodes. This uniform duration of
recurrent depressive episodes suggests that the rate of recovery from a depressive
episode does not change in the course of illness. In other words, recurrent episodes
are equally "treatable" compared to earlier episodes. However, the level of
treatment increased with each recurrent episode, which could have influenced the
duration (Solomon et al., 1997). Also, compliance is higher in recurrent episodes
(Demyttenaere & Bruffaerts, 20ma). Some of our own unpublished data challenge
the above-mentioned findings and indicate that recurrent episodes are more difficult
to treat compared to first episodes, despite the higher treatment adherence
(Demyttenaere & Bruffaerts, 20mb).
Furthermore, the population studied is an important issue when looking at data
regarding duration of depressive episodes. The above mentioned studies all
consisted of patients seeking help at tertiary care centers, which had an influence
on the results. As part of the Dutch NEMESIS study, Spijker et al. (2002) examined
the duration of major depressive episodes in the general Dutch population. In this
study, the median duration of depressive episodes in the whole study was 3 months
and recurrent episodes were shorter in duration compared to first episodes. Severity
of the index episode predicted longer episode duration. However, the level of care
influenced the duration of episodes (median duration was 3 months for those
without professional care, 4.5 months in primary care and 6 months in mental
health care), although the differences did not reach statistical significance (Spijker
et al., 2002). Other longitudinal community-based studies like the Epidemiologic

26
Catchment Area study report similar durations around 12 weeks (Eaton et al., 1997;
Spijker et al., 2002).

Time to Recurrence

A substantial number of patients experience a progressive course of illness,

characterized by "cycle acceleration", In other words, in this group of patients, the
"well interval" (or time to recurrence) decreases with each recurrent episode
(Kessing et al., 1998; Post, 1992). This finding is in line with most previous studies
on this topic, although some studies did not find such a pattern of cycle
acceleration. These discrepancies may be explained by methodological differences
or problems (Kessing et al., 1998).
Moreover, patients with an initial course of illness characterized by decreasing
time intervals between the first three lifetime episodes had a 33% increased risk of
experiencing further recurrence, compared with patients who did not follow such a
pattern of initial illness (Kessing, Mortensen, & Bolwig, 1998).

The Role of Psychosocial Stress and Life Events. "From Precipitated

to Spontaneous Episodes": The Kindling Hypothesis

Several studies suggest that the etiologic role of stressful life events in major
depression is more important in first-onset episodes compared to recurrent episo-
des, at least in an important subgroup of patients. In other words, early episodes are
more likely to be triggered by psychosocial stressors, whereas recurrent episodes
are more likely to occur spontaneously, i.e., not related to any important life event
(Post, 1992). As noted, this pattern was already described by Kraepelin in the early
1920s (Kraepelin, 1921). Nowadays the term "kindling hypothesis" is used to des-
cribe this evolution from precipitated to spontaneous episodes. The term "kindling"
is derived from epileptology (see below). There is quite a lot of evidence supporting
this hypothesis. For example, the effect of never being married decreases during the
course of depression (Kessing, Andersen, & Andersen, 2000). Moreover, in the
National Institute of Mental Health collaborative study on the psychobiology of
depression, the mean number of prior episodes in patients with an "environment-
sensitive" episode was 3.7, whereas the mean number of episodes in patients with
an "autonomous" episode was 13.4 (Post, 1992; Swann et al., 1990). In a meta-ana-
lysis of studies on this topic, 57% of first episode patients experienced a major life
event prior to the onset of their depressive episode, compared to 32% of later episo-
de patients (Post, 1992). Although the studies were methodologically quite diffe-
rent, their findings were rather consistent. A recent study by Mitchell et at. (2003)
suggested that this differential effect of life events in first compared to recurrent
episodes may be specific to non-melancholic depression (Mitchell et al., 2003).
Kendler, Thornton and Gardner (2000) argue that several previous studies on this
"kindling" hypothesis suffer from several methodological limitations (clinical

27
populations, cross-sectional studies, etc.). Moreover, the fact that depressive
subjects may generate more life events compared to non-depressive controls has to
be taken into account (Kendler et al., 2000; Harkness, Monroe, Simons, & Thase,
1999). In a longitudinal population-based twin study (N=2,395), Kendler and
colleagues (2000) interviewed participants four times over a period of nine years.
Only life events that were clearly independent of the respondent's own behavior
were included in the analyses in order to determine "whether the change in
sensitivity to the effects of stressful life events over episodes varied within as well
across individuals" (Kendler et aI., 2000). The results of this study were largely
consistent with the kindling hypothesis: the number of previous episodes had a
strong effect on the association between stressful life events and depressive onsets,
but this relation was biphasic. A strong negative interaction between previous
depressive episodes and stressful life events in the prediction of risk for a successive
depressive episode was found, especially in the first nine episodes. After the ninth
episode, the interaction became weaker. In other words, the odds ratio (OR) for
depressive onset given at least one stressful life event decreased significantly with
an increasing number of previous depressive episodes for the first nine episodes.
From the tenth episode on, the OR remained more or less constant. The authors
conclude: "these findings suggest that, whatever the biological or psychological
process that underlies this phenomenon, it is 'saturable" (Kendler et aI., 2000).
In another well designed study the same authors examined the impact of genetic
risk factors on this "kindling" phenomenon (Kendler, Thornton, & Gardner, 2001).
The results were consistent with a model of "prekindling" in those at high genetic
risk (defined by presence of a history of depression in the co-twin). Kendler et a1.
(200 I) found that subjects at high genetic risk reached the "kindled" stage (i.e., the
stage of spontaneous rather than triggered episodes) after significantly less
depressive episodes compared to subjects who were not at high genetic risk. For
example, a first depressive onset was associated with a significant life event in
59.7% of the low genetic risk group but only in 38.2% of the high genetic risk
group. In summary, "the decline in the association between stressful life events and
risk for major depression as the number of previous depressive episodes increases
was strongest in those at low genetic risk and weaker in those at high genetic risk,
Therefore, there may be potentially distinct environmental and genetic pathways to
a "kindled" or "sensitized" state in which the mindlbrain is predisposed to spon-
taneous depressive episodes".
Robert M. Post was the first to apply the models of kindling (originally described
in epileptology) and behavioral sensitization (derived from psychopharmacology)
to the longitudinal course of affective disorders (Post, Rubinow, & Ballenger,
1986). Kindling can be defined as "the development of major motor seizures in
response to repeated intermittent electrical stimulation of the brain with insufficient
current to produce overt behavioral effects" (Racine, 1978). This phenomenon
seems to be caused by long-lasting, possibly permanent changes in neural
excitability induced by the repeated application of a "subthreshold" current to the
neuron. Moreover, after many repetitions of kindled seizures, spontaneous seizures
may develop (Post et al., 1986). Behavioral sensitization refers to the progressive
changes in behavioral response to psychomotor stimulants (like cocaine) after

28
repeated administration of a constant dose. In a sensitized state, the behavioral
response is developing with a faster onset, increased intensity and longer duration.
Conditioning seems to be important in behavioral sensitization, as for example
response to cocaine is highly dependent on the environment in which cocaine is
administered. Thus, "the underlying changes in brain biochemistry may be
conditioned" (Post et aI., 1986).
Post ar~I!t''i that these kindling and sensitization paradigms provide nonhomo-
logous model, that can help us understand the longitudinal course of affective
disorders in general and the differential impact of psychosocial stressors in first
versus subsequent episodes in particular. These are nonhomologous paradigms
because kindled seizures and behavioral responses to cocaine do not clinically
resemble affective episodes, although the sensitized response to cocaine does share
some features with dysphoric mania. However, both repeated psychosocial stress
and depressive episodes themselves may induce long lasting cognitive and
neurobiological changes, making the subject more vulnerable to subsequent
episodes. Just as kindled seizures appear spontaneously after repeated stimulation
during a long enough period of time, affective episodes eventually occur with-
out environmental triggers (Post. 1992). Type, magnitude and frequency of the
stressor are critical to its long-term effects. For example, stressors involving
psychosocial losses may induce a set of cognitive, behavioral and neurobiological
consequences making an individual more vulnerable to depressive episodes,
whereas acute life-threatening stressors may induce another set of long-lasting
changes more likely to cause post-traumatic stress syndrome (Post, 1992; Post et
aI., 1986).
There is widespread evidence for cortisol hypersecretion both in situations of
psychosocial stress and in depressive episodes (for recent reviews, see Sapolsky,
2000; Raison & Miller, 2003). These elevated levels of cortisol could initiate a
neurobiological cascade of second, third, etc. messenger systems ending in changes
in gene transcription that cause the already mentioned long-lasting neurobiological
changes. From a cognitive point of view, both stressors and depressive episodes
could cause long-lasting changes in cognitive structures. Associations between
previous triggers and their symbolic components may be learned or conditioned so
that the symbolic representation of the stressor is eventually able to cause a
recurrent depressive episode in the absence of a real stressor or loss. However, the
neurobiological and cognitive points of view are not mutually exclusive. The
neurobiological changes induced by stressors and depressive episodes may be the
cause of alterations in memory and cognitive patterns (Post, 1992; Post et aI.,
1986).
In summary, like Nobel Prize winner Eric Kandel (Kandel, 1998), the work of
Post and collaborators tries to build a bridge between neurobiological and
psychosocial theories of depression, providing support for a widely accepted
biopsychosocial stress-vulnerability model of MDD. Nonetheless, further research
is needed as many of the aspects in this kindling theory, especially regarding the
underlying neurobiological mechanisms involved in learning and memory, remain
rather speculative.

29
Cognitive Functions in MDD: The "Scar" Hypothesis

There have so far been quite a lot of contradictory findings in the literature on
cognitive functioning in depression. The importance of cognitive (dys)functioning
in professional and social functioning can hardly be overestimated, though. In an
excellent review on this topic, Austin et al. try to summarize the results and
hypotheses regarding this issue (Austin, Mitchell, & Goodwin, 2001).
Before discussing the pattern of cognitive dysfunction in MOD, it is important
to mention that there are quite a lot of confounding factors that make comparisons
between different studies quite difficult (Austin et aI., 200 I). First, studies on the
effect of depression severity on cognitive function are conflicting. The observation
that depressive individuals scored lower on verbal recall tasks and normal on verbal
recognition tasks led to the hypothesis that the cognitive dysfunction in depression
is secondary to a motivational deficit, causing lower scores on "effortful" and not
on "automatic" tasks (Austin et al., 200 I). However, this hypothesis has been
undermined by several studies that found abnormal performance on "automatic" as
well as on "effortful" tasks. Some tasks may be more sensitive to motivation effects
than others. Similarly, some neuropsychological functions are highly dependent on
current mood, others are not (Austin et aI., 200 I). Porter, Gallagher, Thompson and
Young (2003) suggest that executive functions are relatively stable, whereas
memory functions vary more with severity of depression. In conclusion, there may
be an important interaction between motivation, depressed affect and cognitive
function; further study is needed to clarify this interaction more precisely (Austin
et al., 200 I). Second, subtype of depression (e.g., melancholic versus atypical)
could be associated with a specific pattern of cognitive dysfunction (Austin et aI.,
200 I). Third, the role of medication in cognitive deficits has not been sufficiently
studied. Agents with anticholinergic properties, like most tricyclic antidepressants
(TeA), are particularly known for their negative effects on cognition (POtter,
Gallagher, Thompson, & Young, 2003). Even the mild anticholinergic effect of the
selective serotonin reuptake inhibitor (SSRI) paroxetine causes subtle cognitive
deficits (impaired delayed recall on a word-learning task) in young healthy
volunteers (Schmitt, Kruizinga, & Riedel, 200 I). Finally, age can be an important
confounding factor (Austin et aI., 200 I). Moreover, the role of the microvascular
cerebral changes that are often found in (geriatric) depression remains to be
determined (Austin et aI., 200 I).
There is quite good evidence that depression is associated with deficits in
episodic memory and learning (Austin et aI., 200 I). Both explicit verbal and visual
memory are impaired, whereas performance on implicit memory tasks is spared.
Temporal lobe lesions, especially loss of hippocampal volume potentially caused
by glucocorticoid toxicity, may be responsible for these particular cognitive
dysfunctions (Austin et al., 200 I). We will discuss this extensively further on. The
presence and the pattern of executive dysfunction, mediated by the prefrontal
cortex, is less clear, however. Attentional set-shifting (as measured by the Trail
Making 8 and Digit Symbol Substitution tests) is most consistently reported to be

30
impaired, with more important dysfunction as severity of depression increases
(Austin et aI., 200 I).
In a study under drug-free conditions, Schatzberg et al. (2000) found that
patients suffering from depression with psychotic features show more substantial
cognitive impairment compared to non-psychotic depressive controls, including
deficits in attention, response inhibition (measured by the Stroop test) and explicit
verbal memory. These results are in line with the results of an earlier study by
Basso and Bornstein (1999a).
In another study by Basso and Bornstein (I 999b), 20 single episode (SE) and 46
recurrent episode (RE) young depressed inpatients were administered a compre-
hensive neuropsychological test battery. After controlling for age, severity of
depression, sedation rate and anticholinergic blocking rate, the SE group showed no
significant cognitive impairment. The RE group, on the contrary, showed a
significantly impaired recall on the California Verbal Learning Test compared to the
SE group and published norms. For example, they recalled fewer words during the
five learning trials and also long-delay free and cued recall was impaired. There was
no difference in benefit of recognition cueing relative to long-delay free recall.
Furthermore, the groups showed equivalent performance on the index comparing
long-delay free recall to short-delay free recall. These findings suggest a pattern of
deficient acquisition rather than retrieval or retention. There were no significant
differences in measures of intelligence (subscales of the WAIS), attention (digit
span forwards), visuospatial attention and motor speed (Trail Making Test A) and
various frontal-executive functions (FAS verbal fluency, digit span backwards and
Trail Making Test B) (Basso & Bornstein, 1999b).
Furthermore, there are some studies which examined cognitive functions in
euthymic patients with a history of one or more depressive episodes. First, Tham et
al. (1997) found that the group with a history of depression performed significantly
worse on several tests of cognitive functions (general intelligence, memory,
attention, frontal functions). Moreover, there was a significant association between
number of previous hospitalizations for mood disorder and some of these tests
(reasoning, general intelligence, Trail Making A, set-shifting as measured by Trail
Making B). Second, Kessing (1998) performed a retrospective controlled cohort
study based on the Danish psychiatric case register. Cognitive impairment was
significantly higher in patients with a history of recurrent depressive episodes,
whereas the performance of patients with a history of only one episode did not
differ from normal controls. Moreover, intra-individual analyses showed that the
number of prior episodes correlated with the degree of cognitive impairment. Third,
Paradiso, Lamberty, Garvey and Robinson (1997) found significant cognitive
impairment in euthymic patients with a chronic history of depression. Mainly
attentional and executive functions were impaired, as well as immediate memory.
Finally, Austin et al. (2001) reviewed the literature regarding cognitive deficits
during the euthymic phase of MDD. They concluded that, although the few
longitudinal studies that tested cognitive functions before and after recovery from
depressive episodes suffer from methodological limitations, at least a subgroup of
patients suffers from residual deficits in memory and/or executive functions during
"well intervals". The exact relationship between these deficits and crucial clinical

31
and sociodemographic variables like number of previous episodes, age, treatment
and chronicity needs further research though (Austin et aI., 200 I).
In summary, the above-mentioned studies provide evidence for the hypothesis
that recurrent depressive episodes cause some kind of "cognitive scar", lasting
during episodes of recovery. However, the exact relationship between MOO and
cognitive functions remains to be studied, preferably in longitudinal studies.
As already mentioned above, loss of hippocampal volume in recurrent
depression is an important neurobiological correlate of those residual cognitive
(memory) deficits. This will be discussed more thoroughly in the next paragraph.

Neurobiological Correlates of (Cognitive Deficit in) Depression: Hippocampus

and Amygdala

In a Magnetic Resonance Imaging (MRI) study conducted in 1996, Sheline,

Wang, Gado, Csernansky and Vannier (1996) found a significant negative
association between left and right hippocampal volume and total duration of illness
in euthymic patients with a history of depression compared to matched control
subjects. These results were confirmed by another study by Sheline, Sangha vi,
Mintun and Gado (1999): hippocampal volume decreased as total duration of
depression increased. Moreover, subjects with a history of depression performed
significantly worse on explicit verbal learning tasks. Interestingly, these tasks are
known to be highly hippocampus-dependent. Furthermore, no correlation between
hippocampal volume and age, which could be a confounding factor, was found
(Sheline et al. (1999). Sheline, Gado and Price (1998), in turn, found a 17%
decrease in amygdala core nuclei volume in previously depressed subjects who
were in remission at the time of the study. As the amygdala plays an important role
in emotional processing and memory, these findings are in line with earlier studies
that found impaired performance on emotional tasks in depressed subjects (Sheline
et al., 1998). Finally, Sheline, Gado and Kraemer (2003) found a significant
relationship between degree of hippocampal volume loss and duration of untreated
depressive episodes. These findings provide some more evidence for the hypothesis
that antidepressants provide a neuroprotective effect, which is in line with studies
by Bremner and colleagues in both depression and post-traumatic stress disorder
(PTSO) (Bremner et aI., 2000; Vermetten, Vythilingam, Southwick, Charney, &
Bremner, 2003).
MacQueen et al. (2003) compared hippocampal volume and hippocampal
function (as measured by a recollection memory task) of never-treated first-episode
patients, patients with a history of recurrent episodes and matched healthy controls.
First, the authors found impaired performance on several recollection memory tasks
in first and recurrent-episode patients compared to controls. Measures of habit
memory did not differ between the three groups, which is important as habit
memory is not hippocampus-dependent. These findings indicate that hippocampal

32
dysfunction is present from the first episode on. Second, loss of hippocampal
volume on MRI was found in recurrent-episode patients and not in first-episode
patients or controls. Moreover, the relation between illness duration and
hippocampal volume was best described by a logarithmic curve, indicating that
hippocampal volume loss occurs rather early in the course of the disease, although
it is not yet detectable in first-episode patients (MacQueen et aI., 2003).
There are several possible pathophysiological mechanisms that could lead to
hippocampal volume loss in depression. First, hippocampal neuronal cell death due
to glucocorticoid toxicity is important (Sheline, 2000; Sheline et aI., 1999;
Sapolsky, 2000; Sapolsky, Krey, & McEwen, 1986). It is widely accepted that
dysfunctional feedback mechanisms of the HPA axis cause elevated levels of
cortisol in at least an important subgroup of depressive subjects (Sapolsky, 2000;
Sheline, 2000). Similarly, exposure to repeated stressors causes elevated levels of
cortisol. Both animal and human research has provided substantial evidence for a
toxic effect of glucocorticoids on hippocampal neurons (Sapolsky, 2000; Sapolsky
et al., 1986; Sheline, 2000; Sheline et al., 1998, 1999). For example, patients with
primary Cushing's syndrome (who are exposed to highly elevated glucocorticoid
levels for a long period of time) are found to have both lower hippocampal volume
and impaired hippocampal (memory) functions. Moreover, these changes are
reversible after treatment, resulting in normalization of cortisol levels (Starkman,
Giordani, Berent, Schork, & Schteingart, 2001; Starkman, Giordani, Gebarski, &
Schteingart, 2003). Glucocorticoid toxicity in hippocampal neurons may be
mediated by enhanced vulnerability to the excitotoxic effect of excitatory amino
acids like glutamate (Sheline, 2000). Furthermore, Corticotrophin Releasing Factor
(CRF) also has a direct or indirect neurotoxic effect. Several studies found elevated
levels ofCRF in depression and CRF antagonists are neuroprotective (Sheline et aI.,
1999). Second, loss of glial cells could be a potential mechanism, again resulting in
increased vulnerability to glutamate (Sheline, 2000; Sheline et al., 1999). Third,
stress-induced reduction in neurotrophic factors like Brain-Derived Neurotrophic
Factor (BDNF) could playa role in volume loss. Stress decreases the expression of
BDNF and antidepressants reverse this decrease in expression. This may be one
potential mechanism for the above-mentioned neuroprotective potential of these
agents (Sheline, 2000, 2003; Sheline et al., 1999).
However, many questions remain to be answered. For example, it is not known
whether depression is the cause or the result of the above mentioned structural
changes, or even both (Sheline, 2003). Modern high-resolution brain imaging
techniques like Positron Emission Tomography (PET) and functional Magnetic
Resonance Imaging (fMRI) will undoubtedly contribute to a better knowledge of
the neurobiological correlates of depression and their evolution during the long-
term course of depressive illness.

33
 Categorical Versus Dimensional Diagnosis of Depressive Disorder

A DSM-IV diagnosis of major depressive episode (MDE) requires five

depressive symptoms or more, clinically significant impairment and at least 2
weeks duration. This definition reflects a categorical view of major depression:
either you suffer from MDE, or you do not, depending on criteria defining a "cut-
off'. Similarly, DSM-IV defines criteria for some other depressive syndromes like
minor depressive episode (MinDE). However, there has been increasing debate
about these rather arbitrarily chosen criteria over the past few years. The key
question is whether these "boundaries" set by DSM-IV criteria reflect "real" limits
of distinct depressive syndromes. Alternatively, depressive illness can be seen as a
continuum of depressive symptoms with variable severity and duration over the
course of illness (dimensional diagnosis). In other words: does major depression
differ qualitatively or just quantitatively from minor depression or "subsyndromal"
depressive symptoms? A qualitative difference means that the different unipolar
mood disorders as defined by DSM-IV (MDD, minor depression, dysthymic
disorder) do represent different clinical entities, each with different clinical
characteristics and pathophysiology. In contrast, if the difference between these
depressive syndromes is only quantitative, depression should be seen as a single
disease characterized by a continuum of different and variable levels of symptoms,
but with a single, although multifactorial, etiology and pathogenesis. Recently,
there have been quite a number of studies which tried to answer this important
question.
Judd and colleagues have done a lot of work on this topic. They define
subsyndromal symptomatic depression (SSD) as "depressive symptoms beneath the
diagnostic threshold for minor, dysthymic or major depressive disorder" (Judd et
al., 1998a). In community samples, high prevalences of SSD are found, causing
significant psychosocial impairment. Moreover, some studies indicate that residual
SSD after recovery of an index major depressive episode is a significant predictor
of relapse or recurrence, as we already briefly mentioned (Judd et al., 1998a). To
test this hypothesis, Judd et al. (1998a) followed a large cohort of patients
prospectively for 10 years after recovery from an index episode of MDD, with
special interest in the influence of SSD on duration of recovery. First, asymptomatic
recovery (AR) patients had significantly shorter intake MDEs and significantly
more AR patients did not suffer from a recurrent episode during the 10 years of
follow-up. Second, MDE relapse was more than three times faster for residual SSD
recovery patients. Moreover, SSD recovery patients relapsed 5.5 times faster to any
depressive episode (MinDE or MDE). AR patients remained well for a median of
231 weeks, whereas residual SSD recovery patients had a median time to
recurrence of only 68 weeks. Furthermore, recovery status (AR versus SSD) was a
stronger predictor of relapse/recurrence than history of previous recurrent
depressive episodes. SSD patients received more antidepressants during their first
well interval, but not during the index episode, indicating that the higher relapse
rates in SSD recovery patients are not due to differences in antidepressant treatment

34
(Judd et aI., 1998a). These findings are in line with several earlier studies and
confirm the hypothesis that SSO is an active state of depressive illness rather than
some kind of recovery (Judd et aI., I998a). In other words, the results of this study
provide some evidence for a dimensional rather than a categorical approach to
depressive illness. Furthermore, this study illustrates the need for treatment until
full recovery is reached rather than partial remission. It is possible that SSO
recovery patients need higher doses or combinations of antidepressants, more
frequent psychotherapy or some other forms of intensive treatment to reach
asymptomatical status (Judd et aI., 1998a). Fava, Rafanelli, Grandi, Conti and
Belluardo (1998) found cognitive behavioral therapy (CBT) to be significantly
more effective than clinical management in reducing residual symptoms and
preventing relapse after remission from an index MOE. In this study,
antidepressants were tapered and finally discontinued during the 2-year follow-up
period. After two years, the CBT group had a relapse rate of 25% compared to 80%
in the clinical management group.
In another article, Judd et al. (I 998b) report on the weekly symptomatic course
of MOO in a cohort of patients from the same naturalistic study. Patients were
followed prospectively over a period of 12 years and weekly depressive symptoms
were recorded and divided into 4 levels of severity. The first level was defined as
depressive symptoms at or above the OSM-IV threshold for MOE, the second level
as depressive symptoms at the threshold for MinOE, the third as SSO below the
thresholds for MOE or MinOE and the fourth level was defined as no depressive
symptoms (asymptomatic). Patients were divided into 3 subgroups: first lifetime
MOE, recurrent episode and "double depression" (MOE superimposed on
dysthymia). The authors found that patients presented with multiple levels of
symptoms over time: 59% of patients spent time at each of the 4 levels and only 1%
of patients spent their entire follow-up period at one symptom level. Moreover,
symptom levels changed relatively quickly and often, with an average total of 17.1
changes during follow-up. Furthermore, patients suffering from their first lifetime
MOE at intake were significantly more often symptom-free than the two other
groups in the sample and they had significantly fewer changes. suggesting a more
benign course in first-episode patients. Finally, MinO and SSO symptoms were 3
times more common than MO symptoms (Judd et aI., I998b). This is an important
finding as we have to consider SSO as an active state of illness that highly predicts
MOE relapse, as mentioned above (Judd et aI., I998a). These results provide
further support for the "single clinical illness" (i.e., dimensional) hypothesis of
unipolar depression. Judd et al. (l998b) conclude: "We believe a growing
confluence of scientific evidence supports the hypothesis that unipolar MOO is a
clinically homogeneous illness in which major, minor and subsyndromal depressive
symptoms commonly alternate as different manifestations and levels of illness
activity". This was confirmed by a more recent study from the same group (Judd et
al., 2000).
Kendler and Gardner (1998) examined the relationship between three key
features of the major depressive syndrome and the risk of future episodes of major
depression in both the subject and his/her co-twin, two widely used validating
criteria for psychiatric disorders. The twin pairs were ascertained from a

35
population-based registry; they were followed longitudinally at two time points
after the first interview. The three evaluated features were: numbers of symptoms
listed under DSM-IV criterion A for MDE (at least five required by DSM-IV), level
of severity or impairment (DSM-IV requires either significant distress or significant
impairment in functioning) and duration (minimum of 2 weeks required by DSM-
IV). The risk of major depression in both the subject and his/her co-twin increased
with an increasing number of criterion A symptoms, without any discontinuity
between four and five symptoms, as hypothesized by DSM-IY. Similarly, subjects
who reported five or more depressive symptoms leading to only mild or moderate
impairment still had a significantly greater risk of major depression compared to
subjects who did not report five or more depressive symptoms. The same was true
in the co-twins. However, the increased risk was substantially greater for those who
reported five or more symptoms leading to severe impairment. Finally, major
depressive episodes lasting less than two weeks proved to increase the risk of future
major depression in both subjects and their co-twins, compared to those who did
not report a depressive episode of any duration. In summary, the authors failed to
find empirical evidence for a discontinuity at the boundaries proposed by DSM-IV
(Kendler & Gardner, 1998).
Moreover, our own unpublished data provide some more evidence for a
"differential weight" of the different DSM-IV depressive symptoms, without any
discontinuity between four or five symptoms. Depressed mood, diminished interest
and loss of energy are the most consistently reported symptoms, even at
"subsyndromal" level (Demyttenaere & Bruffaerts, 2002).
In a longitudinal, prospective cohort study of 7,518 older women with a follow-
up of 7 years, Whooley and Browner (1998) found an increased risk of death by
both cardiovascular and non-cancer, non-cardiovascular diseases in patients who
reported "only" subsyndromal or mild depressive symptoms on the IS-item
Geriatric Depression Scale. Moreover, mortality increased with number of
depressive symptoms.
As already mentioned, the relationship between depressive symptoms and
mortality is likely to be multifactorial. One hypothesis is that immunological
changes make depressive patients more vulnerable to "somatic" diseases. In a
recent study, Glaser et al. found that even subsyndromal depressive symptoms in
older adults are associated with higher blood levels of the proinflammatory
cytokine interleukin-6 (IL-6) at baseline. Moreover, subjects reporting depressive
symptoms had an increase in IL-6 levels 2 weeks after influenza vaccination,
whereas the IL-6 levels of non-depressives did not change (Glaser, Robles,
Sheridan, Malarkey, & Kiecolt-Glaser, 2003).
In conclusion, there is a growing body of scientific evidence supporting a
dimensional rather than a categorical concept of MDD. However, we have to take
into account that "disease concepts are always pragmatic in character" (Sadler,
Hulgus, & Agich, 1994). This is especially true in psychiatry because quantifiable
objective measures of disease activity are lacking. Therefore, current psychiatric
classification systems are based on clinical rather than pathophysiological criteria.
Given the above-mentioned body of evidence, scientific rigor may demand a
dimensional view of depression but a categorical disease concept may be more

36
practically useful. We should not forget that comparisons between large studies
were simply not possible before DSM diagnostic criteria, because not everybody
spoke the same "diagnostic language". Moreover, a categorical diagnosis makes the
use of outcome measures in large studies easier, which is important for both
clinicians doing clinical trials and of course for pharmaceutical companies that
want to prove the efficacy of their psychotropic drugs in randomized controlled
trials (RCTs). RCTs are nowadays the "gold standard" in outcome research and
they need some kind of quantification or cut-off that sets the "boundaries" of a
certain disease. This does not mean, however, that the current diagnostic criteria for
MDD do set the "real" boundaries of depression as a discrete disease entity.
Therefore, debate and discussion about diagnostic criteria and validity of disease
concepts are useful and needed to ascertain that these criteria and concepts change
over time rather than remain static. Epidemiological research (especially
prospective studies with a long follow-up period) can teach us a lot about natural
course, risk factors, etiology, pathogenesis and finally validity of the cur-
rent psychiatric syndromes. Thus, the results of this kind of research should be
taken into account when diagnostic criteria are evaluated, discussed and/or
changed.

Conclusions

This review of the "lifecycle of depression" is far from comprehensive, mainly

because of the complex and multifactorial nature of the topic and the growing
number of publications on different aspects of the long-term course of MDD.
Nowadays, it seems quite well established that the reciprocal interaction between
biological, psychological and social factors plays an important role in both the
onset and long-term course of MDD. The exact mechanisms by which these
different pathophysiological factors interact to cause depression remain to be
determined, however.
Some important conclusions can be drawn, though. First, the increasing
prevalence of MDD is at least partly due to psychological biases and differential
mortality. Second, in at least an important subgroup of patients, the course of MDD
deteriorates over time. We discussed some key features of this deteriorating course
such as cycle acceleration, evolution from precipitated to spontaneous episodes and
the induction of psychological/biological scars that possibly deteriorate after each
successive episode. Finally, there is some ongoing scientific debate concerning the
categorical or dimensional approach to MDD, which may have important
implications for the validity of the disease concept of depression as currently
defined by DSM-IV criteria.

37
 References

Alonso, 1., Angermeyer, M. C; Bernert, S.. Bruffaerts, R.. Brugha, T. S., Bryson, H., Girolamo, G., Graaf,
R., Demyttenaere, K., Gasquet. I., Haro, J. M., Katz, S. J.. Kessler, R. C., Kovess, V., Lepine, J. P.,
Ormel, 1., Polidori, G.. Russo, L. J.. Vilagut, G.. Almansa, J., Arbabzadeh-Bouchez, S., Autonell, J.,
Bernal, M., Buist-Bouwman, M. A.. Codony, M" Domingo-Salvany, A., Ferrer, M., Joo, S. S.,
Martinez-Alonso, M.. Matschinger, H., Mazzi, E, Morgan, Z.. Morosini. P.. Palacin, C; Romera. 8.,
Taub, N., Vollebergh, W. A. The ESEMeD/MHEDEA 2000 Investigators (2004). Prevalence of mental
disorders in Europe: Results from the European Study of the Epidemiology of Mental Disorders
(ESEMeD) project. Acta Psychiatrica Scandinavica, 420 (Suppl), 21-27.
Angst, E, Stassen, H. H., Clayton, P. J.. & Angst. J. (2002). Mortality of patients with mood disorcers:
Follow-up over 34-38 years. Journal ofAffectiv« Disorders, 68, 167-181.
Angst. 1. (1986). The course of affective disorders. Psychopathologv, 19 (Supp/), 47-52.
Angst, J. (1997) Fortnightly review. A regular review of the long-term follow up of depression, British
Medical Journal, 315, 1143-1146.
Angst, 1., Baastrup, P.. Grof, P.. Hippius, H.. Poldinger,w., & Weis, P. (1973). The course of monopolar
depression and bipolar psychoses. Psychiatry. Neurology & Neurochirurgy, 76. 489-500.
Austin, M. P" Mitchell, P.. & Goodwin, G. M. (200 I). Cognitive deficits in depression: Possible irr pli-
cations for functional neuropathology. British Journal of Psychiatry. 178, 200-206.
Basso, M. R., & Bornstein, R. A. (1999a). Neuropsychological deficits in psychotic versus nonpsychotic
unipolar depression. Neuropsychology, 13, 69-75.
Basso, M. R" & Bornstein, R. A. (l999b). Relative memory deficits in recurrent versus first-episode major
depression on a word-list learning task. Neuropsychology, 13, 557-563.
Bingefors, K., Isacson, D., Knorring, L. V.. Srnedby, B., & Wicknertz. K. (1996). Antidepressant-treated
patients in ambulatory care. Mortality during a nine-year period after first treatment. British JOUT/wi of
Psychiatry, 169,647-654.
Bremner, J. D., Narayan. M.. Anderson, E. R., Staib, L. H., Miller, H. L.. & Charney, D. S. (2000).
Hippocampal volume reduction in major depression. American Journal of Psychiatry, 157, 115-118.
Bruce, M. L., Leaf, P. J.. Rozal. G. P., Florio, L., & Hoff. R. A. (1994). Psychiatric status and 9-year mor-
tality data in the New Haven Epidemiologic Catchment Area Study. American Journal of Psychiatry,
151,716-721.
Demyttenaere, K.. & Bruffaerts, R. (2002). Differential weight (If DSM-IV symptoms in major and minor
depression, Technical Report, Department of Psychiatry, KU Leuven.
Demyttenaere, K., & Bruffaerts, R. (20ma). Evolution of adherence in first and recurrent episode
depressive patients, Technical Report, Department of Psychiatry, KU Leuven.
Demyttenaere, K., & Bruffaerts, R. (2003b). Evolution (If symptom scores in first and recurrent episode
depressive patients. Technical Report. Department of Psychiatry, KU Leuven.
Eaton, W. W.. Anthony. J. C.. Gallo. J.. Cai. G.. Tien, A.. Romanoski. A.. Lyketsos, C., & Chen, L. S.
(1997), Natural history of Diagnostic Interview Schedule/DSM-IV major depression. The Baltirr.ore
Epidemiologic Catchment Area follow-up. Archives of General Psychiatry. 54, 993-999.
Fava, G. A., Rafanelli, C.. Grandi, S.. Conti. S.. & Belluardo, P. (1998). Prevention of recurrent depression
with cognitive behavioral therapy: Preliminary findings. Archives of General Psychiatry, 55, 816-820.
Frank, E" Prien, R. F.. Jarrett, R. B.. Keller. M. B.. Kupfer. D. J.. Lavori. P. W.. Rush. A. J., & Weissman,
M. M. (1991). Conceptualization and rationale for consensus definitions of terms in major depressive
disorder. Remission, recovery. relapse, and recurrence. Archives ()f General Psychiatry, 48, 851-855.
Giuffra, L. A., & Risch. N. (1994). Diminished recall and the cohort effect of major depression: A
simulation study. Psychological Medicine, 24, 375-383.
Glaser, R., Robles, T. F.. Sheridan. J.. Malarkey, W. B.. & Kiecolt-Glascr, J. K. (2003). Mild depressive
symptoms are associated with amplified and prolonged intlammatory responses after influenza vrus
vaccination in older adults. Archives (If General Psvchiatry, 60, 1009-1014.
Harkness, K. L., Monroe, S. M.. Simons. A. D.. & Thase, M. ( 1999). The generation of life events in recur-
rent and non-recurrent depression. Psychological Medicine. 29, 135-144.
Hasin, D, & Link, B. (1988). Age and recognition of depression: Implications for a cohort effect in major
depression. Psychologicul Medicine. 18,683-688.
w..
Judd, L. L., Akiskal, H. S.. Maser, J. D., Zeller, P. J.. Endicott, J., Coryell. Paulus, M. P., Kunovac, J.

38
 L., Leon, A. c, Mueller, T. L., Rice, 1. A., & Keller, M. B. (1998a). Major depressive disorder: A pros-
pective study of residual subthreshold depressive symptoms as predictor of relapse. Journal ofAffective
Disorders, 50, 97-108.
Judd, L. L., Akiskal, H. S., Maser, J. D., Zeller, P. 1., Endicott, 1., Coryell, w., Paulus, M. P., Kunovac, J.
L., Leon, A. C., Mueller, T. L., Rice, J. A., & Keller, M. B. (l998b). A prospective 12-year study of
subsyndromal and syndromal depressive symptoms in unipolar major depressive disorders. Archives of
General Psychiatry, 55, 694-700.
Judd, L. L., Paulus, M. J.. Schettler, P. J., Akiskal, H. S., Endicott, 1., Leon, A. C., Maser, J. D., Mueller,
T., Solomon, D. A., & Keller, M. B. (2000). Does incomplete recovery from first lifetime major de-
pressive episode herald a chronic course of illness? American Journal of Psychiatry, 157, 1501-1504.
Kandel, E. R. (1998). A new intellectual framework for psychiatry. American Journal of Psychiatry, 155,
457-469.
Kendler, K. S., & Gardner, C. a., Jr. (1998). Boundaries of major depression: An evaluation of DSM-IV
criteria. American Journal of Psychiatry, 155, 172-177.
Kendler, K. S., Thornton, L. M., & Gardner, C. a. (2000). Stressful life events and previous episodes in
the etiology of major depression in women: An evaluation of the "kindling" hypothesis. American
Journal of Psychiatry, 157, 1243-1251.
Kendler, K. S., Thornton, L. M., & Gardner, C. a. (2001). Genetic risk, number of previous depressive
episodes, and stressful life events in predicting onset of major depression. American Journal of
Psychiatry, 158, 582-586.
Kessing, L. V. (1998). Cognitive impairment in the euthymic phase of affective disorder. Psychological
Medicine, 28, 1027-1038.
Kessing, L. V. (1999). The effect of comorbid alcoholism on recurrence in affective disorder: A case
register study. Journal ofAffective Disorders, 53, 49-55.
Kessing, L. v., & Andersen, P. K. (1999). The effect of episodes on recurrence in affective disorder: A case
register study. Journal ofAffective Disorders, 53, 225-231.
Kessing, L. V., Andersen, E. W., & Andersen, P. K. (2000). Predictors of recurrence in affective disorder.
Analyses accounting for individual heterogeneity. Journal ofAffective Disorders, 57, 139-145.
Kessing, L. V., Andersen, P. K., & Mortensen, P. B. (1998). Predictors of recurrence in affective disorder.
A case register study. Journal ofAffective Disorders, 49, 101-108.
Kessing, L. v., Andersen, P. K., Mortensen, P. B., & Bolwig, T. G. (1998). Recurrence in affective disorder.
I. Case register study. British Journal of Psychiatry, 172,23-28.
Kessing, L. V., & Mortensen, P. B. (1999). Recovery from episodes during the course of affective disorder:
A case-register study. Acta Psychiatrica Scandinavica, lOa, 279-287.
Kessing, L. v., Mortensen, P. B., & Bolwig, T. G. (1998). Clinical definitions of sensitisation in affective
disorder: A case register study of prevalence and prediction. Journal ofAffective Disorders, 47, 31-39.
Klerman, G. L., & Weissman, M. M. (1989). Increasing rates of depression. Journal of the American
Medical Association, 261, 2229-2235.
Kraepelin, E. (1921). Manic-Depressive Insanity and Paranoia. Edinburgh: E & S Livingstone.
Lavon, P. W., Dawson, R., & Mueller, T. B. (1994). Causal estimation of time-varying treatment effects in
observational studies: Application to depressive disorder. Statistics in Medicine, 13, 1089-1100.
MacQueen, G. M., Campbell, S., McEwen, B. S., Macdonald, K., Amano, S., Joffe, R. T., Nahmias, C; &
Young, L. T. (2003). Course of illness, hippocampal function, and hippocampal volume in major
depression. Proceedings of the National Academy of Sciences of the United States of America, lOa,
1387-1392.
Mitchell, P. B., Parker, G. B., Gladstone, G. L., Wilhelm, K., & Austin, M. P. (2003). Severity of stressful
life events in first and subsequent episodes of depression: The relevance of depressive subtype. Journal
ofAffective Disorders, 73, 245-252.
Mueller, T. I., Leon, A. c.. Keller, M. B., Solomon, D. A., Endicott, 1., Coryell, w., Warshaw, M., & Maser,
J. D. (1999). Recurrence after recovery from major depressive disorder during 15 years of observational
follow-up. American Journal of Psychiatry, 156, 1000-1006.
Osby, U., Brandt, L., Correia, N., Ekbom, A., & Sparen, P. (2001). Excess mortality in bipolar and unipolar
disorder in Sweden. Archives of General Psychiatry, 58, 844-850.
Paradiso, S., Lamberty, G. 1., Garvey, M. 1., & Robinson, R. G. (1997). Cognitive impairment in the
euthymic phase of chronic unipolar depression. Journal ofNervous and Mental Disease, 185, 748-754.
Porter, R. 1., Gallagher, P., Thompson, J. M., & Young, A. H. (2003). Neurocognitive impairment in drug-
free patients with major depressive disorder. British Journal of Psychiatry, 182, 214-220.

39
Post, R. M., Rubinow, D. R., & Ballenger, 1. C. (1986). Conditioning and sensitisation in the longitudinal
course of affective illness. British Journal of Psychiatry, 149, 191-201.
Post, RM. (1992). Transduction of psychosocial stress into the neurobiology of recurrent affective
disorder. American Journal of Psychiatry, 149,999-1010.
Racine, R (1978). Kindling: The first decade. Neurosurgery, 3, 234-252.
Raison, C. L., & Miller, A. H. (2003). When not enough is too much: The role of insufficient glucocorticoid
signaling in the pathophysiology of stress-related disorders. American Journal of Psychiatry, 160,
1554-1565.
Rice,1. P., Rochberg, N., Endicott, J., Lavori, P. W, & Miller, C. (1992). Stability of psychiatric diagnoses.
An application to the affective disorders. Archives of General Psychiatry, 49, 824-830.
Rovner, B. W., German, P. s.. Brant, L.1., Clark, R., Burton, L., & Folstein, M. E (1991). Depression and
mortality in nursing homes, Journal of the American Medical Association. 265, 993-996. Erratum in
Journal of the American Medical Association, 265, 2672.
Sadler, 1. Z., Hulgus, Y. E, & Agich, G. J. (1994). On values in recent American psychiatric classification.
Journal of Medicine and Philosophy, 19, 261-277.
Sapolsky, R. M. (2000). Glucocorticoids and hippocampal atrophy in neuropsychiatric disorders. Archives
of General Psychiatry, 57, 925-935.
Sapolsky, R M., Krey, L. C, & McEwen, B. S. (1986). The neuroendocrinology of stress and aging: The
glucocorticoid cascade hypothesis. Endocrine Reviews, 7, 284-301.
Schatzberg, A. E, Posener, J. A., DeBattista, c., Kalehzan, B. M., Rothschild, A. J., & Shear, P. K. (2000).
Neuropsychological deficits in psychotic versus nonpsychotic major depression and no mental illiess.
American Journal of Psychiatry, 157, 1095-1100.
Schmitt, 1. A, Kruizinga, M. J., & Riedel, W J. (2001). Non-serotonergic pharmacological profiles and
associated cognitive effects of serotonin reuptake inhibitors. Journal of Psychopharmacology, IS, 173-
179.
Schoevers, R. A., Geerlings, M. I., Beekman, A. Too Penninx, B. W, Deeg, D. J., Jonker, c, & Van Tilhurg,
W. (2000). Association of depression and gender with mortality in old age. Results from the Amsterdam
Study of the Elderly (AMSTEL). British Journal of Psychiatry, 177, 336-342.
Sheline, Y. I. (2000). 3D MRI studies of neuroanatomic changes in unipolar major depression: The role of
stress and medical comorbidity. Biological Psychiatry, 48, 791-800.
Sheline, Y. I. (2003). Neuroimaging studies of mood disorder effects on the brain. Biological Psychiatry,
54, 338-352.
Sheline, Y. I., Gado, M. H., & Kraemer, H. C. (2003). Untreated depression and hippocampal volume loss.
American Journal of Psychiatry. 160, 1516-1518.
Sheline, Y. I., Gado, M. H., & Price, 1. L. (1998). Amygdala core nuclei volumes are decreased in recurrent
major depression. Neuroreport, 9, 2023-2028. Erratum in Neuroreport, 9, 2436.
Sheline, Y. I., Sanghavi, M., Mintun, M. A., & Gado, M. H. (1999). Depression duration but not age
predicts hippocampal volume loss in medically healthy women with recurrent major depression.
Journal of Neuroscience, 19,5034-5043.
Sheline, Y. I., Wang, P. W, Gado, M. H., Csernansky, J. G., & Vannier, M.W (1996). Hippocampal atrophy
in recurrent major depression. Proceedings of the National Academy ofSciences of the United States of
America, 93, 3908-3913.
Solomon, D. A., Keller, M. B., Leon, A. C; Mueller, T I., Lavori, P. Woo Shea, M. T, Coryell, W., Warshaw,
M., Turvey, c., Maser, J. D., & Endicott, J. (2000). Multiple recurrences of major depressive disorder.
American Journal of Psychiatry, 157, 229-233.
Solomon, D. A., Keller, M. B., Leon, A. C., Mueller, T I., Shea, M. T, Warshaw, Moo Maser, 1. D., Coryell,
W., & Endicott, J. (1997). Recovery from major depression. A 10-year prospective follow-up across
multiple episodes. Archives of General Psychiatry, 54, 100I-I 006.
Spijker, J., de Graaf, R., Bijl, R. v., Beekman, A. T, Ormel, J., & Nolen, W. A. (2002). Duration of major
depressive episodes in the general population: Results from The Netherlands Mental Health Survey and
Incidence Study (NEMESIS). British Journal of Psychiatry, 181, 208-213.
Starkman, M. N., Giordani, B., Berent, S., Schork, M. A., & Schteingart, D. E. (2001). Elevated cor:isol
levels in Cushing's disease are associated with cognitive decrements. Psychosomatic Medicine, 63,
985-993.
Starkman, M. N., Giordani, B., Gebarski, S. S., & Schteingart, D. E. (2003). Improvement in learning
associated with increase in hippocampal formation volume. Biological Psychiatry, 53, 233-238.
Swann, A. c., Secunda, S. K., Stokes, P. E., Croughan, 1., Davis, J. M., Koslow, S. H" & Maas, 1. W.

40
 (1990). Stress, depression, and mania: Relationship between perceived role of stressful events and
clinical and biochemical characteristics. Acta Psychiatrica Scandinavica, 81, 389-397.
Tham, A., Engelbrektson, K., Mathe, A. A., Johnson, L., Olsson, E., & Aberg-Wistedt, A. (1997). Impaired
neuropsychological performance in euthymic patients with recurring mood disorders. Journal of
Clinical Psychiatry, 58, 26-29.
Vermetten, E., Vythilingam, M., Southwick, S. M., Charney, D. S., & Bremner, J. D. (2003). Long-term
treatment with paroxetine increases verbal declarative memory and hippocampal volume in
posttraumatic stress disorder. Biological Psychiatry, 54, 693-702.
Whooley, M. A., & Browner, W. S. (1998). Association between depressive symptoms and mortality in
older women. Study of Osteoporotic Fractures Research Group. Archives of Internal Medicine. 158,
2129-2135.
Wulsin, L. R., Vaillant, G. E., & Wells, V. E. (1999). A systematic review of the mortality of depression.
Psychological Medicine, 61, 6-17.

41
 Chapter 2

Mood and Memory: A Cognitive Psychology Perspective

on Maintenance of Depressed Mood

and Vulnerability for Relapse

Dirk Hermans, Filip Raes, & Paul Eelen

Depression is a seriously disabling disorder with high incidence and prevalence

rates. The National Comorbidity Study revealed a lifetime prevalence of 17.1% in
the general American population (Blazer, Kessler, McGonagly, & Swartz, 1994).
Similar results have been found in the Belgian and Dutch community (Bayingana,
Drieskens, & Tafforeau, 2002; Bijl, van Zessen, & Ravelli, 1997; see also Chapter
I, this volume). An important difference with anxiety disorders is that even without
treatment most depressive episodes disappear after a number of months. But at the
same time, depression is associated with a relatively high risk of relapse/recurrence.
Patients who recovered from a first depressive episode have about a 50% chance of
developing a new episode, whereas this risk of recurrence is already 70-80% for
those who have a history of two or more depressive episodes (Belsher & Costello,
1988). Also, for about 6% to 13% of the patients depression becomes chronic, and
every new episode entails a new risk of chronicity.
These figures indicate that interventions should not only target the treatment of
current depressive episodes, but also, or even more so, the primary and secondary
prevention of depression. Such prevention programs require solid knowledge of the
vulnerability factors associated with the development of depression, as well as of
the mechanisms that maintain (chronic) depression or that put individuals at risk of
relapse/recurrence. Thus, in accordance with the changing view of depression as a
recurrent, highly disabling and often chronic or life-long disorder (e.g., Judd,
Akiskal, & Paulus, 1997), throughout this chapter our focus will be on the
prevention of relapse/recurrence and on the study of vulnerability factors of
depression. If researchers from different disciplines or different theoretical
paradigms strive towards integration in the conceptualization and treatment of
depression, we believe that the area of relapse prevention and vulnerability research
is an important domain in which this integration could be pursued.
One of the models that has attracted a lot of attention during the last three
decades, and that has proven fruitful in developing and testing hypotheses
concerning the risk factors and mechanisms of depression, is the cognitive-
behavioral model developed by Beck (Beck, Rush, Shaw, & Emery, 1979).
Although this model has always in the first place been a clinical theory, it has been
heavily inspired by experimental cognitive psychology from the start. This is most
evident from the concepts used to describe the model's crucial constructs and
mechanisms (e.g., memory schemata, automatic negative thoughts, dysfunctional
beliefs). In spite of this obvious inspiration by cognitive science, research
concerning the cognitive-behavioral model of depression has largely relied on self-
report and correlational research designs. Quite simultaneously, there has been a
research tradition that was less clinically inspired, but more experimental in nature,
and more rooted in the general study of human behavior and cognition. These
studies primarily aimed at revealing the information-processing characteristics of
depression and depressed mood (for an overview, see Williams, Watts, Macl.eod, &
Mathews, 1997). Examples of such research topics are: the study of mood and
memory, depression and selective attention, processes of interpretation and
reasoning, learning mechanisms in depression, and processes of social cognition.
As a matter of fact, this field of research turned out to be quite fruitful, and not
without clinical implications. It is thus not surprising that during the last decade
this more experimental approach got increasingly integrated within the cognitive-
behavioral theory (and to a lesser extent also therapy) of depression (e.g., Clark &
Beck, 1999). An experimental cognitive psychology approach surely represents in
our opinion a valuable adjunct to other approaches in the study of depression and
its vulnerability factors (e.g., psychodynamic or personality psychology
approaches, neurobiology, etc.). Biological, genetic, social and psychological
factors interact, and do not just act as independent factors that simply add up to the
complex picture of depression (e.g., Weiss, Longhurst, & Mazure, 1999). Thus, the
respective approaches studying these different factors have to be seen as more than
just adjuncts to one another, given the complexity of depression in general and of
all mechanisms and processes underlying its etiology and course. Future research
should then aim at truly integrating these different approaches, and not just place
them along side one another.
In the present chapter we will focus on one particular topic in the domain of
information processing in depression and depressed mood, namely the study of

44
memory processes in depression. In particular, we will cover research on explicit
memory. The study of implicit memory in depression, which has attracted a lot of
attention during the last ten years, falls outside the scope of our discussion (for a
review, see Watkins, 2002). Two separate but related lines of research will be
highlighted. First, we will discuss the experimental research tradition on the
interaction between mood and memory. In the second part we will discuss more
recent work on autobiographical memory specificity. In doing so, we will pay
particular attention to the clinical implications of this research with respect to the
maintenance of depression and vulnerability for relapse.

Mood and Memory

In 1981, Gordon Bower published an influential article in American Psycho-

logist that can be regarded as the seminal paper for all mood and memory research.
Bower opened his overview of research with a quite pleasant anecdote (Bower,
1981, p. 129). It referred to the movie City Lights in which Charlie Chaplin played
the role of a little tramp and saved a drunk from leaping to his death. The drunk,
who turned out to be a millionaire, befriends Charlie, and the two spend the evening
together drinking. The next day, when he was sober, the millionaire does not
recognize Charlie and ignores him. Later, when the millionaire got drunk again, he
spotted Charlie and treats him as his long-lost companion. The two get drunk again
and Charlie stays over to sleep in the millionaire's mansion. But, the next morning,
the sober millionaire again does not recognize Charlie, treats him as an intruder and
has the butler kick him out. The scene ends with the little tramp telling the audience
his opinion of high society and the evils of drunkenness.
For Gordon Bower, the story exemplified an interesting case of "context-
dependent memory". This more general phenomenon of context dependency was
already known from the work of Godden and Baddeley (1975) who asked subjects
to learn lists of words either on the beach or under approximately 4.5 meters of
water. They were then asked to recall the list learned in one context (e.g., on the
beach) while in the same context on some occasions (i.e., on the beach) and while
in a different context on other occasions (in this case, under water). Godden and
Baddeley observed that there was a decrement in recall of words of over 30% if the
context was changed between learning and recall. The important difference
between the study of Godden and Baddeley and the tale of Charlie Chaplin and his
millionaire friend is that, in the former case, context was defined as an external
situation (on the beach vs. under water), whereas in the latter, context could be
regarded as an internal state (drunk vs. sober). Bower's work aimed at the question
of state-dependent memory, and more specifically of whether moods and emotions
can also act as internal contexts that facilitate or inhibit memory retrieval.
As we will argue later, this question has important clinical relevance with respect
to the understanding of the mechanisms involved in the maintenance of depression

45
and the vulnerability of depressive relapse (see also Gotlib & Krasnoperova, 19(8).
Hence, our discussion of the "mood and memory" literature will be specifically
focused on the studies that address this very issue of depression.
In this context, Williams (1992) gives an interesting case example. He quotes the
case of a patient who, at one therapy session, recalled a swimming event as
particularly pleasant. At that moment, she was in a good mood. At a later session,
however, when she was in a more depressed mood, she remembered different
elements of the same swimming event which made it seem rather unpleasant. She
described that she was afraid of making herself look ridiculous, and that it had been
embarrassing to see herself in a swimsuit because of her weight problem. This case
probably illustrates a more general experience that is shared by many patients
suffering from depression and undoubtedly also by many nondepressed persons:
when you feel down, you seem to remember more negative experiences from your
past than when you feel elated. It is as if a depressed mood is a filter that sifts your
personal memories.
If one takes a closer look at the swimming example, one will notice that it is
conceptually somewhat different from the state-dependent memory phenomenon.
Mood state dependency refers to the fact that stimuli are recalled better if the mood
during recall is the same as the mood during encoding. This is independent of
the affective nature of the stimuli that are encoded and retrieved. The swimming
event example essentially differs from state dependency in that the emotional
tone of the retrieved events is now relevant. Events that are congruent with
the current mood are retrieved more easily. We will refer to this latter pheno-
menon as mood congruency. In fact, two types of mood congruency can be
distinguished: mood congruent encoding and mood congruent recall. After we have
briefly discussed Bower's theory on mood and memory, we will illustrate
both types of mood congruency, as well as mood state dependency, and we will
highlight how these might play a role in the maintenance and exacerbation of
depression.

Mood and the Associative Memory Network

Theories of human memory have been diverse. In addition to more recent

connectionist models, the associative network view of emotion has been one of the
most important and fruitful ways of conceptualizing the way in which knowledge
is stored in memory (Schwartz & Reisberg, 1991). According to the associative
network view, memory is a vast web or network of ideas. The network is built up
of units of knowledge (remembered episodes, remembered facts, sights, etc.),
referred to as "nodes", and associations or associative links between these nodes.
The way this network is structured is different for each individual and will be
dependent upon the learning experiences throughout the individual's life. Some
associative links will be relatively weak, whereas repeated experience (e.g., co-
occurrence) will have made other associations quite strong. When a unit of

46
knowledge is activated - for instance the node "cat" upon seeing a cat - one can
become aware of the content of that node if it has received sufficient input. Also,
through spreading of activation to associatively related nodes, one can become
aware of other things. For instance, seeing a cat might "remind" you to buy food
for your own cat. In that case, spreading of activation has made other mental
contents more accessible to awareness. The activated concept "cat" might have
spread activation to the related concept "cat food" which activated the stored
autobio-graphical event that you ran out of cat food.
Contexts can exert an influence on memory if they are encoded together
with specific knowledge. The accessibility of this knowledge for retrieval might
be enhanced if the context is again present during retrieval. This is because,
through spreading of activation, the context-nodes will already pre-activate the
relevant knowledge. This pre-activation is absent when the context is absent and,
hence, retrieval shall be somewhat less successful. This is basically how the
associative network deals with observations like those of Godden and Baddeley
(1975).
The crux of Bower's theory (1981) is that emotions (mood) can be consi-
dered as individual nodes in the associative memory network. They are embedded
in memory representations of semantic knowledge, autonomic activity and
muscular activity. Information that is encoded during emotional states will likely be
encoded in association with the related "emotion node". Through the spreading of
activation, emotions can then serve as "contexts" and thus influence memory
performance.
For a more detailed description of Bower's "Network Theory of Affect" we refer
to Bower (1981, 1991) and Gilligan and Bower (1984). For the present purposes it
suffices to recapitulate the three most important hypotheses that were derived from
this theory. In subsequent paragraphs we will discuss experimental evidence for
each.
I. State-dependent learning: Emotions serve as contextual cues in learning and
recall. Memory is facilitated when mood state at learning matches mood at
recall.
2. Mood congruent recall: Material with affective tone that is congruent with
current mood is most easily retrieved from memory.
3. Mood congruent encoding: Material with affective tone that is congruent with
current mood is most easily learned (i.e., encoded in memory).

State-dependent Learning

One of the first studies that demonstrated state-dependent learning - sometimes

referred to as "state-dependent memory", given the relevance of both encoding and
retrieval in this phenomenon - stems from Bower, Monteiro and Gilligan (1978;
Experiment 3). In this study a group of 24 participants was asked to study two lists
of 16 words. When studying the first list (List A) a positive mood was induced for

47
half of the participants, whereas a sad mood was induced for the other participants.
The mood induction procedure employed in this study consisted of hypnosis
combined with self-generated imagery. Subsequently, they were asked to study a
second list (List B) while under the same or the opposite mood. After these two
learning phases, participants were asked to recall as many words as possible from
List A. During this free recall phase, participants were again induced with either a
sad or happy mood. Results showed that retention was statistically superior when
memory during the test phase matched the mood they were in when studying List
A. As compared to mood controls, retention was impaired when mood during
studying and testing List A mismatched. These results provided clear support for
the hypothesis that mood can act as a context for the retrieval of memories, A
crucial aspect of this experiment was that participants were asked to study two
different lists, which is why it is also referred to as the "interference paradigm".
Other studies that only employed one word list failed to demonstrate a state-
dependency effect. This points to the fact that the emotional mood was a helpful
feature in distinguishing the target material (List A) from interfering material (List
B). This multiple-list situation seems a closer analogue to real life, where one stores
a multitude of experiences while in different moods.
Later studies have replicated the mood state-dependent effect as well as the
importance of the interference paradigm in obtaining this effect (e.g., Share,
Lisman, & Spear, 1984). Singer and Salovey (1988) provide an overview of these
studies and conclude that the existing data support the mood state-dependency
hypothesis. When translated to the understanding of depression, these data suggest
that information that was encoded during a depressive episode will be more
accessible when one is in a negative mood again later on. Given that experiences
during an episode of depression are more often than not negatively toned, the higher
accessibility of their memory traces during a subsequent negative mood will most
likely fuel this negative emotional state.
Given that the majority of studies on state-dependency have employed
procedures like hypnosis or imagery to induce moods, and have used stimulus
materials like neutral words. phrases, nonsense syllables or pictures of neutral
objects, one can question the ecological validity of this phenomenon. A study by
Weingartner, Miller and Murphy (1977) is worth mentioning in this respect. They
asked a group of patients with bipolar disorder to freely associate to two nouns in
states of depression or mania. Later, they had to recall these associations in sta:es
that were the same or different from which they were originally generated. These
patients showed superior recall when in the same state as their learning state.

Mood Congruent Recall

Several investigations have examined the relation between depression and the
time taken to retrieve personal memories. Lloyd and Lishman (1975) presented

48
neutral cue words to a group of 40 depressed inpatients and instructed them to think
of either a pleasant or unpleasant memory, and to signal to the experimenter when
a suitable memory came to mind. The time taken to retrieve memories was recorded
by stopwatch. They observed that the more severe the depression, as measured by
the Beck Depression Inventory (BOI), the faster the patient retrieved an unpleasant
memory (relative to the pleasant memories). Furthermore, more depressed patients
retrieved more intense negative memories.
Although these data fit nicely with Bower's theory, there are at least two
interpretative problems associated with this study. First, it can not be excluded that
the more severely depressed patients may have had more genuinely depressing
experiences during their life, and, hence, find it easier to retrieve anyone of them.
Second, the more severely depressed patients may simply be interpreting more of
their neutral or ambiguous experiences as more depressive, thus inflating the
number of memories to choose from. In this latter case the data would be sup-
porting an interpretational bias rather than a memory bias.
These alternative explanations were taken into account in an elegant study by
David Clark and John Teasdale (1982). They selected depressed patients for the
presence of diurnal variation of mood. At several moments they asked these patients
to respond with the first personal memory that came to mind. These memories were
later rated for pleasantness. It was observed that happy memories were less
probable (and depressing memories more probable) when patients were more
depressed. However, when the same patients were at the less depressed point in
their cycle, the picture was reversed. These results cannot be explained in terms of
differential frequencies of depressive experiences, because the data stem from the
same person at different moments. With respect to the possibility of an inter-
pretational bias, the pleasantness ratings were indeed also mood-dependent. The
more depressed the current mood, the more negative these ratings, but this effect
was by itself not sufficient to explain the memory bias.
Other studies have approached the mood congruent recall hypothesis using
experimentally induced moods within the laboratory. These studies have the
advantage that they can more easily control for mood during encoding. One such
example is a study by Teasdale and Russell (1983). In this study, nondepressed
college students were asked to study a list of positive and negative trait words.
During the second phase of the study an elated or depressed mood was induced,
during which participants were asked to recall as many words as possible from the
list. The memory data showed a clear effect of mood congruent retrieval:
"depressed" participants recalled more negative than positive words, whereas the
elated participants recalled more positive than negative words. This mood con-
gruency effect has been replicated numerous times and has proven to be a quite
robust phenomenon (for overviews, see Blaney, 1986; Matt, Vazquez, & Campbell,
1992; Singer & Salovey, 1988). One particular finding is that these effects are
stronger when the stimuli are self-referent. It is possible that self-reference induces
more elaborate processing, which would be necessary for the effect to come to
light. Recent research also indicates that mood congruency cannot be reduced to a
heuristic response bias, but truly reflects an increased sensitivity to mood congruent
information (Fiedler, Nickel, Muehlfriede1, & Unke1bach, 2001).

49
 Still, how do the findings on mood congruent retrieval relate to the suggestion
that people suffering from depression are often more accurate in perceptions and
judgments than nondepressed people? Actually, it is sometimes suggested that the
bias resides in nondepressed people, who would tend to look at life through rose-
colored glasses. According to this theory of "depressive realism", depressed people
would be more even-handed, more realistic. Results from a meta-analysis of mood
congruency studies by Matt et al. (1992) shed some light on this proposal. They
concluded that nondepressed persons recall between 6% and 8% more positive than
negative stimuli. Subclinically depressed persons tum out to be even-handed; they
show a balanced recall of affectively valenced stimuli. For clinically depressed
patients, however, a negatively toned bias can be observed, that is estimated at up
to 10% more negative than positive memories.
Given these figures, it is not unrealistic to assume that the process of mood
congruent recall has an important role in maintaining and intensifying depressed
mood. The depressed patient might get stuck in a vicious circle in which depressed
mood heightens the accessibility of negative memories, which in tum amplify the
depressed mood.

Mood Congruent Encoding

Besides mood congruency during retrieval, mood might also influence the type
of information that is encoded. According to the "mood congruent encoding"
hypothesis, material with affective tone that is congruent with current mood is most
easily learned. This too might be part of a vicious circle, in which the depressed
patient spirals between a disproportionate encoding of negative memories and an
escalating negative mood.
Because of the difficulty of control over naturally occurring moods at encoding,
most of the studies on mood congruent encoding have been conducted in the
laboratory. In these experiments, moods are induced followed by a task involving
the reading of a story or the learning of a word list. Then, after the induced mood
has worn off, participants are tested for recognition or recall of this previously
learned material. A good example of mood congruent encoding can be found in the
work of Bower, Gilligan and Monteiro (1981). The mental health professionals and
college students who participated in this study read some simulated psychiatric
interviews. The patient in the narrative briefly describes a series of unrelated happy
and sad incidents from his life. During reading, participants were either in a happy
or sad mood, which was experimentally induced. Later, in a neutral mood, they
were asked to recall the narrative. The results were clear-cut: "happy" readers
recalled about one and a half times as many happy incidents as sad events, whereas
"sad" readers recalled about one and one third times as many sad as happy
incidents. Thus, it can be concluded that being in a sad or happy mood significantly
influences the type of information that is encoded. Later studies have repeatedly
replicated this effect.

50
 One particular problem associated with some of these studies is that their effects
might (in part) be dependent on experimenter demand (Bower, 1987). When
participants are induced to feel happy or sad by an autosuggestion technique (using
either hypnosis, guided fantasy or the Velten technique), they may believe that to
maintain that mood throughout the subsequent learning phase they should attend
more to mood-sustaining material; and as a consequence they learn it better.
However, the data of several studies can not readily be explained by this alternative
explanation. In these studies, mood was induced in a subtle manner, so that demand
effects cannot be invoked to explain the results (e.g., Forgas & Bower, 1987).

Mood and Memory: Some Concluding Remarks

The seminal work of Gordon Bower has been an impetus for several productive
lines of research concerning the interaction between explicit memory and mood.
These studies have particular relevance for the study of depression. The phenomena
of state dependency and mood congruency, when active, will contribute to the
perpetuation or exaggeration of existing mood states. It seems likely that mnemonic
selectivity occurs in depressed persons and that it contributes to the vicious cycle
of deepening depression (Blaney, 1986). When depressed, state dependency will
make information that was encoded during previous depressive episodes more
accessible. Specific memories and feelings that were associated with these
preceding episodes come to mind more easily. It might seem that feelings and
thoughts that were previously encoded in this depressed state seem to "return". This
is sometimes obvious in the experiences of patients who experience relapse and
from the first moment on have the feeling that "everything seems to come back".
In addition to state dependency, mood congruency will further darken a
depressed individual's mood. When depressed, mood congruent retrieval will color
the type of memories that are remembered. This may make one's past life seem
much more negative than it actually was. Negative experiences with others will be
selectively remembered over positive encounters, affecting social behavior. One's
own qualities will be interpreted against a predominant accessibility of memories
of failures over successes, impacting self-esteem and amplifying feelings of
hopelessness. And finally, mood congruent encoding will further increase this
imbalance. Selective learning of negative materials will make a person more
vulnerable because this process actually leads to an objectively higher amount of
negative memories in long-term memory. Each of these processes may lead to a
vicious circle, and each circle can further feed into each of the other cycles
(Wachtel, 1994, 1997).
It will be clear that these processes will only account for part of the processes
that are responsible for the maintenance and amplification of depressed moods.
Moreover, the existence of other regulatory processes is necessary. If not, one
would have to predict that once one starts to drift away from a neutral mood, there

51
will be an unstoppable spiral towards a more and more negative mood, ending up
in severe depression. This is surely not the case, and even the most severe of clinical
depressions will ameliorate at some point. The study of "mood repair" provides
insight in the processes that regulate mood. Under normal circumstances different
types of processes will keep mood changes within an acceptable range. The
discussion of these processes, however, falls outside the scope of this chapter. For
a recent discussion of mood-regulation strategies and mood-congruent memory, we
refer to Rusting and DeHart (2000). In the remainder of this chapter, we will focus
on the study of the specificity of autobiographical memory, which is a research
tradition that was originally rooted in mood and memory research.

Autobiographical Memory Specificity

In the context of mood and memory research, Williams and Broadbent (1986)
studied patients within the first few days after they were hospitalized because of a
severe suicide attempt by overdose. For most of these patients the suicide attempt
had taken place against the background of a depressive disorder. During this study,
they were asked to retrieve specific autobiographical memories to positive or
negative cue words (e.g., happy, angry). Response latencies to retrieve these
memories were compared with those of a group of nondepressed patients from the
same hospital, most of whom were hospitalized for physical investigations, as well
as with a group of nondepressed volunteers who were not hospitalized. Results
from this cuing task showed that the parasuicide patients took longer than controls
to retrieve a memory when given a positive word. This was in line with the mood
congruency hypothesis in that memories that were incongruent with the current
mood were less accessible.
Upon closer inspection of their data, Williams and Broadbent discovered a
striking qualitative difference in the memories produced by the depressed and
nondepressed groups, in addition to the more quantitative differences in response
speed. Despite the fact that they were asked to retrieve specific memories, the
overdose patients tended, as a first response to the cues, to retrieve inappropriately
general memories. Whereas controls produced specific memories, like "When I
passed my driving license (successful)" or "With my supervisor on Monday
(angry)", the depressed patients typically reported overgeneral memories, e.g., "In
my job (successful)" or "When I have had a row (angry)". The difference in memo-
ry specificity between depressed and nondepressed persons was observed for both
positive and negative cue words. Also, this effect could not be attributed to phanna-
cological effects of the overdose, because no differences were observed between
both groups on two other measures of semantic memory that are known to be
sensitive to the effects of drugs.
A few years later, in 1988, the Williams research group published three separate
papers that replicated and extended these findings. In the first paper, Williams and

52
Dritschel (1988) tested a group of 24 overdose patients within the first four days
fol1owing the self-poisoning. The procedure was similar to the procedure employed
by Williams and Broadbent (1986), except that participants were now given only 30
instead of 60 seconds to retrieve a specific memory. If the first response was not a
specific memory, the participants were prompted to produce one ("Can you think
of a specific time - one particular event?"). After each subsequent inappropriately
overgeneral memory the participant was prompted again. As compared to the
control group, the overdose patients again produced less specific autobiographical
memories. A similar effect was observed in a second "overdose" group that took
part in this study. These participants were tested 3 to 14 months after deliberate self-
poisoning and were included to test whether lack of memory specificity was related
to the presence of an ongoing emotional crisis. This group, however, displayed a
lack of autobiographical memory specificity comparable to the "recent overdose"
group. Although both overdose groups displayed similar elevated levels of de-
pressed mood, these patients were not selected on the basis of a current depressive
disorder. In a subsequent study, autobiographical memory was investigated in a
group of patients who met criteria of Major Depressive Disorder according to
DSM-III-R (Williams & Scott, 1988). This study provided clear evidence for a lack
of memory specificity in c1inical1y diagnosed depressed patients. Similar results
were observed by Moore, Watts and WiIliams (1988).
The cue word paradigm first used by WiIliams and Broadbent (1986) has become
known as the Autobiographical Memory Test (AMT). Subsequent studies by other
research groups employed variants of this procedure, typical1y involving the
presentation of 10 or more cue words, alternating in valence (positive or negative).
For each cue word, the participant is asked to retrieve a specific memory that refers
to an event that happened once and lasted not longer than one day (de Decker,
Hermans, Raes, & Eelen, 2003). Different ways of categorizing the responses have
been employed. In our own research, the first response is coded as either a specific
memory or a non-specific memory. The analyses are based on the number of
specific first responses. For exploratory reasons, non-specific memories are further
qualified as either a categoric memory (referring to events that occurred more often,
and are difficult to date exactly, e.g., "Watching my favourite talk show on Friday
evenings"), an extended memory (referring to events that last longer than one day,
e.g., "My holiday in Senegal two years ago"), no memory (e.g., a verbal association
to the cue, thoughts about the future), no response (no response is given to the cue,
"T", "I don't know", etc.), or same event (a response referring to the same event
already mentioned in response to a previous cue). Using this scoring procedure,
Raes, Hermans, de Decker, Eelen and WiIliams (2003) obtained a very good inter-
rater agreement of 98.8% (kappa coefficient of .957) for the categorization of
specific vs. non-specific responses. Other researchers have focused on the number
of categoric memories and refer to a relatively higher proportion of these non-
specific memories as "overgeneral autobiographical memory". I Therefore, we wiII
I It is important to note that increased overgenerality in clinical groups is typically attributable to an excess

of categoric memories, and less of extended memories. This is also important with respect to the
theoretical models concerning the processes involved in overgeneral memory retrieval (e.g., Williams,
1996).

53
use the terms "overgeneral memory" and "lack of memory specificity" interchange-
ably to refer to these related results.
Since the seminal work by Williams and colleagues, several studies employing
the AMT replicated the lack of memory specificity in groups of (clinically)
depressed adults (e.g., Brewin, Watson, McCarthy, Hyman, & Dayson, 1998;
Goddard, Dritschel, & Burton, 1996; Kuyken & Dalgleish, 1995; Puffet, Jehin-
Marchot, Timsit-Berthier, & Timsit, 1991; Wessel, Meeren, Peeters, Arntz., &
Merckelbach, 2001). Recent research has replicated this finding in adolescents with
major depressive disorder (Park, Goodyer, & Teasdale, 2002).
Besides the studies that focused on depression, other studies have investigated
autobiographical memory specificity in anxiety disorders. However, to date, over-
generality does not seem to be a characteristic of anxiety disorders such as
obsessive-compulsive disorder (Wilhelm, McNally, Baer, & Florin, 1997) and
generalized anxiety disorder (Burke & Matthews, 1992). The fact that overgene-
rality is linked to major depressive disorder rather than anxiety disorder was
recently confirmed by Wessel et al. (200 I). On the basis of this latter study it could
also be concluded that overgeneral memory is related to the presence of a diagnosis
of depressive disorder but not to the level of depression. Besides the strong link
between autobiographical memory specificity and depression, there also seems to
be a strong association with trauma. Several studies have shown that lack of
specificity (or overgenerality) is inversely related to different forms of past trauma
experience (e.g., sexual, physical, and parental abuse) (Dalgleish et al., 2003; de
Decker et al., 2003; Henderson, Hargreaves, Gregory, & Williams, 2002; Hermans
et aI., 2004). This particular relation with trauma will be further discussed later on.
In the following sections we will focus on the possible clinical relevance of
reduced autobiographical memory specificity in depression. Is it merely a by-
product of depression, or does it represent a more stable characteristic of persons
vulnerable to depression? Does it have any clinical diagnostic relevance? How is
overgeneral memory engaged in the maintenance and relapse of depression?

Lack of Memory Specificity as a Trait Marker for Depression?

If one could identify the factors that make people vulnerable to depression or
depressive relapse, screening and primary and secondary prevention interventions
could be more efficiently implemented. In the past, a number of psychological
characteristics have been assessed for whether they represent stable characteristics
of depressed patients and place individuals at risk of depressive onset or relapse.
Many of these studies have compared individuals in remission from depression
with groups of currently depressed and never depressed persons. Most of these
cognitive factors (e.g., dysfunctional attitudes), however, turned out to be state-
dependent. Together with remission of the depressive symptoms, these candidate
risk factors disappeared and no longer differentiated between previously depressed

54
and never depressed individuals. With respect to the study of cognitive schemas as
vulnerability factor for depression, these findings of normalization have led some
researchers to hypothesize that these dysfunctional schemas are "latent" and need
to be activated before they can be measured. Activation by priming techniques
(most often mood induction) generally demonstrates that formerly depressed
individuals differ from nondepressed in the negativity of their responses on
information-processing tasks or questionnaires (Hammen, 2001). But only a few
studies (e.g., Segal, Gemar, & Williams, 1999) have actually demonstrated the
association between such negativity and a risk of subsequent depression.
Within the study of autobiographical memory specificity, several studies have
indicated that the presence of overgeneral memory specificity is not state-depen-
dent, but represents a relatively stable characteristic. In the previously discussed
study by Williams and Dritschel (1988), for instance, memory specificity of the
overdose patients that were tested several months after the suicide attempt could not
be distinguished at all from the results of the "recent overdose" group. However,
this study was cross-sectional and no strong inferences can be drawn because a
number of the "ex-patients" were still depressed. In a later longitudinal study by
Brittlebank, Scott, Williams and Ferrier (1993) depressed patients were tested on
admission, 3 months and 7 months later. Although the depressive symptomatology,
as measured by the Hamilton Rating Scale for Depression (HRSD), did
significantly improve over time, the responses to the AMT's cue words did not
become significantly more specific. More recently, Mackinger, Pachinger,
Leibetseder and Fartacek (2000) compared two groups of women who were not
depressed at the time of study, one of which had a history of depression. This group
of women in remission from major depression was significantly more overgeneral
than the group without a lifetime prevalence of depression. The never-depressed
group retrieved about twice as many specific autobiographical memories and half
the number of categoric descriptions than the group in remission.
Together with the finding that overgeneral memory is related to the diagnosis of
depression but not with depression severity (Wessel et al., 2001) these studies
indicate that, in contrast to many other cognitive characteristics of depression, lack
of autobiographical memory specificity is quite stable and does not disappear
during remission. In this respect it seems to be a good candidate for being con-
sidered as a vulnerability factor for depression. But, before one can firmly conclude
that it does represent a vulnerability factor for the onset of depression, more
longitudinal research is needed that investigates whether less specific persons do
indeed have an increased probability of developing a first onset of depressive
disorder. A study by Mackinger, Loschin and Leibetseder (2000) represents a first
step in this direction. They followed a group of 50 pregnant women and assessed
postpartum affective changes, using the Edinburgh Postnatal Depression Scale with
one test during pregnancy and a second test 3 months after delivery. The AMT was
also assessed during pregnancy. Regression analysis showed that the number of
categoric responses to negative cues allowed a significant prediction of affective
changes. The more overgeneral, the greater the extent of affect change in the
negative direction. A weakness of this study was, however, that it did not control
for the presence of possible depressive episodes before pregnancy. Hence, there

55
remains a need for methodologically strong prospective studies to investigate
whether overgeneral memory is more than just a stable "consequence" of depres-
sion, and represents a true vulnerability factor for the development of depressive
disorder.

Lack ofMemory Specificity and Poor Long-Term Outcome

Irrespective of whether overgeneral memory represents a vulnerability factor for

the onset of depression, one can also investigate to what extent overgeneral memory
contributes to the maintenance of a current depressive episode or to relapse after
remission. A study by Brittlebank et al. (1993) has revealed quite impressive results
relevant to these questions. In this study, a group of patients was tested at admission
and 3 and 7 months later. During the first test session, besides the AMT, patients
had to complete two traditional measures of depression severity (the HRSD and the
BOI), as well as a measure of dysfunctional attitudes. This testing was repeated at
the two follow-up moments. Overgeneral memory at baseline (test session I)
significantly predicted outcome at 3 and 7 months. Multiple regression analysis
revealed that overgeneral responses to positively toned words significantly pre-
dicted outcome and accounted for 33% of the variance of the final HRSD score.
Hence, more overgeneral responses were strongly associated with a poorer
long-term outcome. It was also found that only 11% of those who could be
considered as "overgeneral to positive cues" were recovered at seven months,
whereas 80% in the "specific to positive cues" group had recovered. These results
are even more remarkable when compared with those of the traditional depression
measures (HRSD and BOI) for which no significant predictive power was
observed.
Although some researchers failed to find a similar predictive power of the AMT
(e.g., Brewin, Reynolds, & Tata, 1999; see also Scott, Williams, Brittlebank, &
Ferrier, 1995), a number of more recent studies did replicate the findings of
Brittlebank et al. (1993). For instance, Dalgleish, Spinks, Yiend and Kuyken (200 I)
examined autobiographical memory in patients with seasonal affective disorder
(SAD). These patients completed the AMT during winter, when depressed.
Symptom levels were reassessed during the summer, when participants were in
remission. The number of overly general memories to positive cues generated when
depressed predicted symptom levels when in remission, over and above initial
symptom levels during winter. A second study demonstrated that, although the
number of overgeneral responses predicted outcome, the SAD patients were not
more overgeneral than a control group without a history of depression. This
indicates that even when levels of general memories are no greater in a given target
group than in controls, the absolute level of general memories (to positive cue
words) is still independently related to symptom outcome. Another study that
demonstrated the predictive power of the AMT with respect to long term outcome

56
in depression stems from Peeters, Wessel, Merckelbach and Boon-Vermeeren
(2002). Finally, Harvey, Bryant and Dang (1998) made similar observations in the
context of acute stress disorder. On the basis of these studies one can conclude that
overgeneral memory might play an important role in the course of depression, with
lack of specificity being related to less positive outcomes. At the moment there are
several hypotheses about how this relationship could be conceptualized. We will
briefly discuss them in the following paragraph.

Memory Specificity, Social Problem Solving, lmageability of the Future

and Emotional Processing

A number of studies have demonstrated that overgeneral memory is associated

with poor social problem solving. Evans, Williams, O'Loughlin and Howells
(1992) examined the relation between AMT performance and interpersonal
problem solving, using the Means-Ends Problem Solving test (MEPS; Platt &
Spivack, 1975). This test gives a problem scenario together with a positive ending
to the story (e.g., moving to a new neighborhood and having few friends; the story
ends with the person having successfully made some friends). Participants have to
propose solutions to complete the middle of the story. The participants in the study
of Evans and coIleagues were 12 overdose patients within their first 36 hours of
admission. Although it was to be expected that these depressed and suicidal persons
would be less effective in this type of social problem solving than control persons,
these authors additionaIly observed that the effectiveness of the solutions proposed
by the overdose group was significantly related to the specificity of autobiographi-
cal memory. The less specific, the less effective were the solutions (see also PoIlock
& Williams, 200 I; Sidley, Whitaker, Calam, & WeIls, 1997). Similar results were
observed in groups of depressed students and clinically depressed patients
(Goddard, Dritschel, & Burton, 1996, 1997) and persons suffering from bipolar
disorder (Scott, Stanton, Garland, & Ferrier, 2000). It is hypothesized that adequate
problem solving is in part based on the capacity to retrieve specific memories.
Specific memories about past situations might function both at the "problem
definition" and the "generation of alternatives" stages of problem solving by
offering cues from which to build elaborate social problem-solving strategies. The
categoric retrieval style imposes a changeless perspective on one's own past
problem-solving experiences. This inaccessibility of specific autobiographical
memories of previous situations in which one was confronted with similar problems
hinders effective problem solving. In turn, this impaired social problem solving
might negatively affect depression, as it will give rise to negatively toned social
situations and relationships and to a diminished amount of social reinforcement.
A second perspective pertains to the relationship between the specificity of
autobiographical memory and imageability of the future. Several studies of our own
(de Decker, 2001) as well as other labs (e.g., WiIliams et al., 1996) have

57
demonstrated that there is a positive relationship between the specificity with which
people retrieve episodes from their past and the specificity with which they imagine
the future. For persons who are characterized by a more overgeneral memory,
images of the future too are less specific. Evidence is not only correlational in
nature: experimental work by Williams and colleagues (1996) showed that inducing
a generic or specific retrieval style affected the specificity of images of the future
accordingly. This might entail important clinical implications, given that reduced
specificity in imagining (especially positive aspects of) the future might fuel
feelings of hopelessness.
A third way in which overgeneral autobiographic memory might contribute to
the maintenance of depression or vulnerability for relapse, relates to a relative
absence of exposure to negatively toned memories or feelings. Repeated
confrontation with past negative or traumatic events can substantially reduce the
negative emotions associated with these events (Littrell, 1998). Because of low
specificity of the related autobiographical memories, this naturally occurring
exposure effect might be impeded in persons that are characterized by an over-
general memory style. In the short run, this lack of exposure might have a positive
effect (due to the absence of confrontation with negative memory traces), but in the
long run this will have a strongly unfavourable effect on psychological well-being
(Hermans & de Decker, 2001).
We can thus conclude that overgeneral autobiographical memory can exert a
maintaining influence on depressed mood through impaired social problem solving,
reduced imageability of the future and limited exposure to negative memories and
feelings.

Memory Specificity and Affect Regulation

Besides depression, a series of studies demonstrates that overgeneral memory

can also be observed in persons who have experienced traumatic sexual or physical
abuse (e.g., Henderson et aI., 2002; for a recent overview, see Hermans et aI., 2004).
Kuyken and Brewin (1995), for example, investigated the relationship between
early adversities and autobiographical memory retrieval style. They found that in a
clinical sample of depressed women, those who reported childhood sexual abuse
(CSA) retrieved significantly more overgeneral memories on the AMT than those
who did not report CSA. These results have recently been extended by Hermans et
al. (2004) to the presence of physical abuse. Also, in the group of depressed
inpatients investigated by Hermans and colleagues, 10 out of the 18 patients
reported being sexually approached before the age of 18. Within this group, the
number of specific responses was significantly correlated with the age at which this
sexual approach started. The younger the person was when the sexual approaches
took place and the more distressing these events, the less specific were the answers
on the AMT.

58
 There are several ways in which this relation between trauma and lack of
memory specificity can be explained. Besides explanations in terms of the psycho-
biological or neuroanatomical impact of trauma on brain functioning, two
psychological accounts are viable. First, in cases of traumatic experiences that took
place early in childhood (before the age of 5), children might have missed a normal
developmental stage during which they learn to inhibit relatively automatic cate-
goric descriptions, allowing the child greater strategic control over the recollection
process (including accessing specific memory traces). Together with the biological
explanation, this account can be considered as a "deficit model". A second, and
more functional, account states that being overgeneral might diminish the
emotional impact of traumatic or negative events from the past. Persons who
experienced traumatic adversities might adopt a generic style in retrieving auto-
biographical memories as an affect regulating strategy. Being less specific then
reduces the risk of confrontation with painful memories concerning the traumatic
experiences. This strategy only works if it is not only applied with respect to the
traumatic memories, but to memory in general. This hypothesis, referred to here as
the affect-regulation hypothesis, has received experimental support from recent
research by Raes et al. (2003). They demonstrated that being less specific in the
retrieval of autobiographical memories has a positive effect on the affective impact
of a negative event. This could be seen as an extension of Williams' original affect-
regulation hypothesis (1996): being less specific not only regulates affect
associated with past negative events, but also makes one less vulnerable to future
adversities. It appears that being less specific in some cases is advantageous - at
least in the short run (!) - and might be termed as functional, protective or adaptive.
In the long run, however, this reduced specificity may very well prove to be
maladaptive and to have detrimental effects for reasons described above. This more
functional approach to the study of autobiographical memory seems to be an
appealing route to the understanding of the mechanisms of overgeneral memory in
depression.

Autobiographical Memory Specificity: Some Concluding Remarks

The study of autobiographical memory specificity is still in its infancy, but has
already proven to be quite fruitful. The fact that depression and the presence of
traumatic experiences have a unique linkage with reduced memory specificity has
been the impetus for some developmental hypotheses. One possibility is that
trauma leads to overgeneral memory as a way of coping with adverse memories.
When this strategy loses its flexibility and gets more and more engrained it puts the
person at risk for depression for reasons mentioned earlier. Although attractive, this
hypothesis is in need of good prospective studies to be put to the test. Irrespective
of this developmental issue, reduced memory specificity seems to be a quite stable
characteristic of many depressed persons. It is predictive of poor long-term

59
outcome, and it is associated with poor social problem solving and limited
imageability of the future. In as much as it represents an affect-regulation strategy
it might have positive short-term consequences, but will be detrimental in the long
run because it deprives the person of a lot of naturally occurring self-exposure to
negative memories and feelings. Although it is most probable that overgeneral
memory is a characteristic that prolongs depressed mood and renders the depressed
person more vulnerable for relapse, future research will have to unveil its precise
role in the onset of depression.

Discussion and Clinical Implications

The present chapter reviewed cogmtrve psychology research on mood and

(explicit) memory. In particular we addressed aspects relevant to the maintenance
of depression and vulnerability for depressive relapse. First, the research tradition
on the interaction between mood an memory was covered. Next, findings of the
related but more recent line of research on autobiographical memory specificity
were discussed.
It is interesting to see how paradigms from experimental cognitive psychology,
discussed in the present chapter, are being used to further support psychological
theories on depression that originated outside experimental cognitive psychology
and of which some are reviewed elsewhere in this volume. For example, in Chapter
3 and 4 in this volume, recent theorizing from Beck (Beck, 1987; Beck, Epstein, &
Harrison, 1983) on the role of sociotropy and autonomy in precipitating depression
is discussed, which relates to Blatt's (1974) ideas on the distinction between
anaclitic and introjective depression. Moore and Blackburn (1993) looked for
evidence for this sociotropy/autonomy distinction in depressed patients, using a
paradigm from experimental cognitive psychology research: the autobiographical
memory paradigm. They found that sociotropy, as measured by the Sociotropy
Autonomy Scale (SAS; Beck, Epstein, Harrison, & Emery, 1983), was associated
with faster retrieval of negative sociotropic autobiographical memories. A similar
relationship between autonomy and increased (biased) accessibility of negative
autonomous autobiographical events, however, was not found (the authors note that
this might be due to validity problems of the Autonomy subscale of the SAS, see
also Chapter 4, this volume). These results then partly support the sociotropy/
autonomy distinction and related selective information processing.
Not only in the memory domain, but also in the field of attentional processing,
experimental paradigms are being applied to investigate evidence in support of the
sociotropy/autonomy idea. Nunn, Mathews and Trower (1997) applied an
experimental Stroop task to investigate whether sociotropy and autonomy would be
related to selective attentional processing of "sociotropic" and "autonomous"
words respectively. Results indicated that there was a bias for all types of negative
words and no specific relation between the two domains of concern (sociotropy/

60
auto-nomy) and biased processing of corresponding word types. These authors too,
suggest that psychometric problems of the Sociotropy-Autonomy Scale may partly
explain why no such relation was found. It should be clear that these results are not
to be regarded as real counter-evidence for the sociotropy/autonomy distinction and
related biased cognitive processing. Moreover, the fact that there seems to be some
evidence for a biased processing, related to the sociotropy/autonomy distinction, in
the memory study (Moore & Blackburn, 1993), and not in the attention study (Nunn
et al., 199"1). actually should come as no surprise: cognitive biases in memory
processes are more commonly found in depression as compared to biased
attentional processing, which seems to be more related to anxiety disorders
(Williams et al., 1997).
Aside from integrating - or at least relating - different psychological
perspectives or approaches to depression, another issue is the coupling of
psychological theories with biological, neuroanatomical and neurophysiological
approaches. In that respect, brain imaging techniques opened the way to the study
of the neuroanatomical and neurophysiological bases of memory in general, and of
autobiographical memory retrieval processes in particular (see Conway, 200 I;
Rugg & Wilding, 2000). A viable route, then, for future studies in the above-
mentioned research on autobiographical memory specificity will be the application
of brain imaging expertise to study the neurophysiological correlates of the
overgenerality phenomenon in the retrieval of autobiographical memories in
depressed patients.
Let us now tum to the possible clinical implications of the research findings
discussed in the present chapter. Irrespective of the input these results provide to the
development of new therapeutic techniques and procedures or the improvement of
old ones, we believe that the therapeutic relevance of research on depression and
memory also lies in its psycho-educational potential. Informing patients about this
research, and in particular about the mechanisms at work in depression (e.g., mood-
congruent recall, overgeneral memory, etc.) might in itself already be a valuable
part (or first step) of treatment. This "first step" also nicely matches patients' first,
and at the outset of treatment often most predominant question: "How did it come
this far?", rather than "How can I solve my problem?". That way, patients gradually
gain insight into depression, in particular into the processes and mechanisms
relevant to maintenance of depression and vulnerability for depressive relapse. And
in so doing, a preventive way of thinking is introduced and encouraged which will
considerably facilitate the task of motivating patients to start working on (relapse)
prevention. For example, when depressed patients, at the beginning of a session, are
asked to report on the past week, they will often talk about all that went wrong and
will articulate a stream of negative thoughts, feelings, experiences, etc. At such a
point, a therapist might introduce and explain or - if already discussed in a previous
session - again point to the mood congruency mechanisms, and check how they are
"at work" here in patients' reports of the preceding week. It is our feeling and
experience that when patients are able to look at their symptoms with some
background of knowledge concerning the psychological mechanisms involved in
depression, this will enable them to take on a meta-perspective (see below) on their
depression. Whereas otherwise negative thoughts and feelings are often viewed as

61
objective representations of reality, this meta-perspective will allow them to take
distance. For example, negative thoughts about the world and the others are no
longer seen as evidence for the fact that the world indeed "is" a bad place, but can
now be viewed as products of biased information processing. Or in the words of a
patient: "It is my depression that is talking now". Hence, when a patient is in a
negative mood and catches himself dwelling on negative thoughts about past
failures and losses, he is able to correct himself by realizing that it is his depression
that is "misleading" him, or "playing tricks" on him. The more this meta-
perspective can be generalized to situations outside the therapeutic sessions, the
more a downward spiral might be stopped in an early stage, and a further
amplification of depressed mood is averted. Future research will need to establish
the therapeutic significance of such psycho-education.
Aside from this psycho-educational aspect, research findings on mood and
memory have also led researchers to develop or refine therapeutic methods to
counteract the various memory processes involved in the maintenance of depression
and depressive relapse. For example, given the evidence reviewed above for the
negative impact of autobiographical memory overgenerality on functioning (e.g.,
poor social problem solving) and on the course of depression, researchers and
therapists have pointed to the importance for patients "to go beyond general
statements such as 'I've always been a failure' or 'I used to be so happy' to describe
the details of particular instances when they had failed or felt fulfilled" (Williams
& Dritschel, 1988, p. 232). In this respect, researchers often refer to techniques
such as diary keeping and self-monitoring (e.g., Williams & Scott, 1988). A
consequence of a tendency to overgeneralize or to not attend to specific details of
experiences is that positive events or positive elements of events go unnoticed. For
example, Fava (1999), in his "well-being therapy", asks patients to monitor and to
write down in a structured diary the detailed circumstances of all events leading to
feelings of well-being. Also, positive instances that could counter-argue a more
overgeneral negative memory (e.g., "I've always disappointed other people") might
go unnoticed, due to this mechanism of overgeneral memory. For example, in
response to the cue word "cowardly", a depressed patient in our autobiographical
memory research retrieved the following overgeneral "categoric" memory: "Since
my depression, I've had no longer the courage to attend any of the weekly meetings
at work on Fridays". When he was further prompted to recall a specific instance of
that overgeneral memory, he responded: "Oh now wait a minute, two months ago,
I did attend the Friday meeting". Further "detailed" elaboration of that memory
revealed that he had rather enjoyed that particular meeting (he received a
compliment from his boss, and from one of his closest colleagues). This specific
memory provided a nice counterexample for the overall "negative" overgeneral
memory of not attending the weekly meetings.
Williams, Teasdale, Segal and Soulsby (2000) further showed how a more
elaborate treatment package, called "mindfulness-based cognitive therapy"
(MBCT; Segal, Williams, & Teasdale, 2002; Teasdale, Segal, & Williams, 1995),
significantly reduces overgenerality in formerly depressed patients. MBCT is
about making patients aware of specific aspects of their experiences and about
promoting the adoption of a different perspective on one's own thoughts and

62
worries, namely a meta-perspective in which one "observes" one's own ruminative
thoughts without judging or trying to suppress them. This reflects an attitude of
"decentering" or distraction where one no longer actively engages in further
dwelling on negative thoughts. It may come as no surprise then, that research has
shown ruminative thinking to be involved in the maintenance of overgenerality
(e.g., Watkins & Teasdale, 2001). Rumination is a clear manifestation of depressed
patients' difficulty with disengaging from all negative material (thoughts,
memories, events). As such, this "decentering" philosophy might not only be of use
in tackling overgeneral memory, but might also be of importance in countering or
breaking through vicious circles such as mood-congruent recall and mood-
congruent encoding in the prevention of depression and depressive relapse, as
already hinted at above.
Thus, it seems clear that as a response to the changing view of depression as a
recurrent and often chronic disorder (see above), treatment of depression is
broadening its perspective as well. There is now broad agreement that the focus in
treatment of depression should extend beyond treatment of a current depressive
episode. More important, treatment should focus also, and maybe even more so, on
the prevention of future relapse or recurrence. And this is what MBCT is all about:
it is an approach to preventing relapse in depression (Segal et al., 2002). In essence,
MBCT could be regarded as a short-term, eight-session, treatment program. But in
the end, it has a long-term vision, as it is about providing patients with life-long
skills for adequately dealing with negative emotion in the future (Segal et aI., 2002).
Thus, whether we need short-term or long-term treatment in depression seems to be
not the most important issue. What is important is that the focus in treatment lies
on relapse prevention, and that it extends beyond treating a single depressive
episode.

References

Bayingana, K., Drieskens, S.• & Tafforeau, J. (2002). Depressie. Stand van zaken in Belgie. Richtlijnen
voor een gezondheidsbeleid [Depression. State of the art. Guidelines for mental health policy].
Wetenschappelijk Instituut Volksgezondheid IPHIEPI Reports nr. 2002 -DI I.
Beck, A. T. (1987). Cognitive model of depression. Journal of Cognitive Psychotherapy. l, 2-27.
Beck. A. T., Epstein, N., & Harrison. R. (1983). Cognitions, attitudes and personality dimensions in
depression. British Journal of Cognitive Psychotherapy. l, 1-16.
Beck, A. T., Epstein. N., Harrison. R.• & Emery, G. (1983). Development of the Sociotropy-Autonomy
Scale: A measure of personality factors ill psychopathology. Unpublished manuscript, University of
Pennsylvania, Philadelphia.
Beck, A. T., Rush, A., Shaw, B., & Emery, G. (1979). Cognitive therapy of depression. New York: Guilford
Press.
Belsher, G., & Costello, C. G. (1988). Relapse after recovery from unipolar depression: A critical review.
Psychological Bulletin. /04. 84-96.
Bijl, R.v., van Zessen, G., & Ravelli, A. (1997). Psychiatrische morbiditeit onder volwassenen in Neder-
land: het NEMESIS-onderzoek II. Prevalentie van psychiatrische stoornissen [Psychiatric morbidity in
adults in The Netherlands: The NEMESIS study II: Prevalence of psychiatric disorders]. Nederlands
Tijdschrift voor Geneeskunde, /4/. 2453-2460.

63
Blatt, S. 1. (1974). Level of object representation in anaclitic and introjective depression. Psychoanalytic
Study of the Child. 29. 107-157.
Blaney, P. H. (1986). Affect and memory: A review. Psychological Bulletin. 99.229-246.
Blazer, D. G., Kessler, R. C.. McGonagle, K. A., & Swartz, M. S. (1994). The prevalence and distribution
of major depression in a national community sample: The National Comorbidity Survey. American
Journal of Psychiatry. 151. 979-86.
Bower, G. H. (1981). Mood and memory. American Psychologist. 36. 129-148.
Bower, G. H. (1987). Commentary on mood and memory. Behaviour Research and Therapy, 25, 443-455.
Bower, G. H. (1991). Mood congruity of social judgements. In 1. P. Forgas (Ed.), Emotion and social
judgments (pp. 31-54). Oxford: Pergamon Press.
Bower, G. H., Gilligan, S. G., & Monteiro, K. P. (1981). Selective learning caused by affective stales.
Journal of Experimental Psychology: General. 110. 451-473.
Bower, G. H., Monteiro, K. P., & Gilligan, S. G. (1978). Emotional mood as a context of learning and
recall. Journal of Verbal Learning and Verbal Behavior, 17.573-585.
Brewin, C. R., Reynolds, M.. & Tata, P. (1999). Autobiographical memory processes and the course of
depression. Journal ofAbnormal Psychology. 108. 51 1-517.
Brewin, C. R., Watson, M., McCarthy, S.. Hyman, P.. & Dayson, D. (1998). Intrusive memories and
depression in cancer patients. Behaviour Research and Therapy. 36. 1131-1142.
Brittlebank, A. D., Scott, J., Williams, J. M. G., & Ferrier, I. N. (1993). Autobiographical memory ill
depression: State or trait marker? British Journal of Psychiatry. 162. 118-121.
Burke, M., & Mathews, A. (1992). Autobiographical memory and clinical anxiety. Cognition and Emotion,
6. 23-35.
Clark, D. A" & Beck, A. T. (1999). Scientific foundations of cognitive theory and therapy (if depression.
New York: John Wiley & Sons.
Clark, D. M., & Teasdale, J. D. (1982). Diurnal variation in clinical depression and accessibility of
memories of positive and negative experiences. Journal ofAbnormal Psychology. 91, 87-95.
Conway, M. A. (2001). Sensory-perceptual episodic memory and its context: Autobiographical memory,
Philosophical Transactions of the Royal Society of London. 356, 1375-1384.
Dalgleish, T., Spinks, H., Yiend, J.. & Kuyken, W. (2001) Autobiographical memory style in seasonal
affective disorder and its relationship to future symptom remission. Journal of Abnormal Psychology.
110. 335-340.
Dalgleish, T., Tchanturia, K., Serpell, L., Hems, S.. Yiend, J.. de Silva, P., & Treasure, J. (2003). Self-
reported parental abuse relates to autobiographical memory style in patients with eating disorders.
Emotion. 3. 211-222.
de Decker, A. (200 I). The specificity of the autobiographical memory retrieval style in adolescents witn a
history of trauma. Unpublished doctoral dissertation, University of Leuven, Belgium.
de Decker, A., Hermans, D., Raes, F.. & Eelen P. (2003), Autobiographical memory specificity and trauma
in inpatient adolescents. Journal of Clinical Child and Adolescent Psychology. 32, 22-31.
Evans, J., Williams, J. M. G., O'Loughlin, S.. & Howells, K. (1992). Autobiographical memory and
problem solving strategies of individuals who parasuicide. Psychological Medicine, 22. 399-405.
Fava, G. A. (1999), Well-being therapy: Conceptual and technical issues. Psychotherapy and
Psychosomatics. 68. 171-179.
Fiedler, K, Nickel, S., Muehlfriedel, T.. & Unkelbach, C. (200 I). Is mood congruency an effect of genuine
memory or response bias. Journal (if Experimental Social Psychology. 37. 201-214.
Forgas, J. P., & Bower, G. H. (1987). Mood effects on person-perception judgments. Journal of Personality
and Social Psychology. 53. 53-60.
Gilligan, S. G., & Bower, G. H. (1984). Cognitive consequences of emotional arousal. In C. E. Izard, J.
Kagan & R. Zajonc (Eds.), Emotions. cognitions; and behaviour (pp. 547-588). New York: Cambridge
Press.
Goddard, L., Dritschel, B., & Burton, A. (1996). Role of autobiographical memory in social problem
solving and depression. Journal ofAbnormal Psychologv. 105.609-616.
Goddard, L., Dritschel, B., & Burton, A. (1997). Social problem-solving and autobiographical memory in
non-clinical depression. British Journal (If Clinical Psychology. 36. 449-451.
Godden, D. R., & Baddeley, A. D. (1975). Context dependency in two natural environments: On land and
underwater. British Journal of Psychology. 91. 99-104.
Gotlib, I. H., & Krasnoperova, E. (1998). Biased information processing as a vulnerability factor for
depression. Behavior Therapy. 29. 603-617.

64
Hammen, C. (2001). Vulnerability to depression in adulthood. In R. E. Ingram & J. M. Price (Eds.),
Vulnerability to psychopathology. Risk across the life-span (pp. 226-257). New York: The Guilford
Press.
Harvey, A. G., Bryant, R. A., & Dang, S. T. (1998). Autobiographical memory in acute stress disorder.
Journal of Consulting and Clinical Psychology, 66, 500-506.
Henderson, D., Hargreaves, I.. Gregory, S., & Williams, 1. M. G. (2002). Autobiographical memory and
emotion in a non-clinical sample of women with and without a reported history of childhood sexual
abuse. British Journal of Clinical Psychology, 41, 129-141.
Hermans, D., & de Decker, A. (200 I, December). Adaptive aspects of overgeneral memory: Part 2. Paper
presented at the Second Autobiographical Memory Workshop, Cambridge, England.
Hermans, D., Van den Broeck, K., Belis, G., Raes, E, Pieters, G., & Eelen, P. (2004). Trauma and
autobiographical memory specificity in depressed inpatients. Behaviour Research and Therapy, 42,
775-789.
Judd, L. L., Akiskal, H. S., & Paulus, M. P. (1997). The role and clinical signiticance of subsyndromal
depressive symptoms (SSD) in unipolar major depressive disorder. Journal ofAffective Disorders, 45,
5-17.
Kuyken, w., & Brewin, C. R. (1995). Autobiographical memory functioning in depression and reports of
early abuse. Journal ofAbnormal Psychology, 104,585-591.
Kuyken, w., & Dalgleish, T. (1995). Autobiographical memory and depression. British Journal of Clinical
Psychology, 34, 89-92.
Littrell, J. (1998). Is the reexperience of painful emotion therapeutic? Clinical Psychology Review, 18,
71-102.
Lloyd, G. G., & Lishman, W. A. (1975). Effect of depression on the speed of recall of pleasant and
unpleasant memories. Psychological Medicine, 5, 173-180.
Mackinger, H. E, Loschin, G. G., & Leibetseder, M. M. (2000). Prediction of postnatal affective changes
by autobiographical memories. European Psychologist, 5, 52-61.
Mackinger, H. E, Pachinger, M. M., Leibetseder, M. M., & Fartacek, R. R. (2000). Autobiographical
memories in women remitted from major depression. Journal ofAbnormal Psychology, 109, 331-334.
Matt, G. E., Vazquez, C; & Campbell, K. W. (1992). Mood congruent recall of affectively toned stimuli:
A meta-analytic review. Clinical Psychology Review, 12,227-255.
Moore, R. G., & Blackburn, I.-M. (1993). Sociotropy, autonomy and personal memories in depression.
British Journal of Clinical Psychology. 32, 460-462.
Moore, R. G., Watts, EN., & Williams, J. M. G. (1988). The specificity of personal memories in
depression. British Journal of Clinical Psychology, 27, 275-276.
Nunn, J. D., Mathews, A., & Trower, P. (1997). Selective processing of concern-related information in
depression. British Journal (If Clinical Psychology, 36, 489-503.
Park, R. 1., Goodyer, I. M., & Teasdale, J. D. (2002). Categoric overgeneral autobiographical memory in
adolescent Major Depression. Psychological Medicine, 32, 267-276.
Peeters, E, Wessel, I., Merckelbach, H., & Boon-Vermeeren, M. (2002). Autobiographical memory
specificity and the course of major depressive disorder. Comprehensive Psychiatry, 43, 344-350.
Platt, 1. J., & Spivack, G. (1975). Manual for the Means-Ends Problem-Solving (MEPS): A measure of
interpersonal problem-solving skill. Philadelphia: Hahnemann Medical College and Hospital.
Pollock, L. R., & Williams, J. M. G. (2001). Effective problem solving in suicide attempters depends on
specific autobiographical recall. Suicide and Threatening Behavior, 31, 3l\6-396.
Puffer, A., Jehin-Marchot, D., Timsit-Berthier, M., & Timsit, M. (1991). Autobiographical memory and
major depressive states. European Psychiatry, 6, 141-145.
Raes, E, Hermans, D., de Decker, A., Eelen, P., & Williams, J. M. G. (2003). Autobiographical memory
specificity and affect-regulation: An experimental approach. Emotion, 3, 201-206.
Rugg, M. D., & Wilding, E. L. (2000). Retrieval processing and episodic memory. Trends in Cognitive
Sciences, 4, 108-115.
Rusting, C. L., & DeHart, T. (2000). Retrieving positive memories to regulate negative mood:
Consequences for mood-congruent memory. Journal of Personality and Social Psychology, 78, 737-
752.
Schwartz, B., & Reisberg, D. (1991). Learning and memory. New York: Norton.
Scott, J., Williams, J. M. G., Brittlebank, A., & Ferrier, I. N. (1995), The relationship between premorbid
neuroticism, cognitive dysfunction and persistence of depression. Journal of Affective Disorders, 33,
67-72.

65
Scott, J., Stanton, B., Garland, A., & Ferrier, I. N. (2000). Cognitive vulnerability in patients with bipolar
disorder. Psychological Medicine. 30. 467-472.
Segal, Z. v., Gemar, M. G., & Williams, S. (1999). Differential cognitive response to a mood challenge
following successful cognitive therapy or pharmacotherapy for unipolar depression. Journal of
Abnormal Psychology. /08.3-10.
Segal, Z. V., Williams, J. M. G., & Teasdale. J. D. (2002). Mindfulness-based Cognitive Therapy for
depression. A new approach to preventing relapse. New York: Guilford Press.
Share, M. L., Lisman, S. A., & Spear, N. E. (1984). The effects of mood variation on state-dependent
retention. Cognitive Therapy and Research. 8. 387-408.
Sidley, G. L., Whitaker, K., Cal am, R. M., & Wells, A. (1997). The relationship between problem solving
and autobiographical memory in parasuicide patients. Behavioural and Cognitive Psychotherapy. 25.
195-202.
Singer, J. A., & Salovey, P. (1988). Mood and memory: Evaluating the network theory of affect. Clinical
Psychology Review. 8. 211-251.
Teasdale, J. D., & Russell, M. L. (1983). Differential effects of induced mood on the recall of positive,
negative and neutral words. British Journal of Clinical Psychology. 22. 163-171.
Teasdale, J. D., Segal, Z. V.. & Williams, J. M. G. (1995). How does cognitive therapy prevent depressive
relapse and why should attentional control (mindfulness) training help? An information processing
analysis. Behaviour Research and Therapy. 33. 25-39.
Wachtel, P. L. (1994). Cyclical processes in personality and psychopathology. Journal of Abnormal
Psychology, /03,51-66.
Wachtel, P. L. (1997). Psychoanalysis. behavior therapv, and the relational world. Washington, DC:
American Psychological Association.
Watkins, E., & Teasdale, J. D. (2001). Rumination and overgeneral memory in depression: Effects of self-
focus and analytic thinking. Journal ofAbnormal Psychology. //0, 353-357.
Watkins, P. C. (2002). Implicit memory bias in depression. Cognition and Emotion. /6.381-402.
Weingartner, M., Miller, H., & Murphy, D. L. (1977). Mood-state-dependent retrieval of verbal
associations. Journal ofAbnormal Psychology. 86. 276-284.
Weiss, E. L., Longhurst, J. G., & Mazure, C. M. (1999). Childhood sexual abuse as a risk for depression
in women: Psychological and neurobiological correlates. American Journal of Psychiatry. /56. 816-
828.
Wessel, I., Meeren, M., Peeters, E, Arntz, A., & Merckelbach, H. (200 I). Correlates of autobiographical
memory specificity: The role of depression, anxiety and childhood trauma. Behaviour Research and
Therapy. 39. 409-421.
Wilhelm, S., McNally, R. J., Baer, L., & Florin, I. (1997). Autobiographical memory in obsessive-
compulsive disorder. British Journal of Clinical Psychology. 36. 21-31.
Williams, J. M. G. (1992). The psychological treatment of depression: A guide to the theory and practice
of cognitive behaviour therapy. London: Routledge.
Williams, J. M. G. (1996). Depression and the specificity of autobiographical memory. In D. C. Rubin
(Ed.), Remembering our past. Studies in autobiographical memory (pp, 244-267). Cambridge, NY:
Cambridge University Press.
Williams, J. M. G., & Broadbent, K. (1986). Autobiographical memory in suicide attempters. Journal of
Abnormal Psychology. 95, 144-149.
Williams, J. M. G., & Dritschel, B. H. (1988). Emotional disturbance and the specificity of
autobiographical memory. Cognition and Emotion. 2. 221-234.
Williams, J. M. G., Ellis, N. c., Tyers, c., Healy, H., Rose, G., & MacLeod, A. K. (1996). Specificity of
autobiographical memory and imageability of the future. Memory and Cognition. 24. 116-125.
Williams, J. M. G., & Scott, J. (1988). Autobiographical memory in depression. Psychological Medicine.
/8.689-695.
Williams, J. M. G., Teasdale, J. D., Segal, Z. v.. & Soulsby, J. (2000). Mindfulness-based cognitive therapy
reduces overgeneral autobiographical memory in formerly depressed patients. Journal of Abnormal
Psychology, 109. 150-155.
Williams, 1. M. G., Watts, E, MacLeod. c.. & Mathews, A. (1997). Cognitive psychology and emotional
disorders (2nd ed.). Chichester: John Wiley & Sons.

66
 Chapter 3

The Convergence Among Psychodynamic and Cognitive-

Behavioral Theories of Depression: Theoretical Overview

Patrick Luyten, Sidney J. Blatt, & Jozef Corveleyn

Over the last decades, there has been a growing trend towards integration among
psychodynamic and cognitive-behavioral theories (Jones & Pulos, 1993; Milton,
200 I; Ryle, 1995; Wachtel, 1997; Westen, 2000), specifically concerning the
conceptualization and treatment of depression (Clark & Beck, 1999; Blatt, 2004;
Blatt & Maroudas, 1992). In general, two related factors are responsible for this
trend. First, as mentioned in the introduction of this volume, findings on the natural
course of depression have changed our view of this disorder in important ways.
Research has shown that depression tends to be a recurrent and even chronic
disorder (e.g., Costello et al., 2002; see also Chapter I, this volume), which has a
high comorbidity with personality disorders (Klein & Hayden, 2000; Mulder,
2002). Second, so-called "evidence-based" brief treatments of depression, such as
"traditional" cognitive-behavioral treatments and interpersonal therapy, have
limited effects (Luyten, Lowyck, & Corveleyn, 2003; Shea et al., 1992; Westen &
Morrison, 200 I; Westen, Novotny, & Thompson-Brenner, 2004). These findings
have not only led to a renewed interest in theories and techniques that are based on
long-term treatments of depression, such as psychodynamic theories (Blatt, 2004;
Jones & Pulos, 1993; Kwon, 1999), in which considerations concerning personality
and particularly personality organization often playa central role (Westen, 2000),
but also in the relationship between personality and depression in general (Klein,
Durbin, Shankman, & Santiago, 2002; Klein, Kupfer, & Shea, 1993; Gunderson et
al., 1999).
In this chapter, we focus on a very specific move towards the integration between
psychodynamic and cognitive-behavioral formulations concerning the conceptuali-
zation and treatment of depression (Blatt, 2004; Robins, 1993; Sabbe, 2002). More
specifically, we will focus on four groups of authors that have proposed similar
personality dimensions as vulnerability factors for both clinical and non-clinical
forms of depression and dysphoria. Blatt (1974, 1998, 2004; Blatt & Zuroff, 1992),
from an object-relational and cognitive-developmental perspective, has described
dependency and self-critical perfectionism 1 as vulnerability factors for depression,
resulting in a dependent (or anaclitic) versus a self-critical/perfectionistic (or
introjective) type of depression, as well as possible mixed forms of dependent and
self-critical depression. Anaclitic depression is characterized by feelings of
helplessness, loneliness, weakness, and intense and often chronic fears of being
abandoned. Introjective depression, in contrast, is characterized by self-criticism
and feelings of guilt, inferiority, failure, and a chronic fear of being disapproved and
criticized.
Beck (1983, 1991, 1999; Clark & Beck, 1999) has made a similar distinction
from a cognitive-behavioral perspective between sociotropy and autonomy as
cognitive-affective personality styles that confer a vulnerability for depression,
which at a descriptive level shows many similarities with Blatt's concepts of
dependency and self-critical perfectionism respectively. Bowlby (1980, 1988) has
distinguished from a psychodynamic/ethological perspective between anxious and
compulsive self-reliant attachment styles predisposing to depression. Arieti and
Bemporad (Arieti & Bemporad, 1978; Bemporad, 1992), finally, distinguish from
an interpersonal psychodynamic perspective between dominant other and dominant
goal personality structures predisposing to depression.
However, most research has concentrated on Blatt's and Beck's formulations of
dependency/sociotropy and self-critical perfectionism/autonomy (e.g., see Blatt,
1998, 2004; Clark & Beck, 1999). Despite almost three decades of empirical
research, this area of research is often little known, even among those who have
specialized in mood disorders research or treatment (Luyten et al., 2003). In this
chapter, we therefore present an overview of the main theoretical assumptions of
these models. Because both Blatt's and Beck's views of depression are part of a
more general theory of normal and pathological personality development, we first
outline these general views briefly. Next, we focus specifically on Blatt's and
Beck's conceptualizations concerning depression, including their views of the
treatment of depression. Relevant empirical research is reviewed in the next
chapter. Despite the fact that these theories offer important avenues towards
integration concerning the conceptualization and treatment of depression, we also
believe that there are some important barriers that limit further integration among
1 Blatt (e.g., Blatt, 1974) originally used the term "self-criticism" to refer to this second personality dimen-

sion. However, in keeping with his more recent views (see Chapter 5, this volume) and given the faci that
self-criticism is only one characteristic of this personality dimension. we prefer the notion "self-cruical
perfectionism" to refer to this personality dimension as a whole.

68
these theories. These barriers are discussed in a separate section. In a final section,
we discuss more general issues concerning these theories and their potential for
further integration in research and treatment of depression.

Blatt's and Beck's Theory of Normal and Pathological

Personality Development

Blatt's Theory of Normal and Psychopathological Personality Development

According to Blatt (e.g., Blass & Blatt, 1996; Blatt, 1995a; Blatt & Blass, 1990;
Blatt & Shichman, 1983; Guisinger & Blatt, 1994), personality development can be
conceptualized as a dialectical interaction between two fundamental developmental
lines, namely (1) an anaclitic or relatedness line that normally leads to increasingly
mature, complex, and mutually satisfying interpersonal relations, and (2) an
introjective or self-definitional line that normally leads to the development of a
stable, realistic, and essentially positive self and identity. In "normal" development,
these two developmental lines are in constant reciprocal or dialectic interaction.
Thus, an increasingly stable and differentiated self-definition leads to more
differentiated and complex interpersonal relationships and vice versa. Hence, in the
case of "optimal" development, high levels of identity and autonomy go together
with the capacity to form and maintain complex, differentiated and satisfying
interpersonal relationships, and vice versa. One is at the same time able to relate to
others in complex, reciprocal ways, without losing one's sense of self and identity.
Such differentiated and reciprocal relationships in turn contribute to the
development of even more integrated and positive feelings of identity and
autonomy.
Based on this theory of normal personality development, psychopathology can
be conceptualized as an overemphasis on or an exaggeration of one of these two
developmental lines and the neglect or defensive avoidance of the other line (e.g.,
Blatt, 1991a, 1991b; Blatt & Maroudas, 1992; Blatt & Shichman, 1983). Thus,
following this view, psychopathology is characterized by a disruption of the
dialectical interaction of these two developmental lines - by a rigid, one-sided
overemphasis of one line (e.g., interpersonal relationships), and the neglect and/or
defensive avoidance of the other line (e.g., feelings of identity and autonomy).
Whereas in "normal" development anaclitic and introjective developmental lines
interact synergistically, and thus are in constant dialectical interaction,
psychopathology is characterized by a rigid overemphasis of one developmental
line at the expense of the other. Thus two clusters or configurations of
psychopathology can be distinguished, depending on which developmental
line predominates and which developmental line is neglected or defended against,

69
i.e., an anaclitic and an introjective cluster. Blatt distinguishes several symp-
tom ("Axis I") and personality ("Axis II") disorders in each developmental line
on three structural levels, i.e., the psychotic, borderline, and neurotic level, which
can be seen as different rigid exaggerations of one developmental line and
the neglect or defensive avoidance of the other (see below; Blatt & Shichman,
1983). To distinguish between adaptive and maladaptive dimensions on each
developmental line, Blatt has introduced the terms relatedness versus dependency
or neediness, and efficacy versus self-criticism or self-critical perfectionism
respectively (Blatt, 2004; Zuroff, Mongrain, & Santor, 2004; see also Chapter 5,
this volume).
Thus, the name of these two developmental lines not only refers to
the central dominant developmental theme in each line, but also to the cen-
tral problem in case of disturbance (Blatt, 1974; Blatt & Shichman, 1983). The
notion "anaclitic" refers to the development of interpersonal relations that
prototypically find their origin in the relation with the first significant other, mostly
the mother. Disturbances in this developmental line can give rise to anaclitic
psychopathology, which is primarily characterized by problems concerning
relatedness, such as intimacy, caring, and sexuality. These concerns can range from
desires and conflicts concerning primitive, dyadic relationships, in which
(primitive) symbiotic desires are central, as in nonparanoid schizophrenia
(psychotic level) and, on somewhat higher levels, in hysteroid borderline
personality disorder, histrionic (borderline level) and dependent personalities (low
neurotic level). At the higher neurotic level, desires and conflicts concern more
triadic, oedipal relationships, as in hysteria. According to Blatt (Blatt & Levy,
1998), the DSM-IV Axis II dependent, histrionic, passive-aggressive and borderline
personality disorders belong to the anaclitic cluster (Blatt & Levy, 1998; Shahar,
Blatt, & Ford, 2003).
The notion "introjective" in turn refers to the development of a stable, dif-
ferentiated and integrated self by means of identification or introjection.
Introjective pathology is according to Blatt primarily characterized by a focus on
issues such as self-control, guilt, identity, and autonomy (Blatt & Shichman, 198.3).
Again, this focus can range from very primitive desires for autonomy, for instance
in striving for isolation and defensive separation, such as in paranoid schizophrenia
(psychotic level) and to a lesser extent in more "ideational" borderlines (borderline
level), and in paranoid personalities, to a rigid control over impulses, as in
obsessive-compulsive neurosis, and to an exaggerated focus on self-worth, power
and identity in introjective depression and phallic narcissism (or narcissistic
personality disorder) (neurotic level). The DSM-IV Axis II schizoid, schizotypic,
paranoid, narcissistic, antisocial, avoidant, self-defeating, and obsessive-
compulsive personality disorders belong in this cluster according to Blatt (Blatt &
Levy, 1998; Shahar et al., 2003).
In general, introjective patients are expected to be more ideational than anaclitic
patients and more concerned with identity and autonomy than with interpersonal
relations and affects (Blatt & Shichman, 1983). Moreover, introjective patients are
assumed to have a more sequential cognitive style that focuses on details,
responsibility, and cause-effect relationships, whereas the cognitive style of

70
anaclitic patients is expected to be more impressionistic and global. Conflicts or
problems with anger and aggression dominate the clinical picture in introjective
patients, while libidinal issues predominate in anaclitic patients.
Finally, Blatt (e.g., Blatt & Maroudas, 1992; Blatt & Shichman, 1983) hypo-
thesizes that there are important gender differences in the emphasis on these two
developmental lines and thus perhaps also in the two clusters of psychopathology.
First, several psychological theories suggest that the primary identification figure
throughout female development is typically the mother, whereas the male child has
to switch in development from the mother to the father figure (Chodorow, 1978;
Baker Miller, 1976). Thus, the male child is typically forced to de-identify with the
mother and thus to de-emphasize relationships to attain a stable identity and
autonomy, whereas there is much more continuity in the emphasis on relationships
in the female child. In addition, at least in our western society there (still) is on
average more emphasis on relatedness for females and achievement and autonomy
for males (Chevron, Quinlan, & Blatt, 1978; Gilligan, 1982). Accordingly, one
could expect that anaclitic psychopathology is more prevalent in women and
introjective psychopathology in men. However, according to Blatt (e.g., Blatt,
Quinlan, Chevron, McDonald, & Zuroff, 1982), gender incongruence might also
play an important role. That is, for men high dependency and for women high self-
critical perfectionism might be associated with an increased risk for
psychopathology.
Possible limitations of this model of normal and pathological personality
development include the fact that personality development is perhaps
conceptualized in a too linear-causal way. However, Blatt (1974; Blatt & Shichman,
1983) explicitly refers to the fact that on each line progression and regression is
possible. Thus, on each line, patients will tend to show - especially in stressful
situations - a mix of different forms and levels of psychopathology that are typical
for that developmental line. Moreover, the theoretical model also leaves room for
the role of deferred action (Nachtriiglichkeit) (Blatt, 1974). A second possible
limitation concerns the assumption in Blatt's original formulations (Blatt, 1991 a;
Blatt & Shichman, 1983) that individuals "choose" early in life between the two
developmental lines. However, even Blatt's original formulations did not suggest a
rigid dichotomy between anaclitic and introjective (Zuroff et aI., 2004), but
included the possibility that patients show both anaclitic and introjective features.
Furthermore, it is a common misunderstanding to interpret the distinction between
anaclitic and introjective as a qualitative or even dichotomous distinction, instead
as one of relative emphasis. Blatt's theory emphasizes that the distinction between
anaclitic and introjective should be made on the basis of which developmental line
predominates and which line is relatively neglected or defended against. In
addition, according to Blatt (e.g., Blatt & Shichman, 1983), "normality" is precisely
defined by a synergistic dialectic interaction between these two developmental
lines (although this does not mean that also within the "normal" range individuals
can differ in relative emphasis on both lines), whereas psychopathology is
characterized by a rigid emphasis on one line and the neglect or defensive
avoidance of the other. For both reasons, anaclitic and introjective characteristics
should not be considered to be mutually exclusive. Yet, notwithstanding these

71
important nuances, there has been a dearth of theoretical and especially empirical
research on so-called "mixed anaclitic-introjective" patients. Recent theoretical
developments and empirical research that addresses these issues are discussed in
detail by Blatt and Shahar in Chapter 5 (see also Zuroff et aI., 2004; Shahar, Blatt,
& Ford, 2003).

Beck's Theory of Normal and Psychopathological Personality Development

Over the last two decades, Beck's views have evolved from a fairly simple
schema theory towards a psychoevolutionary theory that conceptualizes
pathological personality development, as does Blatt, as an exaggerated or distorted
form of normal personality development (Beck, 1991, 1999; Beck & Freeman,
1990).
According to Beck ( 1999), psychological disorders can be seen as basic patterns
of behavior that once had survival value, but that have often become maladaptive in
our current modern society (the so-called "evolutionary friction rub"). For instance,
depression according to Beck (1999) is a reaction to a loss that reactivates an
"innate program consisting of giving up and withdrawal (in other words,
depression) [which] would serve to reduce the individual's "needs" until new
resources were developed" (Beck, 1999, p. 413). This innate program once clearly
had survival value, because in the wild the loss of a partner, for instance, could
mean that one had to survive for a considerable amount of time with limited
resources. However, in our society, this is no longer the case and thus such innate
programs often become maladaptive. Hence, such innate programs should largely
be seen as "anachronistic symbols inherited from the past" (Beck, 1999, p. 413).
Beck (Beck, 1999; Beck & Freeman, 1990) has proposed that such patterns or
strategies make up normal personality development. A certain "harmony" with and
adaptation to the environment characterizes, according to Beck, "normal" or
"functional personality". However, in an exaggerated form these patterns or
strategies give rise to different forms of psychopathology. Thus, psychopathology
arises when these strategies are used excessively, compulsively and/or
inappropriately (Beck, 1999). Moreover, Beck proposes that these strategies are
associated with specific beliefs, attitudes, and assumptions (Beck, 1999). These
strategies and accompanying core beliefs form the basis of Beck's classification of
personality disorders summarized in Table 1.
Furthermore, Beck's (1983, 1999) most recent views clearly exemplify the so-
called "psychoanalytic drift" (Power, 1991; Milton, 2001) in cognitive psychology.
Under the influence of psychodynamic formulations, Beck (1983, 1999) now
distinguishes between two basic "self-concepts": on the one hand the self-concept
"I am helpless", which according to Beck refers to survival; on the other hand, a
self-concept centered around the core belief "I am unlovable", which refers to
relatedness and attachment. Beck (1983, 1999) also calls these two basic self-

72
concepts sociotropic and autonomous personality dimensions or "modes".
Sociotropic individuals according to Beck are fundamentally oriented toward
interpersonal relationships. Their self-esteem is mainly dependent on receiving
love, care and acceptance from others. Autonomous individuals on the other hand,
are predominantly oriented toward autonomy, control, and independence, and thus
their self-esteem is mainly dependent on the achievement of goals and control.
Although some important differences have been identified, sociotropy and
autonomy resemble in important respects, at least at a descriptive level, Blatt's
concepts of relatedness and self-definition (Blatt & Maroudas, 1992; Clark & Beck,
1999; Zuroff et aI., 2004).
Moreover, like Blatt (Blatt & Levy, 1998; Blatt & Shichman, 1983), Beck (1999)
has proposed that in an exaggerated form, these two personality dimensions give
rise to two clusters ofpsychopathology that match in general Blatt's categorization
of personality disorders into anaclitic and introjective clusters. According to Beck
(1999) the dependent, histrionic, avoidant, passive-aggressive and borderline'
personality disorder belong to the sociotropic cluster, whereas the schizoid,
paranoid, narcissistic, antisocial, and obsessive-compulsive personality disorders
make up the autonomous cluster.

TABLE I
Strategies and core beliefs in personality disorders (Based on Beck. 1999)

Strategy Core belief Personality disorder

Predatory People are there to be taken Antisocial

Help-eliciting I am helpless Dependent
Competitive I am special Narcissistic
Exhibitionistic I need to impress Histrionic
Autonomous I need plenty of space Schizoid
Defensive People are potential adversaries Paranoid
Withdrawal I may get hurt Avoidant
Ritualistic Errors are bad Compulsive

Like Blatt, Beck's theoretical formulations have some important limitations (see
Blatt & Maroudas, 1992). First, Beck's assumption of a link between evolutionary
inherited strategies and Axis I and Axis II disorders is very speculative. Moreover,
we believe that Beck's conceptualization of depression as an "anachronistic symbol
inherited from the past" (Beck, 1999, p. 413) is not only speculative, but is also
based on an overly rationalistic human anthropology which assumes that loss (e.g.,
loss of a partner, of a job, etc.) in our contemporary society should in fact not be
associated with feelings of sadness and depression. However, we believe that such
feelings are part and parcel of the human condition precisely because human beings
possess the capacity to emotionally invest in what they value. Thus, feelings of
sadness and depression after loss appear to be the other side of the coin of the
human capacity for caring. Third, it has been repeatedly pointed out that Beck's
terminology is often imprecise and vague (Clark & Beck, 1999). For instance, the
relationship between concepts such as schemas, beliefs, automatic thoughts,

, Note that the borderline personality disorder is not included in Table I. For a discussion, see below.

73
personality dimensions, and "modes" is often not clear. Fourth, Beck's theory only
refers to content, but not to the structural level of psychopathology (e.g., psychotic,
borderline, neurotic). This lack of a structural perspective on personality
development and organization appears to be an important shortcoming of Beck's
views (see also Westen, 2000). This becomes clear, for example, when Beck
describes the reasons why the borderline personality disorder is not included in his
views about the relationships between evolutionary inherited strategies and beliefs.
According to Beck, this is due to the fact that patients with borderline personality
disorder not only combine several "core beliefs", but also "are uniquely
characterised by 'ego defects' in impulse control, affect stability, and reality testing
rather than by a specific content" (Beck, 1999, p.422), a description which is
heavily loaded with concepts that refer to a (psychodynamic) structural perspective
on psychopathology.

Advantages over the DSM Approach

Despite these possible limitations, both Blatt's and Beck's views offer some
important advantages over the atheoretical approach of DSM (Blatt, 2004; Blatt &
Levy, 1998). It is well known that precisely because of its symptomatic and
atheoretical approach, patients diagnosed with major depressive disorder (and other
mood disorders) according to the DSM show considerable heterogeneity in terms
of etiology, pathogenesis, prognosis, and treatment responsivity (Blatt, 2004; Van
Praag, 1998; Westen et aI., 2004). Moreover, the DSM is clearly inspired by a
disease model, which assumes that a disease is either present or absent based on a
counting of symptoms. However, research tends to favor a more continuous view of
depression, going from mild dysphoria to full-blown depression (Judd et aI., 1998;
Kendler & Gardner, 1998; see also Chapter I, this volume). Thus, in sum, a DSM
diagnosis gives little information concerning theoretically and clinically important
variables. In addition, the reification of the construct of depression by DSM, as if it
were a disease entity that is either present or absent and that has its own specific
etiology, pathogenesis, and treatment, hampers further research.
Blatt's and Beck's approach, in contrast, is clearly theoretical and geared
towards providing the clinician (and researcher) with clinically important
information. Blatt as well as Beck propose that psychopathology can be seen as an
exaggerated form of normal personality development, thus linking the field of
"normal" and "abnormal", "clinical" and "positive" psychology, and the study of
psychopathology and developmental psychopathology (see also Chapter 7, this
volume). In particular, their views imply that depression should not be seen as a
distinct disease entity, for which only some people are vulnerable. To the contrary,
depression is situated within a broader model of normal and pathological
personality development: it can range from mild dysphoria to clinical depression.
Moreover, clinical depression is not seen as a distinct disease, but as one possible

74
disorder among many other related disorders. Thus, as we will show in detail later
in this chapter, Blatt's and Beck's view that depression is related to personality
development and structure has immediate implications for theoretically and
clinically important information, such as etiology, pathogenesis, and treatment
responsivity. This opens up several new perspectives from a research as well as
from a clinical perspective.

Dependency/Sociotropy, Self-Critical Perfectionism/Autonomy,

and Depression

Introduction

Blatt's and Beck's views about depression share two basic assumptions which
are often misunderstood (Zuroff et aI., 2004). A first assumption is that both
dependency/sociotropy and self-critical perfectionism/autonomy are continuous
dimensions. Thus, individuals can show all combinations of dependency/sociotropy
and self-critical perfectionism/autonomy. However, Blatt's and Beck's views are
often interpreted as attempts to distinguish between two types of depression, i.e., an
anaclitic/sociotropic versus a perfectionistic/autonomous type of depression. The
fact that both Blatt and Beck often present their views, mainly for didactical purpo-
ses, in a way that suggests a typology, has considerably added to this misunde-
standing (Zuroff et aI., 2004). However, both authors clearly hold the view that all
combinations of dependent and perfectionistic features are possible and that only at
the extreme ends of the continuum do relatively "pure" cases of anaclitic and per-
fectionistic depression exist. We will return to this issue at the end of this chapter.
A second assumption concerns the fact that Blatt and Beck also advocate a
continuum view of depression (Blatt, 1974; Clark & Beck, 1999). This view holds
that there is a continuum going from "normal" or mild forms of dysphoria to
clinical, diagnosable depression, differing only in the severity (and perhaps also the
persistence) of symptoms. A discontinuity or categorical view of depression on the
other hand argues that there are not only quantitative, but also qualitative
differences between feelings of dysphoria and diagnosable depression (e.g., see
Vredenburg, Flett, & Krames, 1993 for a review). Although most research evidence
seems to support a continuum view, the debate on this issue is far from settled (e.g.,
Clark & Beck, 1999). However, if diagnosable depression is qualitatively different
from dysphoria, then studies investigating Blatt's and Beck's views in nonclinical
samples might not generalize to clinical samples (Coyne & Whiffen, 1995; Flett,
Hewitt, Endler, & Bagby, 1995; Ingram & Hamilton, 1999). In the next chapter,
where we will review empirical research concerning Blatt's and Beck's view, we
will return to this issue.

75
 Taking these remarks into account, we will in the remainder of this chapter
outline the main aspects of recent psychodynamic and cognitive-behavioral theories
of depression. Because Blatt's model is the most encompassing, it will form the
basis of our overview. However, we will also discuss important differences between
Blatt's views and those of Beck, Arieti and Bemporad, and Bowlby. In short, Blatt
(Blatt, 1974, 1998, 2004; Blatt & Shichman, 1983) proposes that general
predisposing factors (genetic, biological and environmental), in interaction with
specific environmental factors, can lead to disturbances in the anaclitic and/or
introjective developmental line, resulting in two personality dimensions or
structures that confer a specific vulnerability for depression. Moreover, both Blatt
(e.g., Blatt & Zuroff, 1992) and Beck (1983, 1999) have proposed that
dependentJsociotropic and self-critical/autonomous individuals are particularly
vulnerable for life events that match their personal vulnerability. This hypothesis,
which has been termed the personality-event congruency hypothesis (Robins,
1995), suggests that dependentJsociotropic individuals are more likely to become
dysphoric/depressed following negative interpersonal events, such as rejection or
loss, whereas selfcritical/autonomous individuals are more likely to become
depressed following negative achievement-related events, such as failure at work or
school. Both personality dimensions are also hypothesized to be associated with a
specific relational style that enhances the risk for depression. Final1y, when
depressed, both personality dimensions are expected to influence the clinical
presentation of depression. In what follows, we will for didactical reasons describe
these features in relatively "pure" types, keeping in mind that patients may show
"mixed" characteristics.

AnacliticlSociotropic Depression

Clinical presentation. The clinical presentation of dependent depressed patients

is characterized by feelings of loneliness, helplessness, weakness, and fears of
abandonment (Blatt, 1974). Often, these patients seek refuge in the use of alcohol,
drugs, or excessive eating ("oral traits") (Blatt et al., 1982). Anxiety and agitation
would often color the clinical picture ("anxious depression"). Depression would
also often be masked by or show itself in somatic complaints (Blatt & Shichman,
1983). Suicide attempts would be less violent or more "passive" than in
perfectionistic patients (e.g., overdose of medication) (Beck, 1983; Blatt et al.,
1982). In addition, Beck (1983) has proposed that the mood of these patients would
be more reactive to both positive and negative events (e.g., a new relationship
ameliorates symptoms). They are also expected to be very sensitive to even minor
frustrations. However, aggression would be denied or inhibited for fear of losing the
care and love of others on whom one is dependent (Blatt & Shichman, 1983). In
other respects, dependent depressed patients according to Blatt (1974) are
characterized by marked impulsivity and problematic impulse control, and/or a

76
hypomanic mood, which can result in "intense seeking and clinging to objects"
(Blatt, 1974, p. 118; Blatt & Shichman, 1983). They would also readily ask for
(professional) help, although often in a clinging and claiming way (Blatt & Levy,
1998). Finally, Beck (1983) argues that these patients are often overly optimistic
about treatment, resulting in a significant (but often temporary) relief of symptoms
(see also Blatt & Maroudas, 1992).
Personality structure. According to Blatt (e.g., Blatt & Shichman, 1983),
anaclitic depression is situated at the lower neurotic level. At this level, individuals
in whom anaclitic needs predominate show a number of specific personality
characteristics, which also differentiate them from individuals functioning
predominantly on higher (e.g., hysterical) or lower (e.g., hysteroid borderline)
anaclitic levels. First, they are overly dependent on others for their self-esteem.
They have a tendency to idealize significant others and prefer symbiotic-like
relationships, reminiscent of what Kohut and Wolf (1978) have called "merger-
hungry" or "ideal-hungry" persons. When depressed, Arieti and Bemporad (1978)
describe their depression in terms of a dominant other or claiming type of
depression, because these patients typically are overly dependent on one or more
"dominant others". Sometimes their exaggerated needs for dependency are
expressed in what Bowlby (1980) has described as "compulsive caregiving",
Compulsive caregivers typically feel guilty and anxious about leaving home, while
simultaneously feeling resentful at being forced to stay at home, mostly to care for
a disturbed or sick parent. However, they are highly dependent on the person(s) they
care for and/or on the caring as such in that they care for others in the way they
would like to be cared for (Blatt et al., 1982; Blatt & Maroudas, 1992).
Second, concerning ego functioning, dependent individuals heavily rely on
denial, especially of aggression towards significant others. They are very sensitive
to experiences of loss and abandonment (Blatt, 1974; Blatt & Shichman, 1983) and
attempt to deny the importance of such painful experiences by frenetic activity. This
denial would sometimes lead to hypomanic episodes, compulsive caregiving,
somatic complaints, eating disturbances, alcohol or drug abuse or the frantic search
for substitutes, which can result in hypersexual and/or promiscuous behavior. In
addition. these patients show a tendency to externalize their problems, often
accompanied by feelings of entitlement (see Kernberg, 1975). These feelings of
entitlement show up not only in a claiming attitude, but also in an often "silent"
rebelIion against authority and rules ("passive-aggressive"). For instance, in therapy
these patients demand extra therapeutic sessions or try to violate time limits of
sessions (Blatt, 1974; Beck, 1983). Finally, from a drive perspective, libidinal
issues and conflict dominate the clinical picture in dependent individuals (Blatt &
Shichman, 1983).
Characteristic interpersonal style. Relationships of dependent individuals would
be predominantly dyadic in nature, with little differentiation (Blatt, 1974; Blatt &
Zuroff, 1992). Others are primarily valued only for (immediate) need gratification
("self-objects"), and not as separate individuals with their own needs and desires.
They often have a strong need for immediate, visual contact with others (e.g., the
importance of "touching", "hugging"), resulting in a claiming relational style, with
much difficulty in tolerating delay and postponement. According to Blatt (1974),

77
self-representations and object-representations in these individuals are
predominantly on a sensorimotor level. This level refers to a stage in development
where others are only valued for direct, immediate need gratification.
Internalization of the object is incomplete, and thus there is much need for the real
and immediate physical presence of the object (Blatt, 1974).
Dependent individuals would also often have excessive fears of abandonment
and of object loss in general (Blatt & Shichman, 1983). Typically, aggression
towards significant others is denied or displaced, for fear of losing these significant
others ("rage threatens the very hand that feeds"). In addition, others are often put
in a domineering position in that dependency "pulls" for dominance from others.
Paradoxically, despite the fact that dependent individuals seek stability in
relationships, this stability is mostly not found. According to Blatt (e.g., Blatt &
Zuroff, 1992), the claiming, demanding, consuming relational style of these
individuals tends to evoke irritation and dissatisfaction in others, and therefore
often rejection and abandonment. Arieti and Bemporad (1978), in contrast,
maintain that these individuals often have long and stable, but overly dependent
relationships with one or more dominant others (e.g., parent, partner).
Proximal precipitating factors: The interaction with life events. It is well
established that there is a modest, though important, relationship between negative
life events and depression (Kessler, 1997; Tennant, 2002). Blatt and Beck take this
finding one step further in that they propose that dependent and perfectionistic
individuals are particularly vulnerable for life events that match their specific
personal vulnerability ("personality-event congruency hypothesis") (Blatt &
Zuroff, 1992; Clark & Beck, 1999; Robins, 1995). Thus, dependent individuals are
expected to be particularly vulnerable for "interpersonal" life events, such as
separation, divorce, or death of a significant other. However, dependent individuals
are not considered to be mere passive recipients of their environment (Blatt &
Zuroff, 1992; Zuroff, 1992). First, congruent with their underlying fear of rejection
and abandonment, they are expected to show a tendency to interpret life events in
terms of rejection or abandonment. Second, dependent individuals are also
expected to enhance the risk of congruent life events (Blatt & Zuroff, 1992; Zuroff
et al., 2004). Because of their claiming, consuming relational style, dependent
individuals are expected to invoke irritation and frustration in others, often leading
to rejection and abandonment and thus "congruent" life events.
Distal precipitating factors. The four groups of authors differ the most in their
descriptions of the distal antecedents of both personality dimensions. Beck (1983,
1999), consistent with his emphasis on the present, refers only in general terms to
"early life experiences" influencing the development of personality. Among the
psychodynamically oriented authors, there seems to be general agreement that
dependency is associated with parental inconsistency, neglect, or abandonment
and/or overprotection, resulting in (defensive) excessive dependency on others and
fear of rejection and abandonment. Beyond this point, there is much disagreement.
Blatt (1974) links the origin of dependency to disturbances in preoedipal phases,
i.e., issues related to the early separation-individuation phases. Bowlby (1980)
emphasizes that as a child, these individuals are made to feel guilty and/or
neglected by their parents, resulting in guilt over separation, excessive need for

78
love, and fear of rejection. Whereas Blatt and Bowlby propose that dependency and
self-critical perfectionism are associated with different parental styles, Arieti and
Bemporad (1978) maintain that both personality dimensions have their origin in
similar parenting styles. According to Arieti and Bemporad (1978), initially the
mother is appropriately responsive and often even too responsive to the child's
needs, resulting in a strong, often symbiotic-like relationship. However, with the
advent of the child's desires for autonomy and independence towards the second
year of life, the mother's (and later on father's) attitude to the child drastically
changes. From then on, love and care are given only conditionally and the child is
forced to conform to parental expectations in order to receive love. Depending on
the specific circumstances, the child then can either be forced to remain in a
dependent, submissive position, resulting in a clinging, dependent personality style,
or is forced to conform to parental expectations concerning achievement, resulting
in a dominant goal (or self-critical) type of personality.

IntrojectiveiAutonomous Depression

Clinical presentation. When depressed, perfectionistic/autonomous individuals

typically show self-criticism, guilt, shame, worthlessness, and often a chronic fear
of being criticized or disapproved. There is constant self-scrutiny, often together
with a feeling of having failed to live up to expectations (Beck, 1983; Blatt, 1974).
They often have the feeling that they are constantly being watched and criticized.
Self-criticism and guilt can become psychotic (e.g., delusions of poverty, feelings
of immortal sin, etc.). Instead of feeling unloved, a feeling that dominates
anaclitic/sociotropic depression, they would rather feel unlovable (Blatt, 1974).
Obsessive-compulsive and paranoid-like symptoms can be present (e.g., distrust, a
feeling of being constantly evaluated, delusions of punishment, etc.) (Arieti &
Bemporad, 1978; Blatt, 1995b; Blatt & Shichman, 1983). Suicide would be more
active and violent in these individuals (e.g., use of firearms) (Beck, 1983; Blatt et
aI., 1982).
According to Beck (1983), when depressed, these individuals withdraw from
personal contact, seek isolation, and are less likely to seek (professional) help.
Moreover, they are expected to be more pessimistic about being helped (e.g., about
psychotherapy), despite the fact that they often have a good capacity for
introspection (Beck, 1983; Blatt & Maroudas, 1992). Their depressed mood is
according to Beck (1983) also less reactive to positive and negative events and
immediate precipitating events can often not be identified. Instead, depression is
caused more by internal factors (e.g., the feeling of having not lived up to one's own
high standards). Hence, not surprisingly, Beck (1983) has proposed that his
distinction between a sociotropic and autonomous depression matches the
"classical" distinction between a reactive and an endogenous type of depression
(e.g., see Gillespie, 1929; Kiloh & Garside, 1963). Indeed, the differences between

79
autonomous and sociotropic depression in terms of symptomatology, reactivity of
mood, and presence or absence of proximal precipitating events overlaps with the
distinguishing features of endogenous and reactive depression. Blatt (1991 a, 1998;
Blatt, Schaffer, Bers, & Quinlan, 1992), however, maintains that distinctions
between types of depression based on symptoms are unproductive because they are
unlikely to lead to significant advances in our understanding of depression.
Personality structure. According to Blatt (e.g., Blatt & Shichman, 1983), the
personality structure of introjectively depressed individuals is situated at a higher
neurotic level. First, they typically have high and often rigid personal standards, in
combination with strong self-criticism. In psychodynamic terms this means that
they possess an overly harsh and rigid superego, in combination with an ego ideal
that is characterized by strivings for perfectionism and control (Blatt & Shichman,
1983; Blatt, I995b). Interestingly, Beck (1983) emphasizes in his theoretical
descriptions more the control and independence component, while Blatt (1974)
emphasizes more the high achievement standards and self-critical perfectionism of
these individuals. Arieti and Bemporad (1978) coined the notion of a dominant goal
personality. This dominant goal most often concerns professional issues (e.g., being
successful in business), but according to Arieti and Bemporad (1978) it can also
manifest itself in relationships. Relationships are then not so much valued for the
interpersonal aspect as such, but rather for fulfilling certain role expectations or
being "successful" in relations (e.g., the ideal of being a perfect husband, father and
lover).
A second characteristic personality feature of these indidivuals is their high level
of premorbid functioning (Arieti & Bemporad, 1978; Beck, 1983; Blatt, 1995b).
Before becoming depressed, they are often very successful and praised by their
environment for their achievements. However, equally characteristic is that they
often have little lasting satisfaction (Arieti & Bemporad, 1978; Blatt. 1974). In
addition, because of their "hypermorality", they frequently have the feeling of
being superior to "ordinary" people, who they look down upon (Arieti &
Bemporad, 1978). Thus, self-critical individuals remind us of what Nietzsche said
about Christians: "They say that they have been redeemed, but then they should
look more redeemed".
Traditionally, these individuals are expected to show a conflict over ambivalence
in the strict sense of the term, i.e., aggression is repressed and thus absent on the
manifest level (Freud, 1917/1957; Vergote, 1976). However, Blatt (1974; Blatt &
Zuroff, 1992), as well as Arieti and Bemporad (1978), Beck (1983), and Bowlby
(1980), maintain that these patients show a marked manifest ambivalence.
Relationships of perfectionisticlautonomous individuals in particular would be
characterized by much manifest ambivalence, and hence conflict and distance.
However, in this regard there is a small, but important difference in the theoretical
views of Beck, Bowlby, and Blatt. According to Beck (1983) and Bowlby (1980),
autonomous individuals want to remain relatively distant and aloof from others
("defensive separation"), while Blatt (Blatt & Maroudas, 1992) maintains that
introjective individuals do desire interpersonal contact and need others for
appreciation and approval, but that they also simultaneously fear their disapproval
and criticism.

80
 The ego functioning of perfectionistic individuals is expected to be characterized
by a tendency to assume blame and responsibility, and to attribute failures to
oneself (Blatt & Shichman, 1983). This self-blaming, internalizing attitude is often
combined with expectancies of punishment (Blatt, 1974). Not surprisingly,
perfectionism in depression has traditionally been linked to pathological
masochism (e.g., see Kernberg, 1992; Markson, 1993). From a drive perspective,
aggression and its derivatives (e.g., competitiveness) are expected to dominate the
personality structure of perfectionistic individuals.
Characteristic interpersonal style. As noted above, the interpersonal
relationships of perfectionistic individuals are described as rather distant and cold
(Blatt, 1974; Blatt & Zuroff, 1992). Beck (1983) and Bowlby (1980) maintain that
autonomous individuals strive for independence and control, and thus avoid close
interpersonal relationships, whereas Blatt (Blatt & Maroudas, 1992) stresses the
conflictual and ambivalent nature of the relationships of perfectionistic individuals.
On the one hand, they would desire interpersonal relations, especially for approval,
while at the same time fearing disapproval and criticism from others.
Moreover, according to Blatt (Blatt & Shichman, 1983), relationships of
introjective individuals are more triadic and thus more differentiated compared to
anaclitic individuals. Object-representations and self-representations of introjective
individuals are supposed to be at the developmentally more advanced perceptual
and iconic levels (Blatt, 1974). This means that objects are recognized beyond their
need-gratifying role, but that self-representations and object representations are still
fragmented (i.e., based on part properties), contradictory and ambivalent. Negative
part features of objects (such as criticism) and self (such as self-criticism) are
exaggerated and not integrated with other features of self and others.
Proximal precipitating events. In line with the congruency hypothesis, both Blatt
(Blatt & Zuroff, 1992) and Beck (1983) have proposed that perfectionis-
tic/autonomous individuals are particularly vulnerable for life events that match
their personal vulnerability. Thus, they are expected to be specifically vulnerable
for life events related to achievement and control, such as missing a job promotion
or failing at school or work. Again, however, Blatt (Blatt & Zuroff, 1992; Blatt et
al., 200 I) and Beck (1983; Clark & Beck, 1999) underline that introjective/auto-
nomous individuals, in part, actively select, interpret and create their own environ-
ment. Hence, they are not only seen as particularly sensitive to failure, but they are
also expected to enhance the risk of failure and thus congruent life events because
of their often excessively high perfectionistic standards.
Distal precipitating factors. As for dependency/sociotropy, there is also
disagreement among the four groups of authors concerning the developmental
origins of perfectionism/autonomy. Whereas Beck's (1983) theory does not address
the developmental origin of autonomy, Arieti and Bemporad (1978) emphasize high
parental expectations concerning achievement. Bowlby (1980) has proposed that
his anxious-avoidant type of attachment is associated with early loss and
subsequent inadequate care and/or overcritical attachment figures, resulting in a
tendency to be overly self-reliant and aloof. In a similar vein, Blatt (1974) suggests
that self-critical perfectionism results from the identification with overcritical,
hostile parental figures. In addition, reaction formation and overcompensation are

81
used in reaction to supposed failures to live up to personal or parental standards.
Like Arieti and Bemporad, Blatt (Blatt, 1974; Blatt & Homann, 1992) suggests that
parents of future perfectionistic individuals link love too strongly to the attainment
of parental standards and expectations, which prevents the development of real
autonomy and results in the inhibition of "normal" impulsivity and creativity (Blatt
& Homann, 1992). Finally, Blatt (1974) has proposed that the origin of self-critical
perfectionism is situated at a higher developmental level than dependency, namely
at the phallic and oedipal level, whereas Arieti and Bemporad (1978) maintain that
for both dependency and self-critical perfectionism the reaction of the parents to the
advent of strivings for autonomy around the age of two is crucial.

Clinical Implications: Do Different Kinds of Folks

Need Different Kinds of Strokes?

All four groups of authors have outlined detailed treatment principles that are
directly derived from their theoretical conceptualizations (see Blatt, 2004). Because
most research has concentrated on Blatt's and Beck's formulations, we will restrict
our discussion mainly to their views. In short, Blatt (e.g., Blatt & Felsen, 1993;
Blatt, Shahar, & Zuroff, 2002) as well as Beck (1983) have argued that (predomi-
nantly) dependent/sociotropic and perfectionistic/autonomous individuals have
very different needs and expectations about treatment and thus respond differently
to (and often to different) treatment-related events (Blatt et al., 2002). Accordingly,
both Blatt and Beck have suggested that these patients demand different therapeutic
approaches and perhaps even different forms of psychotherapy. As Blatt puts it:
"different folks may need different kinds of strokes" (Blatt & Felsen, 1993).
Especially in the early phases of psychotherapy, dependent/sociotropic and
perfectionistic/autonomous patients are expected to show marked differences in the
central transference and countertransference themes. Transference (and counter-
transference) themes in anaclitic/sociotropic patients are likely to center on
libidinal issues, such as care, intimacy, loss, rejection, and abandonment; whereas
in introjective/autonomous patients, issues such as aggression, power, control,
autonomy, self-worth, criticism, self-criticism, and guilt are likely to be central.
Thus, in the early phases, the therapeutic relationship is colored by "personality-
congruent" themes, while themes from the "other" developmental line only emerge
after some therapeutic work has been done.
Dependent/sociotropic and perfectionistic/autonomous patients are also
expected to respond initially to different dimensions in the psychotherapeutic
setting and thus to demand somewhat different therapeutic approaches (Beck, 1983;
Blatt et al., 2002). According to Blatt (Behrends & Blatt, 1985; Blatt & Behrends,
1987), one can distinguish two general therapeutic factors in any form of
psychotherapy, namely insight (explanation, interpretation) on the one hand and the
therapeutic relationship on the other hand. In the initial phase of therapy,

82
dependent/sociotropic patients are likely to be more responsive to the interpersonal
and supportive dimension of psychotherapy and thus to the therapeutic relationship
as such. Although explanations are readily accepted, it is not so much the concrete
content or the correctness of these explanations that are important, but rather the
fact that these explanations are interpreted in terms of empathy and support and
thus strengthen the therapeutic relationship (Beck, 1983; Blatt & Maroudas, 1992).
Self-critical/autonomous patients, on the other hand, because of their more
"intellectual" style and their difficulties in relating to others, respond initially better
to the interpretative aspects of psychotherapy (e.g., explanation, interpretation,
insight), than to aspects related to the therapeutic relationship. The therapeutic
relationship is initially likely to be colored by the ambivalence, need for control,
and even somewhat distrustful, paranoid features that characterize these patients.
Thus, self-critical/autonomous patients are expected to need more time to establish
a safe and "good enough" therapeutic alliance (Blatt & Maroudas, 1992).
However, congruent with Blatt's theory of "normal" personality development
and his characterization of psychopathology as a rigid overemphasis on one
developmental line at the expense of the other, eventually both type of patients must
deal with and thus work through issues of both developmental lines. Thus, during
the therapeutic process, dependent/sociotropic patients eventually have to deal with
introjective issues (such as self-definition, autonomy, and identity), whereas self-
critical/autonomous patients eventually have to deal with interpersonal themes
(such as intimacy, love, and care) (Blatt & Maroudas, 1992). If treatment goes well,
discussion and working through of interpersonal issues in dependent/sociotropic
patients leads to a working through of themes related to self-definition and
autonomy. After some time, they come to realize that they have been overly
dependent on one or more significant others and that their life has often been
completely dominated by their tendency to please and comfort others. According to
Arieti and Bemporad (1978), this is an important turning point in the psychotherapy
of these patients. Moreover, this phase is often characterized by much anger (which
previously was denied or displaced) towards significant others. Because of this
anger, and because of the fact that the patient is changing and is becoming more
assertive in general, significant others, such as the partner, often try to undermine
the therapeutic process at this point. Thus, according to Arieti and Bemporad
(1978), couple or family therapy should routinely be considered for these patients.
In self-critical/autonomous patients, the therapeutic relationship appears to form
the vehicle for discussing and working through interpersonal themes, such as trust,
love, and intimacy. The therapeutic situation thus can be seen in these patients as
an "interpersonal laboratory", where they can "experiment" with relationships. In
our view, benevolent neutrality, which counteracts these patients' fear of criticism
and disapproval in combination with the interpretation of transference and counter-
transference issues concerning relatedness once a good therapeutic alliance has
been established, is vital in this context. However, it can take some time to establish
such a good therapeutic alliance. And even then it might often not be simple to
overcome these individuals' fear of criticism and sometimes even downward
distrust of the good intentions of others, as well as their tendency to rationalize or
intellectualize emotions and relationships.

83
 In sum, Blatt has proposed that one can conceptualize the therapeutic process as
a reinitiation of normal psychological development or the "reactivation of [a]
disrupted developmental process" (Blatt & Shichman, 1983, p. 249), namely of a
dialectical interaction between relatedness (anaclitic line) and separation-
individuation (introjective line) towards more differentiated and integrated inter-
personal relationships and self-definition. Conceptualizing the therapeutic process
in this way clearly shows that these formulations are not limited to patients that can
easily be categorized as predominantly anaclitic or introjective (i.e., relatively
"pure" types), because issues of attachment/relatedness and separation/identity are
important issues in all patients. In addition, these formulations are relevant for each
psychotherapist, regardless of theoretical orientation and regardless of whether
short-term or long-term treatment is considered (Blatt & Maroudas, 1992).
According to Blatt it is moreover likely that anaclitic and introjective patients
need different kinds or forms of psychotherapy (Blatt & Felsen, 1993). In this
context, Blatt refers to the fact that decades of psychotherapy research has not only
identified few specific psychotherapeutic factors, but also few differences in the
effectiveness of different forms of psychotherapy (the famous "Dodo bird verdict")
(Blatt et al., 2002; Luborsky et al., 1993). In this respect, Blatt (Blatt & Felsen,
1993) argues that patient (and therapist) characteristics are potentially much more
important than supposed differences between the various forms of psychotherapy
and thus that one can expect patient-therapy interactions (see also Beutler, 1991;
Beutler, Clarkin, & Bongar, 2000). For instance, anaclitic patients can be expected
to respond better to more supportive, and possibly standardized forms of
psychotherapy, in which the therapist is more active, supportive, and directive.
Introjective patients are expected to benefit more from long-term, interpretative
forms of psychotherapy, in which insight plays a central role and the autonomy and
control, so important for these patients, is not restricted by the directive stance of
the therapist or by the arbitrary amount of sessions fixed in advance, as is
characteristic in short-term, standardized forms of treatments.
However, it is important to underline that we are not proposing here that because
dependent patients might respond well to short-term psychotherapy, which is often
more supportive, that they should not be referred to long-term psychotherapy, and
vice versa for introjective patients. While short-term treatment of depression may
result in symptom reduction and sometimes also in lasting changes in personality
structure, many, if not most, patients do not show significant changes in personality
structure after short-term psychotherapy. This is not mere speculation. In fact,
research has shown that short-term treatment has limited effects, even in terms of
symptom reduction (Westen et aI., 2004). As noted in the introduction to this
volume, on average only about 50% of patients show a substantial reduction in
symptoms after brief psychotherapy. Moreover, studies concerning the long-term
effects of short-term psychotherapy are largely lacking and the few studies that do
exist show relapse rates around 70% after two years (Westen & Morrison, 20(1).
Congruent with these findings, we believe that for many patients, short-term
psychotherapy does not result in lasting changes in personality structure, which
could partly explain high relapse rates after such treatments. Thus, not only
perfectionistic, but many dependent patients may also need long-term

84
psychotherapy to work through both anaclitic and introjective issues (Luyten et aI.,
2003). This issue will be discussed in more detail in the next chapter, in which we
will review relevant empirical research.

Barriers to Integration between Cognitive

and Psychodynamic Theories of Depression

The literature reviewed in this chapter clearly indicates a clear trend towards
integration between psychodynamic and cognitive thinking concerning depression.
However, it should be underlined that up to this point we mainly emphasized the
similarities between the theories of Blatt, Beck, Arieti and Bemporad, and Bowlby.
However, differences between these theories have been identified (e.g., Blatt &
Maroudas, 1992), as well as between psychodynamic and cognitive-behavioral for-
mulations in general (Millon, 200 I; Westen, 2000). These differences might pose
important barriers towards further integration.
We believe that such differences are of two types. First, there are what we
refer to as minor differences between these theories which concern concrete
hypotheses at low levels of theorizing. In our view, these minor differences do
not constitute a real barrier towards further integration because they can be easily
translated into research hypotheses, and thus can subsequently lead to changes
in theoretical assumptions. For instance, the four groups of authors discussed
in this chapter differ in their conceptualization of the developmental origins
of dependency and perfectionism. Future research could investigate these
various hypotheses and then the four groups of authors could subsequently mo-
dify their views depending on the outcome of these studies. Another example
of such a minor difference is the fact that Blatt and Beck differ with regard
to the capacity for introspection of dependent individuals. According to Beck
(1983), these individuals show a good capacity for introspection, whereas Blatt
(1974) maintains that dependent individuals have very limited capacities for
introspection.
Whereas these minor differences can be easily put to the test, other more fun-
damental or substantial differences are more difficult, if not impossible, to evaluate
empirically. Moreover, these differences do form important barriers towards further
integration. In this context, the views of Safran and Inck (1995) concerning
integration between various psychotherapeutic orientations or "schools" are
particularly interesting. Briefly, they have argued that the possibilities of integration
between various theoretical orientations vary according to the level of theorizing.
At lower levels of theoretical conceptualizations, often few differences are
identifiable between various theoretical models, offering considerable possibilities
for integration. This certainly seems to be true for Beck's and Blatt's models, which
show many similarities at concrete or low levels of theorizing. On higher levels of
theoretical reasoning, however, substantial differences often exist between different

85
theoretical views, making integration much more difficult, if not impossible.
Ultimately, on the highest level of theorizing, differences concern the concept of
human nature and the worldview underlying a particular theoretical model or
orientation (Safran & Inck, 1995). At this particular level, major differences exist
between the psychodynamic views of Blatt (and Bowlby and Arieti and Bemporad)
and Beck's cognitive-behavioral theories (see also Milton, 2001).
Whereas both Beck and Blatt conceptualize dependency/sociotropy and self-
critical perfectionism/autonomy as central dimensions in personality development,
their underlying concept of human nature is very different. These differences center
particularly on Blatt's and Beck's ideas concerning the relationship between
"normality" and psychopathology and their subsequent views on psychotherapy. In
accordance with the cognitive-behavorial perspective, Beck (Beck & Freeman,
1990; Clark & Beck, 1999) sees human beings as rational scientists that can be
dominated by (inherited or acquired) irrational schemas, which can be adjusted or
"replaced" in short-term psychotherapy. Thus, both implicitly and explicitly, Beck
assumes a more or less clear-cut difference between "normality" and
psychopathology, and this despite the fact that Beck (like Blatt) proposes that
psychopathology is in fact an exaggeration of normal personality development.
Beck's view is intimately associated with a strong belief in reason and rationality
and with a strong therapeutic optimism. From a cognitive-behavioral perspective,
"we know what is wrong" or we will at least be able in the future to identify what
is "wrong". For instance, we know when a particular depressed patient has a
"maladaptive" belief. Put differently: we know what a "normal", "rational" or
"adaptive" schema is and thus we can change or try to change "irrational" schemas.
Although in recent years Beck's therapeutic optimism has been tempered somewhat
(e.g., see Beck, 1999), his theory (e.g., Beck, 1991) - and the cognitive-behavioral
literature in general - is still characterized by a strong belief in the possibilities of
(short-term) cognitive behavioral therapy (for a good recent example of this
optimism, see Hollon, 2003).
In contrast, and consistent with classic psychoanalytic thinking, Blatt's theory
implies an essential continuity between normality and psychopathology. This
means, in other words, that human beings are expected to be always troubled by
conflict and frustration. Conflict is considered to be - to a certain extent - normal
and thus is viewed as an intrinsic part of the human condition. A life without
conflict and frustration is impossible and thus the possibility of psychopathology is
present in everyone, and not just in some unhappy few who happen to have some
special vulnerability or dysfunctional schemas. Moreover, for some (possibly for
many), psychological problems we have no "rational", "normal" or "good"
solutions. For instance, how should one deal with the death of a loved one? Is there
a "rational", "adaptive", "normal" way to deal with this? Or, is there a "rational"
way to deal with an overcritical parent? In addition, are some patients not correct
in seeing themselves as, for instance, aloof, overcritical, and unlovable (Westen,
2oo0)?
This more "tragic" view of human nature (Westen, 2000) is associated with a
completely different outlook on treatment. To begin with, in the therapeutic process
there is not only an emphasis on reason and rationality (insight), but also on what

86
since Alexander and French (1946) has come to be known as (corrective) emotional
experience. Congruent with this view, Blatt (e.g., Blatt & Behrends, 1987; Blatt &
Maroudas, 1992) describes the therapeutic process as a series of gratifying
involvements at a succession of different developmental levels of relatedness and
self-definition, leading to the internalization of these experiences. Thus, the
therapeutic process enables the patient to reenact and subsequently renegotiate
these issues. In addition, ultimately, the goal of psychodynamic psychotherapy is
not only to obtain symptomatic improvement or to teach patients more adaptive
modes of dealing with their current life circumstances, but to produce a significant
change in personality structure, thus reducing vulnerability to relapse. While this is
also the ultimate goal of cognitive-behavioral therapy (e.g., Beck & Freeman,
1990), studies of cognitive-behavioral therapy have generally tended to focus on
short-term symptom improvement and not on changes in long-term reduction in
vulnerability (Westen & Morrison, 200 I). Moreover, from a psychodynamic point
of view, even changing vulnerability does not imply a "total cure", whatever one
may understand by that term, since vulnerability to psychopathology is part and
parcel of the human condition. Stated otherwise, as Freud (1895/1953, p. 305) once
said, psychotherapy can be nothing but transforming misery into common human
unhappiness. Thus, this view implies a more pessimistic - or should we perhaps say
realistic - view of therapeutic success.
These substantial differences are bound to limit the possibility of further
integration on a theoretical level (Arkowitz, 1997; Messer, 1986; Westen, 2000)
and this despite the fact that there is an increasing tendency towards technical
integration (i.e., the use of cognitive-behavioral techniques in psychodynamic
psychotherapy and vice versa, e.g., see Jones & Pulos, 1993) and an emphasis on
common factors (such as the therapeutic relationship, e.g., see Waddington, 2002).

Conclusions and Future Perspectives for Integration

Psychodynamic and cognitive-behavioral theories of depression have proposed

similar personality dimensions, i.e., interpersonal dependency/sociotropy and self-
critical perfectionism/autonomy, as vulnerability factors for clinical and nonclinical
forms of depression. Moreover, these theories are embedded in more general
theoretical frameworks concerning normal and disrupted personality development.
Having their origin in clinical experience as well as in empirical research, these
theoretical perspectives provide both the researcher and the clinician with a rich
heuristic theoretical framework.
Though these theories have a remarkable synthetic and heuristic power, it is clear
that they also have a number of limitations. First, despite the fact that both Blatt and
Beck have conceptualized dependency/sociotropy and perfectionism/autonomy as
continuous dimensions, there is an urgent need for more theoretical and empirical
research concerning "mixed" types (Coyne & Whiffen, 1995; Flett et al., 1995).

87
One likely candidate for such a "mixed" type of depression seems to be the so-
called depressive-masochistic personality (e.g., Kernberg, 1992). Traditional
descriptions suggest that these individuals are characterized by a severe superego,
marked dependency, and difficulties in the expression of aggression, thus
combining important aspects of both perfectionism and dependency. The issue of
mixed types is further discussed in Chapter 5.
Second, because of their focus on broad personality dimensions, the dynamics
involved in depression are somewhat less emphasized by these theories. In classical
psychodynamic theorizing, for instance, depression is conceptualized as an active
and basic psychobiological reaction to loss (e.g., see Bibring, 1953; Engel, 1962;
Sandler & Joffe, 1965; Spitz, 1946). This implies that depression is not just a
passive state, but an active and to a certain extent adaptive reaction to loss. In
addition, classic psychodynamic descriptions of depression have emphasized the
dynamics involved in discrepancies between ego and ego ideal (Bleichmar, 1996;
Lax, 1989; Miller, 1979; Milrod, 1988; Morrison, 1989; Vergote, 1976). Because
Blatt and Beck focus on the role of personality dimensions in depression, they place
less emphasis on these detailed dynamic formulations. Future theoretical and
empirical research should therefore concentrate more on the dynamics involvec in
dependency/sociotropy and perfectionism/autonomy. For instance, psychodynamic
formulations suggest that both dependent/sociotropic and self-critical/autonomous
individuals have much difficulty in abandoning unattainable goals or formulating
more realistic goals, even in the face of repeated failures and/or frustrations. Future
studies could focus on the dynamics (e.g., defensive processes, underlying
narcissistic fantasies, etc.) involved in this inability to relinquish unattainable goals.
Despite these limitations, these theories emphasize the importance of
considering personality factors in depression. Thus, they see depression not as an
isolated disease entity, but as the result of distortions of normal development. As we
said earlier, this view opens up several interesting perspectives for the integration
between various psychological and biological approaches to depression (Blatt &
Maroudas, 1992). First, concerning the integration between psychological theories
of depression, this view makes integration possible between various domains in
psychological research that have often developed in relative isolation, such as the
relationships between "normal" and "psychopathological" development, between
life-span developmental theory and developmental psychopathology, and between
"clinical" and "positive" psychology. Second, concerning the integration between
psychological and biological approaches, research has made increasingly clear that
biological, social, and psychological processes are intimately interwoven in both
normal and pathological personality development (Rutter et al., 1997). Research on
depression could therefore benefit from being embedded in a more encompassing
biopsychosocial theoretical framework concerning normal and pathological
personality development, such as the one proposed by Blatt and Beck (Abramson,
Alloy. Hankin, Haeffel, MacCoon, & Gibb, 2002). For instance, recent studies have
lent considerable evidence for passive and active person-environment correlations
and interactions (see Rutter et aI., 1997, for an overview), linking temperamental
and personality characteristics to neurobiological and genetic factors on the one
hand and environmental factors on the other hand. These views are congruent With

88
both Blatt's (Blatt & Zuroff, 1992) and Beck's (1983) emphasis on the active
influence of anaclitic/sociotropic and self-critical/autonomous individuals on their
environment. Moreover, Blatt's (Blatt, Cornell, & Eshkol, 1993; Blatt & Maroudas,
1992) and Beck's (1983) formulations explicitly refer to the (neuro)biological
underpinnings of personality. Blatt has hypothesized that both dependency and self-
critical perfectionism may have genetic roots (Blatt & Shichman, 1981) and are
related to biological processes involved in neoplastic and cardiovascular disease,
respectively (Blatt, Cornell, & Eshkol, 1993). In addition, Blatt (Blatt & Maroudas,
1992) has proposed that dependent individuals may initially show a positive
placebo effect to antidepressant treatment, because of their optimism regarding
treatment, especially when this treatment can be experienced by the patient as
"being cared for and fed" by a physician. Self-critical patients, on the contrary, are
expected to show a negative placebo effect, because of their pessimism regarding
treatment and because they would feel unworthy of treatment. Beck ( 1983), in tum,
has hypothesized that the autonomous type of depression is similar to the concept
of endogenous depression and hence is more biological in origin. Thus,
autonomous patients are expected to show a better response to antidepressant
treatment than sociotropic patients.
Moreover, Blatt's and Beck's formulations are also congruent with recent studies
showing that early life stress (ranging from severe trauma or neglect to low quality
of parental care) is associated with biopsychosocial vulnerability to current life
stress (e.g., Davidson, Pizzagalli, & Nitschke, 2002; Gunnar, 1998; see also
Chapter 8, this volume). In this context, an increasing number of studies show that
early adversity and later life stress have a profound impact on the hypothalamic-
pituitary-adrenocortical (HPA) axis (Kandel, 1999; Gunnar, 1998; Tsigos &
Chrousos, 2002), which has been linked to depression (Raison & Miller, 2003).
Blatt's theory is also compatible with studies on "kindling" and "scarring" (see
Chapter I, this volume) since it emphasizes the recursive, cyclical interaction
between personality, environment, and depression (Zuroff et al., 2004). In sum, we
believe these findings and formulations clearly emphasize the urgent need for more
integrative biopsychosocial theoretical frameworks regarding normal and
pathological personality development (see also Chapters 1 and 8 in this volume).
Research concerning such integrative theories could not only improve our insight
into mood and other disorders, but is also likely to have important therapeutic
implications. We will return to these issues in the Epilogue of this volume.
However, the conceptualization of depression, not as an isolated disorder, but as
a distortion of normal personality development that can range from mild dysphoric
reactions to full-blown clinical depression, may also constitute a main obstacle to
the further integration between these theories and mainstream research on
depression. Currently dominant theories of depression, in line with the DSM, view
various forms of depression and other disorders as categorically distinct disease
entities. Moreover, the DSM explicitly avoids any etiological considerations in the
classification of mental disorders. Despite the fact that many of the key assumptions
underlying DSM have been invalidated and thus hamper further research (Blatt &
Levy, 1998; Westen & Shedler, 1999; Wampold, 1997), including the search for
genetic and other biological factors involved in depression (e.g., Van Praag, 1998),

89
the DSM remains the gold standard in mainstream depression research. Until this
tension between descriptive and etiological viewpoints is solved, integration among
the views of Blatt, Beck, Arieti and Bemporad and Bowlby with DSM-inspired
research on mood and other disorders will be difficult.

References

Abramson, L. Y., Alloy, L. B., Hankin, B. L.. Haeffel. G. J., MacCoon, D. G., & Gibb, B. E. (2002).
Cognitive-vulnerability-stress models of depression in a self-regulatory and psychobiological context.
In I. H. Gotlib & C. L. Hammen (Eds.), Handbook ofdepression (pp. 268-294). New YorkILondon: The
Guilford Press.
Alexander, E, & French. 1'. M. (1946). Psychoanalytic therapy. Principles and application. New York: The
Ronals Press Company.
Arieti, S., & Bemporad, J. (1978). Psychotherapy of severe and mild depression. Northvale/London: Jason
Aronson.
Arkowitz, H. (1997). Integrative theories of therapy. In P. Wachtel & S. Messer (Eds.), Theories of
psychotherapy: Origins and evolution (pp. 227-288). Washington. DC: American Psychological
Association Press.
Baker Miller, J. (1976). Toward a nell' psychology of women. London: Penguin Books.
Beck, A. T. (1983). Cognitive therapy of depression: New perspectives. In P. J. Clayton & J. E. Barrett
(Eds.), Treatment o] depression: Old contro-versies and nell' approaches (pp. 265-290). New York:
Raven Press.
Beck. A. 1'. (1991). Cognitive therapy: A 30-year retrospective. American Psychologist. 46, 368-375.
Beck. A. 1'. (1999). Cognitive aspects of personality disorders and their relation to syndromal disorders: A
psychoevolutionary approach. In C. R. Cloninger (Ed.), Personality and psychopathology (pp. 411-
429). Washington. DC/London: American Psychiatric Press.
Beck. A. 1'., & Freeman. A. (1990). Cognitive therapy (if personality disorders. New York: Guilford Press.
Behrends. R. S.• & Blatt. S. J. (1985). Internalization and psychological development throughout the life
cycle. Psychoanalytic Study ofthe Child. 40. 11-39.
Bemporad, J. R. (1992). Psychoanalytically orientated psychotherapy. In E. S. Paykel (Ed.), Handbook of
affective disorders (2nd ed.) (pp. 465-473). Edinburgh: Churchill Livingstone.
Beutler. L. E. (1991). Have all won and must all have prizes? Revisiting Luborsky et aI.'s verdict. Journal
of Consulting and Clinical Psychologv, 59, 226-232.
Beutler. L. E.. Clarkin, J. E. & Bongar, B. (2000). Guidelines for the systematic treatment ofthe depressed
patient, New York/Oxford: Oxford University Press.
Bibring, E. (1953). The mechanism of depression. In P. Greenacre (Ed.), Affective disorders.
Psychoanalytic contribution to their study (pp. 13-48). New York: International Universities Press.
Blass, R. B.. & Blatt. S. J. (1996). Attachment and separateness in the experience of symbiotic relatedness.
Psychoanalytic Quarterly, 65. 711-746.
Blatt, S. (1974). Levels of object representation in anaclitic and introjective depression. The Psvchoanatvtic
Study of the Child. 29, 107-157.
Blatt. S. J. (l99Ial. A cognitive morphology of psychopathology. Journal of Nervous and Mental Disease,
/79. 449-458.
Blatt. S. 1. (1991 b). Depression and destructive risk-taking behavior in adolescence. In L. P. Lipsitt & L.
L. Mitnick (Eds.), Self-regulatory behavior and risk-faking: Causes and consequences (pp. 285-:,0'1).
Norwood. NJ: Ablex Press.
Blatt. S. J. (1995a). Representational structures in psychopathology. In D. Cicchetti & S. L. Toth (Eds.),
Emotion, cognition, and representation (Rochester Symposium on developmenal psychopathology \-'.,1.
6) (pp. 1-33). New York: University of Rochester Press.
Blatt. S. J. (l995b). The destructiveness of perfectionism. Implications for the treatment of depression,
American Psychologist, 50. 1003-1020.

90
Blatt, S. 1. (1998). Contributions of psychoanalysis to the understanding and treatment of depression.
Journal of the American Psychoanalytic Association, 46, 722-752.
Blatt, S. J. (2004). Experiences ofdepression: Theoretical, clinical and research perspectives. Washington,
DC: American Psychological Association.
Blatt, S. J., & Behrends, R. S. (1987). Internalization, separation-individuation, and the nature of
therapeutic action. International Journal of Psycho-Analysis, 68, 279-297.
Blatt, S. J., & Blass, R. B. (1990). Attachment and separateness. A dialectic model of the products and
processes of development throughout the life cycle. Psychoanalytic Study of the Child, 45,107-127.
Blatt, S. J., Cornell, C. E., & Eshkol, E. (1993). Personality style, differential vulnerability and clinical
course in immunological and cardiovascular disease. Clinical Psychology Review, 13, 421-450.
Blatt, S. J., & Felsen, I. (1993). Different kinds of folks may need different kinds of strokes: The effect of
patients' characteristics on therapeutic process and outcome. Psychotherapy Research, 3, 245-259.
Blatt, S. J., & Levy, K. N. (1998). A psychodynamic approach to the diagnosis of psychopathology. In J.
W. Baron (Ed.), Making diagnosis meaningful. Enhancing evaluation and treatment of psychological
disorders (pp. 73-109). Washington, DC: American Psychological Association.
Blatt, S. J., & Maroudas, C. (1992). Convergence among psychoanalytic and cognitive-behavioral theories
of depression. Psychoanalytic Psychology, 9, 157-190.
Blatt, S. J., Quinlan, D. M., Chevron, E. S., McDonald, C., & Zuroff, D. (1982). Dependency and self-
criticism: Psychological dimensions of depression. Journal of Consulting and Clinical Psychology, 50,
113-124.
Blatt, S. J., Schaffer, C. E., Bers, S. A., & Quinlan, D. M. (1992). Psychometric properties of the
Depressive Experiences Questionnaire for adolescents. Journal of Personality Assessment, 59, 82-98.
Blatt, S. J., Shahar, G., & ZurotT, D. C. (2002). AnacliticlSociotropic and Introjective/Autonomous
dimensions. In 1. C. Norcross (Ed.), Psychotherapy relationships that work. Therapist contributions
and responsiveness to patients (pp. 315-333). Oxford: Oxford University Press.
Blatt, S. J., & Shichman, S. (1983). Two primary configurations of psychopathology. Psychoanalysis and
Contemporary Thought, 6, 187-254.
Blatt, S. J., & Zuroff, D. C. (1992). Interpersonal relatedness and self-definition: Two prototypes for
depression. Clinical Psychology Review, 12,527-562.
Bleichmar, H. B. (1996). Some subtypes of depression and their implications for psychoanalytic treatment.
International Journal of Psycho-Analysis, 77, 935-961.
Bowlby, J. (1980). Attachment and loss Vol. 3 Loss: Sadness and depression. London: The Hogarth Press.
Bowlby, J. (1988). A secure base: Clinical applications of attachment theory. London: Routledge.
Chevron, E. S., Quinlan, D. M., & Blatt, S. J. (1978). Sex roles and gender differences in the experience
of depression. Journal ofAbnormal Psychology, 87, 680-683.
Chodorow, N. (1978). The reproduction of mothering. Psychoanalysis and the sociology of gender.
Berkeley: University of California Press.
Clark, D. A., & Beck, A. T. (1999). Scientific foundations of cognitive theory and therapy of depression.
New York: John Wiley & Sons.
Costello, E. J., Pine, D. S., Hammen, c., March, J. S., Plotsky, P. M., Weissman, M. M., Biederman, J.,
Goldsmith, H. H., Kaufman, J., Lewinsohn, P. M., Hellander, M., Hoagwood, K., Koretz, D. S., Nelson,
C. A., & Leckman, J. F. (2002). Development and natural history of mood disorders. Biological
Psychiatry, 52, 529-542.
Coyne, J. C; & Whiffen, V. E. (1995). Issues in personality as diathesis for depression: The case of
sociotropy-dependency and autonomy-self-criticism. Psychological Bulletin, 118,358-378.
Davidson, R. J., Pizzagalli, D., & Nitschke, J. B. (2002). The representation and regulation of emotion in
depression: Perspectives from affective neuroscience. In I. H. Gotlib & C. L. Hammen (Eds.),
Handbook of depression (pp. 219-244). New York/London: The Guilford Press.
Engel, G. L. (1962). Anxiety and depression-withdrawal: The primary affects of unpleasure. International
Journal of Psycho-Analysis, 43,89-97.
Flett, G. L., Hewitt, P. L., Endler, N. S., & Bagby, R. M. (1995). Conceptualization and assessment of
personality factors in depression. European Journal of Personality, 9, 309-350.
Freud, S. (1953). Studies on Hysteria. In 1. Strachey (Ed. & Trans!.), The standard edition of the complete
psychological works of Sigmund Freud (Vol, 2). London: Hogarth Press. (Original work published
1895).
Freud, S. (1957). Mourning and melancholia. In 1. Strachey (Ed. & Transl.), The standard edition of the
complete psychological works of Sigmund Freud (Vol, 14, pp. 243-258). London: Hogarth Press.

91
 (Original work published 1917).
Gillespie, R. D. (1929). The clinical differentiation of types of depression. Guy's Hospital Reports, 9, 306-
344.
Gilligan, C. (1982). In a different voice. Psychological theory and women's development.
CambridgelLondon: Harvard University Press.
Guisinger, S., & Blatt. S. J. (1994). Individuality and relatedness. American Psychologist. 49, 104-1 J I.
Gunderson, J. G., Triebwasser, 1., Philips, K. A., & Sullivan, C. N. (1999). Personality and vulnerability to
affective disorders. In C. R. Cloninger (Ed.), Personality and psychopathology (pp, 3-32). Washington.
DCILondon: American Psychiatric Press.
Gunnar. M. E. (1998). Quality of early care and buffering of neuroendocrine stress reactions: Potential
effects on the developing human brain. Preventive Medicine, 27, 208-211.
Hollon, S. D. (2003). Does cognitive therapy have an enduring effect? Cognitive Therapy and Research,
27,71-75.
Ingram, R. E., & Hamilton. N. A. (1999). Evaluating precision in the social psychological assessment of
depression: Methodological considerations. issues. and recommendations. Journal of Social and
Clinical Psychology. 18, 160-180.
Jones, E. E.. & Pulos, S. M. (1993). Comparing the process of psychodynamic and cognitive-behavioral
therapies. Journal of Consulting and Clinical Psychology, 61,306-316.
Judd, L. 1., Akiskal, H. S., Maser, 1. D., Zeller, P. J., Endicott. 1., Coryell, w., Paulus, M. P., Kunovac, J.
L., Leon, A. C; Mueller, T. I., Rice, J. A.. & Keller. M. B. (1998). A prospective 12-year study of
subsyndromal and syndromal depressive symptoms in unipolar major depressive disorders. Archives of
General Psychiatry, 55, 694-700.
Kandel, E. R. (1999). Biology and the future of psychoanalysis: A new intellectual framework for
psychiatry revisited. American Journal of Psychiatry, 156. 505-524.
Kendler, K. S.• & Gardner. C. O. (1998). Boundaries of major depression: An evaluation of DSM-IV
criteria. American Journal of Psychiatry, 155, 172-177.
Kernberg, O. F. (1975). Borderline conditions and pathological narcissism. New York: Jason Aronso i,
Kernberg, O. F. (1992). Aggression in personality disorders and perversions. New HavenlLondon: Yale
University Press.
Kessler, R. C. (1997). The effects of stressful life events on depression. Annual Review of Psychology, 48,
191-214.
Kiloh, L. G., & Garside, R. F. (1963). The independence of neurotic depression and endogenous
depression. British Journal of Psychiatry, 109,451-463.
Klein, D. N.. Durbin, E.. Shankman, S. A.. & Santiago. N. J. (2002). Depression and personality. In I. H.
Gotlib & C. L. Hammen (Eds.), Handbook of depression (pp, 115-140). New YorkILondon: The
Guilford Press.
Klein, D. N., & Hayden, E. P. (2000). Dysthymic disorder: Current status and future directions. Current
Opinion in Psychiatry, 13. 171-177.
Klein, M. H.• Kupfer, D. 1., & Shea, M. T. (1993). Personality and depression. A current view. New
YorkILondon: The Guilford Press.
Kohut, H., & Wolf, E. S. (978). The disorders of the self and their treatment: An outline. lnternauonal
Journal of Psycho-Analysis, 59, 413-425.
Kwon, P. ( 1999). Attributional style and psychodynamic defense mechanisms: Toward an integrative model
of depression. Journal of Personality. 67. 645-658.
Lax, R. F. (1989). The narcissistic investment in pathological character traits and the narcissistic
depression: Some implications for treatment. International Journal of Psycho-Analysis, 70. 81-90.
Luborsky, L., Diguer, L.. Luborsky, E., Singer, B., Dickter, D.. & Schmidt. K. A. (1993). The efficacy of
dynamic psychotherapies: Is it true that "Everyone has won and all must have prizes"? In N. E. Miller,
L. Luborsky, 1. P. Barber & J. P. Docherty (Eds.), Psychodynamic treatment research. A Handbook for
clinical practice (pp. 497-516). New York: Basic Books.
Luyten. P., Lowyck, B., & Corveleyn, J. (2003). Teoria y tratamiento de la depression: Haci.i su
integracion? [Theory and treatment of depression: Towards integration?]. Persona, 6, 81-97.
Markson, E. R. (1993). Depression and moral masochism. International Journal of Psycho-Analysis, 74,
931-940.
Messer, S. (1986). Behavioral and psychoanalytic perspectives at therapeutic choice points. American
Psychologist. 41, 1261-1272.
Miller, A. (1979). Depression and grandiosity as related forms of narcissistic disturbances. International
Review of Psycho-Analysis. 6. 61-76.

92
Milrod, D. (1988). A current view of the psychoanalytic theory of depression. With notes on the role of
identification. orality. and anxiety. The Psychoanalytic Study of the Child. 43. 83-99.
Milton. J. (2001). Psychoanalysis and cognitive behaviour therapy. Rival paradigms or common ground?
International Journal of Psychoanalysis, 82. 431-447.
Morrison. A. P. (1989). Shame. The underside of narcissism. Hillsdale. NJ: The Analytic Press.
Mulder. R. T. (2002). Personality pathology and treatment outcome in major depression: A review.
American Journal of Psychiatry. 159.359-371.
National Insiiture of Mental Health (2003). Breaking ground, breaking through: The Strategic Plan for
Mood Di.II.,,. 'rs Research. Retrieved September 3. 2004. from
http.z/www.nuu:•.nih.gov/strategic/mooddisorders.pdf
Power. M. (1991). Cognitive science and behavioural psychotherapy: Where behaviour was. there shall
cognition be? Behavioural Psychotherapy. 19. 20-41.
Raison. C. L.. & Miller. A. H. (2003). When not enough is too much: The role of insufficient glucocorticoid
signaling in the pathophysiology of stress-related disorders. American Journal of Psychiatry. /60.
1554-1565.
Robins. C. J. (1993). Implications of research in the psychopathology of depression for psychotherapy
integration. Journal of Psychotherapy Integration. 3. 313-330.
Robins. C. J. (1995). Personality-event interaction models of depression. European Journal of Personality.
9.367-378.
Rutter. M.• Dunn. J.• Plomin, R.• Simonoff, E.• Pickles. A.. Maughan. B.• Orrnel, J.• Meyer. 1.. & Eaves. L.
(1997). Integrating nature and nurture: Implications of person-environment correlations and
interactions for developmental psychopathology. Development and Psychopathology. 9. 335-364.
Ryle, A. (Ed.), (1995). Cognitive Analytic Therapy: Developments in theory and practice. New York:
Wiley.
Sabbe. B. G. C. (2002). Psychotherapie bij depressie: Integratie? IPsychotherapy for depression:
Integration?] In W. Trijsburg. S. Colijn, E. Collumbien & G. Lietaer (Eds.), Handboek voor integratieve
psychotherapie [Handbook of integrative psychotherapy] (VI. 5.1-32). Leusden, The Netherlands: De
Tijdstroom.
Safran. 1. D.• & Inck, T. A. (1995). Psychotherapy integration: Implications for the treatment of depression.
In E. E. Beckham & W. R. Leber (Eds.), Handbook of depression (2nd ed.) (pp. 425-434). New
York/London: The Guilford Press.
Sandler. J.• & Joffe. W. G. (1965). Notes on childhood depression. International Journal of Psycho-
Analysis. 46. 88-96.
Shahar, G.. Blatt. S. J.. & Ford. R. Q. (2003). Mixed anaclitic-introjective psychopathology in treatment-
resistant inpatients undergoing psychoanalytic psychotherapy. Psychoanalytic Psychology. 20. 84-102.
Shea. M. T.. Elkin. I.. Imber. S. D.• Sotsky, S. M .• Watkins. J. T.. Collins. 1. E. Pilkonis, P. A.• Beckham.
E.. Glass. D. R.• & Dolan. R. T. (1992). Course of depressive symptoms over follow-up. Findings from
the NIMH treatment of depression collaborative research program. Archives (if General Psychiatry. 49.
782-787.
Spitz. R. A. (1946). Anaclitic depression. The Psychoanalytic Study of the Child. 2. 313-342.
Tennant. C. (2002). Life events. stress and depression: A review of recent findings. Australian and Ne....
Zealand Journal of Psychiatry. 36. 173-182.
Tsigos, C .. & Chrousos, G. P. (2002). Hypothalamic-pituitary-adrenal axis. neuroendocrine factors and
stress. Journal of Psychosomatic Research. 53. 865-871.
Van Praag, H. M. (1998). Intlationary tendencies in judging the yield of depression research.
Neuropsvchobiology, 37. 130-141.
Vergote, A. (1976). Nevrose depressive [Depressive neurosis]. Topique, 17. 97-126.
Vredenburg. K.• Flett. G. L.. & Krames, L. (1993). Analogue versus clinical depression: A critical
reappraisal. Psychological Bulletin. 113. 327-344.
Wachtel. P. L. (1997). Psychoanalysis. behavior therapy. and the relational world. Washington. DC:
American Psychological Association Press.
Waddington. L. (2002). The therapy relationship in cognitive therapy: A review. Behavioural and Cognitive
Psychotherapy. 30. 179-191.
Warnpold, B. E. (1997). Methodological problems in identifying efficacious psychotherapies.
Psychotherapy Research. 7. 21-43.
Westen. D. (2000). Integrative psychotherapy: Integrating psychodynamic and cognitive-behavioral theory
and technique. In C. R. Snyder & R. E. Ingram (Eds.), Handbook of psychological change (pp. 217-
242). New York: Wiley.

93
Westen, D., & Morrison, K. (200 I). A multidimensional meta-analysis of treatments for depression, panic,
and generalized anxiety disorder: An empirical examination of the status of empirically supported
therapies. Journal of Consulting and Clinical Psychology, 69, 875-899.
Westen, D., & Shedler, J. (1999). Revising and assessing Axis II, Part I: Developing a clinically and
empirically valid assessment method. American Journal of Psychiatry, 156, 258-272.
Westen, D., Novotny, C. M., & Thompson-Brenner, H. (2004). The empirical status of empirically
supported psychotherapies: Assumptions, findings, and reporting in controlled clinical trials.
Psychological Bulletin, 130, 631-663.
Zuroff, D. C. (1992). New directions for cognitive models of depression. Psychological Inquiry, 3, 274-
277.
Zuroff, D. C; Mongrain, Moo & Santor, D. (2004). Conceptualizing and measuring personality vulnerab.lity
to depression: Comment on Coyne and Whiffen (1995). Psychological Bulletin, 130,489-511.

94
 Chapter 4

The Convergence among Psychodynamic

and Cognitive-Behavioral Theories of Depression:

A Critical Review of Empirical Research

Patrick Luyten, Jozef Corveleyn, & Sidney J. Blatt

The previous chapter presented a theoretical overview of the growing integration

among psychodynamic and cognitive-behavioral theories of depression. This
chapter reviews empirical research relevant to the theoretical formulations of
Sidney Blatt and Aaron Beck about the psychological dimensions of depression.
First, we discuss the assessment of Dependency/Sociotropy and Self-Critical
Perfectionism/Autonomy and then consider the three most central tenets of Blatt's
and Beck's theories: (1) the relationship between Dependency/Sociotropy and Self-
Critical Perfectionism/Autonomy and depression, (2) the relationship of these
dimensions to stressful life events, and (3) the characteristic interpersonal styles
and attitudes associated with these personality dimensions. We close this chapter
with a review of research on the clinical implications of these formulations and
some conclusions.
Assessment of Dependency/Sociotropy and Self-Critical Perfectionism/Autonomy

Despite a growing literature that employs other research methods, such as

experiments and interviews, most research on the psychological dimensions of
depression has relied on self-report instruments to assess dependent/sociotropic and
self-critical perfectonistic/autonomous personality dimensions. Four such
instruments are currently used to assess these dimensions: the Depressive
Experiences Questionnaire (DEQ), developed by Blatt and his colleagues (Blatt,
D' Afflitti, & Quinlan, 1976), I two instruments developed to assess Beck's
cognitive-behavioral views, namely the Sociotropy-Autonomy Scale (SAS; Beck,
Epstein, Harrison, & Emery, 1983) and the Dysfunctional Attitude Scale (DAS;
Cane, Olinger, Gotlib, & Kuiper, 1986), and finally the Personal Style Inventory
(PSI; Robins et aI., 1994), which was developed to integrate both Blatt's and Beck's
constructs.
Notwithstanding the fact that these instruments have produced interesting
results, theoretical and/or psychometrical problems have been identified with each
of these instruments (e.g., Blatt, 2004; Luyten, Fontaine, Soenens, Meganck et aI.,
2004). For instance, although the internal factor structure of the DEQ has been
replicated in several cultures in nonclinical samples, it has not always been fully
replicated in clinical samples (e.g., Bagby, Parker, Joffe, & Buis, 1994). The
subscales of the DAS are highly correlated (Blaney & Kutcher, 1991; Blatt, Shahar,
& Zuroff, 2002), suggesting that the two DAS subscales have substantial overlap.
One of the most important assessment problems centers on the assessment of
Beck's autonomy concept. It has repeatedly been shown that the SAS Autonomy
scale measures counterdependency rather than autonomy (Blaney & Kutcher, 1991;
Rude & Burnham, 1993). Attempts to improve the construct validity of the SAS
Autonomy subscale have only been partially successful (Bieling, Beck, & Brown,
2000; Clark & Beck, 1991; Clark, Steer, Beck, & Ross, 1995; Sato & McCann,
1997), and the PSI Autonomy subscale appears to suffer from similar problems as
the SAS Autonomy subscale (Bagby, Parker, Joffe, Schuller, & Gilchrist, 1998;
Hong & Lee, 2001; Kwon, Campbell, & Williams, 200 I; Sato & McCann, 1997). 2
In addition to these instruments, several other scales measure aspects of
Dependency/Sociotropy and Self-Critical Perfectionism/Autonomy. The most well-
known include the Interpersonal Dependency Inventory (IDI; Hirschfeld, et aI.,
1977) and two multidimensional measures of perfectionism, both entitled
Multidimensional Perfectionism Scale (MPS), one developed by Frost, Marten,
Lahart and Rosenblate (1990) (MPS-F), the other one (MPS-H) by Hewitt and Flett

I Blatt and his colleagues also developed a version of the DEQ for adolescents (DEQ-A; Blatt. Schaffer,

Bel'S, & Quinlan, 1992).

, Although somewhat imprecise, for didactical reasons in our discussion we will consider these four
instruments (DEQ, SAS, DAS, PSI) as equivalent measures of the constructs of Dependency!Sociotropy
and Self-Critical Perfectionism!Autonomy, except when results differ depending upon the instrument used.
For the same reason, we will in general not distinguish between Blatt's and Beck's constructs, unless this
is important for a correct interpretation of results.

96
(1991). Factor analytic studies suggest that two dimensions underlie these two
perfectionism scales, namely adaptive or "healthy" and maladaptive or "unhealthy"
perfectionism (Enns, Cox, & Clara, 2002). Similarly, recent research suggests that
interpersonal dependency also is multidimensional and has both adaptive and
maladaptive aspects (Blatt; 2004; Bornstein, 1992; Pincus & Gurtman, 1995).
Consistent with these findings, Blatt and his colleagues have recently distinguished
between adaptive and maladaptive forms of self-definition and interpersonal
relatedness. This chapter, however, concentrates mainly on research concerning
maladaptive aspects of self-definition and relatedness. In the next chapter, Blatt and
Shahar will focus on both adaptive and maladaptive aspects of these two
dimensions. In addition, our review is mainly limited to research with the DEQ,
SAS, DAS, and PSI for two reasons. First, we believe that it is vital to retain a clear
distinction between results from studies that have employed instruments that
explicitly assess Blatt's and Beck's views and instruments that were developed
based on other theoretical frameworks.' Second, specifically concerning the MPS-
F and MPS-H, we believe that further research is needed to investigate whether
these instruments indeed contain two underlying dimensions that can be interpreted
as adaptive versus maladaptive perfectionism.'

The Relationship Between Dependency/Sociotropy,

Self-Critical Perfectionism/Autonomy, and Depression

Cross-sectional Studies

Severity of depression. Over the past decades, cross-sectional studies in various

nonclinical populations (adolescents, students, adults, elderly) and in inpatient and
outpatient clinical populations have in general shown a consistent relationship
between Dependency/Sociotropy, Self-Critical Perfectionism/Autonomy, and self-
• For instance. recent research shows that the DEQ and the DAS Perfectionism subscales both capture and
thus confound Hewitt and Flett's socially prescribed and self-oriented perfectionism (Enns & Cox. 1999;
Hewitt & Flett, 1993; Hewitt, Flett, Besser, Sherry, & McGee, 2003), which makes it difficult to compare
results obtained with the Hewitt and Flett scales and results from scales that have been developed from
Blatt's and Beck's theories.
, In fact. several studies have found that "adaptive" perfectionism scales are related to measures of distress
and psychopathology (e.g., Bieling, Israeli, & Antoni, 2004; Enns, Cox, Sareen, & Freeman, 2(01) as well
as somatic complaints and Chronic Fatigue Syndrome (Luyten, Van Houdenhove, Cosyns, &
Vandenbroucke, 2004; Saboonchi & Lundh, 2(03). Thus, we believe that caution should be exercised in
calling these scales "adaptive", It appears that these so-called adaptive scales represent less pathological
forms of perfectionism. Second, although some forms or aspects of perfectionism (e.g., high personal
standards) might be adaptive temporarily (e.g., when one has to finish a project in time) or in some
circumstances (e.g., during an exam period), in the long term they might be quite detrimental for one's
psychical and physical health and/or for others (Blatt, 1995; Shafran & Mansell, 2(01).

97
report and interview-based measures of the severity of depression (for overviews,
see Blatt, 2004; Clark & Beck, 1999; Nietzel & Harris, 1990). A typical finding in
this respect is that Self-Critical Perfectionism shows a stronger relation to severity
of depression than Dependency/Sociotropy. Nietzel & Harris (1990), for instance,
in a meta-analysis of cross-sectional studies, found a mean effect size of r = .28
for Dependency/Sociotropy and r = .49 for Self-Critical Perfectionism/Autonomy
in their association with measures of depression. No differences were found,
however, in association with severity of depression between Autonomy and
Dependency/ Sociotropy in studies using the SAS. This finding, consistent with
findings from a number of other studies, again points to psychometric problems
with the SAS Autonomy scale (Clark & Beck, 1999). While the relationship
between measures of depression and Dependency/Sociotropy is typically smaller
than for Self-Critical Perfectionism/Autonomy, in some studies Dependency/
Sociotropy is even unrelated to severity of depression (e.g., Fichman, Koestner, &
Zuroff, 1996; Ouimette & Klein, 1993; Overholser & Freiheit, 1994; Santor &
Zuroff, 1997). One possible explanation for this finding is that traditional measures
of the severity of depression are more heavily weighted with symptoms typical for
self-criticism, leading to an underdiagnosis of anaclitic depression, which is more
concerned with loneliness, abandonment, and somatic complaints than with self-
criticism and guilt (Blatt, 1998; Blatt, Quinlan, Chevron, McDonald, & Zuroff,
1982).
Consistent with Blatt's and Beck's predictions, females generally have higher
levels of Dependency/Sociotropy than males, but gender differences are usually not
found on Self-Critical Perfectionism/Autonomy (Blatt, 2004). Some evidence
indicates that gender incongruence (high Dependency/Sociotropy in males and high
Self-Critical Perfectionism/Autonomy in females) is associated with a higher risk
for depression (Mongrain & Zuroff, 1994; Sanathara, Gardner, Prescott, & Kendler,
2003; Smith, O'Keeffe, & Jenkins, 1988) and a greater number of negative life
events (Mongrain & Zuroff, 1994; Fichman, Koestner, & Zuroff, 1994; Little &
Garber, 2000; Smith et aI., 1988).
Diagnosis of depression. Several cross-sectional studies have shown that both
Dependency/Sociotropy and Self-Critical Perfectionism/Autonomy are consistent-
ly associated with current major depression (e.g., Cox, McWilliams, Enns, &
Clara, 2004; Mazure, Bruce, Maciejewski, & Jacobs, 2000; Mazure &
Maciejewski, 2003; Mazure, Maciejewski, Jacobs, & Bruce, 2002; Nelson,
Hammen, Daley, Burge, & Davila, 2001; Raghavan, Le, & Berenbaum, 2002).
Also, case-control studies, comparing clinical and control groups, have consistently
found that depressed patients exhibit higher levels of both Dependency/Sociotropy
and Self-Critical Perfectionism/Autonomy than nonclinical groups of students and
adults (Bagby, Schuller, Parker, Levitt, Joffe, & Shafir, 1994; Bieling & Alden,
2001; Blatt et aI., 1982; Fairbrother & Moretti, 1998; Franche & Dobson, 1992;
Gudleski & Shean, 2000; Klein, Harding, Taylor, & Dickstein, 1988; Lehman et al.,
1997; Mazure, Bruce, Maciejewski, & Jacobs, 2000; Mazure, Maciejewski, Jacobs,
& Bruce, 2002; Mazure, Raghavan, Maciejewski, Jacobs, & Bruce, 2001; Moore &
Blackburn, 1996; Robins et al., 1994; Rosenfarb, Becker, Kahn, & Mintz, 1998;
Sahin, Ulusoy, & Sahin, 1993).5 In addition, Lehman et al. (1997) found that

98
depressed inpatients scored higher on both dimensions than depressed outpatients
(see also Morse, Robins, & Gittes-Fox, 2002). One study comparing depressed and
mixed psychiatric patients (Luyten, 2002) found that depressed inpatients exhibited
significantly higher scores on Dependency/Sociotropy than mixed psychiatric
inpatients, but no difference was found for Self-Critical Perfectionism/Autonomy,
suggesting that Self-Critical Perfectionism/Autonomy is a more general vul-
nerability factor than Dependency/Sociotropy.
Ouimette, Klein, Clark and Margolis (1992) found that only Self-Critical
Perfectionism/Autonomy was related to life time major depression as well as to
interview assessed depressive personality (see also Zuroff & Blatt, 2002). Other
studies have found that both Dependency/Sociotropy and Self-Critical Perfec-
tionism/Autonomy are associated with life time major depression (Cox et aI., 2004;
Nordahl & Stiles, 2000; Ouimette, Klein, & Pepper, 1996; Sakado et aI., 1999;
Sanathara et aI., 2003). Sanathara et al. (2003), for instance, reported that
Interpersonal Sensitivity, as assessed by the IDI (Hirschfeld et aI., 1977), which is
closely related to the construct of Dependency/Sociotropy, was strongly associated
with life time major depression as assessed by the Structured Clinical Interview for
DSM-III-R in the Virginia Twin Registry population-based sample of 7,174 twins.
Cox et al. (2004) reported that both personality dimensions were associated with
life time major depression as assessed by the Composite International Diagnostic
Interview in the National Comorbidity Survey (NCS), a study of a population based
representative sample (N=5,877).
Dependency/Sociotropy and Self-Critical Perfectionism/Autonomy have also
been related to dysthymic disorder (Goldberg, Segal, Vella, & Shaw, 1989;
Ouimette et al., 1996). Moreover, Klein, Taylor, Dickstein and Harding (1988)
reported that patients with early-onset Dysthymic Disorder had higher scores on
Self-Critical Perfectionism/Autonomy, but not Dependency/Sociotropy, compared
to patients with major depressive disorder.
Finally, some studies have investigated the stability of both personality
dimensions using cross-sectional designs by comparing currently depressed and
recovered patients. Results of these studies have been somewhat mixed (Fairbrother
& Moretti, 1998; Hartlage, Arduino, & Alloy, 1998; Klein, Harding et aI., 1988;
Rosenfarb et aI., 1998; Zuroff, Blatt, Sanislow, Bondi, & Pilkonis, 1999).
Recovered depressed patients, however, have consistently been found to exhibit
higher levels of both Dependency/Sociotropy and Self-Critical Perfectionism/Auto-
nomy than never depressed controls (Solomon, Haaga, Brody, Kirk, & Friedman,
1998; Solomon, Arnow, Gotlib, & Wind, 2003). Thus, while depression may
influence levels of Dependency/Sociotropy and Self-Critical Perfectionism/Auto-
nomy (state effect), recovered depressed patients still show higher levels of both
personality dimensions compared to normal controls, which is congruent with a

, It should be noted that four studies did not find a significant difference in levels of Autonomy between
depressed patients and normals controls (Mazure, Maciejewski. Jacobs. & Bruce. 2002; Moore &
Blackburn. 1996; Sahin. Ulusoy, & Sahin, 1993; Vogel. Stiles. & Nordahl. 2000). However. all these
studies employed the SAS Autonomy subscale which. as we noted. has serious psychometric limitations.
In fact. recent research indicates that the SAS Autonomy scale also contains both adaptive and maladaptive
dimensions (Blatt. 2004).

99
vulnerability model. Family studies have also investigated the vulnerability status
of Dependency/Sociotropy and Self-Critical Perfectionism!Autonomy because it is
well known that biological offspring of depressed patients are at high risk for
affective disorders. However, Ouimette et al. (1992) found no differences in levels
of both personality dimensions between a high risk group (offspring of patients
with affective disorders) and normal controls. While these findings may contradict
the vulnerability hypotheses, Self-Critical Perfectionism!Autonomy in the total
sample was associated with lifetime major depression and depressive personality.
In another study, Ouimette et al. (1996) found evidence for a scar effect in that
relatives of depressed patients with a history of affective disorders had higher levels
of both Dependency/Sociotropy and Self-Critical PerfectionismlAutonomy
compared to those relatives without a history of affective disorders.
Though cross-sectional studies clearly suggest that Dependency/Sociotropy and
Self-Critical Perfectionism!Autonomy are possible etiological or pathogenetic
factors associated with both severity of depression and depression diagnosis, these
studies are limited in the investigation of these dimensions as possible causal
antecedents or vulnerability factors in depression (Barnett & Gottlib, 1988). Simple
cross-sectional studies, for example, cannot rule out that these personality
dimensions are mere concomitants or consequences C'scar-hypothesis'', see
Chapter I, this volume) of depression. Likewise, cross-sectional case control
studies that compare remitted and currently depressed patients as well as family
studies of offspring of depressed patients are also limited in drawing causal
conclusions. Even though state effects may influence personality, this does not
eliminate the possibility of a vulnerability component. In fact, Blatt and Beck do
not imply absolute stability of Dependency/Sociotropy and Self-Critical
Perfectionism!Autonomy. Rather, their theoretical views imply both state and trait
effects, thus environmental stressors can intensify dependentlsociotropic or self-
critical/autonomous needs and conflicts (Zuroff, 1992). Conversely, pharmacolo-
gical or psychotherapeutic treatment may deactivate or modify these personality
dimensions. Moreover, Blatt's formulations (e.g., Blatt & Shichman, 1983) also
imply the possibility of regression as well as progression on each of these
dimensions. Hence, only longitudinal studies can definitely test the stability and
temporal antecedence of these personality dimensions.

Longitudinal Research

Stability. Evidence for the stability of both personality dimensions comes from
several studies that have shown good test-retest reliability of the DEQ, SAS, PSI.
and DAS in both clinical and nonclinical samples over periods ranging from thn:e
weeks to up to three years (Brewin & Firth-Cozens, 1997; Hammen, Marks.
deMayo, & Mayol, 1985; Hammen, Ellicott, & Gitllin, 1989; Kasch. Klein, & Lara.
200 I; Ouimette & Klein, 1993; Overholser & Freiheit, 1994; Segal, Shaw, Vella, &

100
Katz, 1992; Voyer & Cappeliez, 2002; Zuroff, Moskowitz, Wielgus, Powers, &
Franko, 1983; Zuroff, Igreja, & Mongrain, 1990; Zuroff et al., 1999). In addition,
Koestner, Zuroff and Powers (1991) reported considerable stability of self-criticism
as measured by a specially constructed scale in adolescent and young adult females,
but not in males. Self-criticism in males at age 12, however, predicted inhibition of
aggressive impulses at age 31.
Follow-up studies. Follow-up studies of depressed patients have shown that
levels of both Dependency/Sociotropy and Self-Critical Perfectionism!Autonomy
remain elevated in remitted depressed patients compared to normal controls, even
after brief pharmacological and/or psychotherapeutic treatment (Bagby, Schuller et
a\., 1994; Bagby et a\., 2001; Enns, Cox, & Pidlubny, 2002; Frank et al., 1997;
Kasch et aI., 200 I; Mazure et aI., 2000; Moore & Blackburn, 1996; Ouimette &
Klein, 1993; Segal et a\., 1992; Solomon et aI., 2003; Voyer & Cappeliez, 2002;
Zuroff et al., 1999). Moreover, Zuroff et al. (1999) showed in a re-analysis of data
from the National Institute of Mental Health Treatment of Depression Collaborative
Research Program (TDCRP) that changes over time in mean scores that are found
in some studies (e.g., Enns, Cox, & Pidlubny, 2002; Moore & Blackburn, 1996;
Ouimette et al., 1996; Rector, Bagby, Segal, Joffe, & Levitt, 2000; Rosenfarb et
al., 1998) do not necessarily imply that Dependency/Sociotropy and Self-Criti-
cal Perfectionism/Autonomy are mere concomitants of depression. Using
Structural Equation Modeling (SEM), these investigators found that a model that
included both state and trait effects better fitted the data than a pure state or pure
trait model. Both during, but particularly after treatment, both personality
dimensions showed remarkable stability. These findings have been replicated since
in two other studies (Cox & Enns, 2003; Shahar, Blatt, Zuroff, Kuperminc, &
Leadbeater, 2004). Consistent with formulations by Blatt and Beck, these findings
suggest that the measurement of Dependency/Sociotropy and Self-Critical
Perfectionism!Autonomy in both clinical and non-clinical samples includes both
state and trait components, with trait components accounting for most of the
variance.
Prospective studies. While some studies found that neither Dependency/Socio-
tropy nor Self-Critical Perfectionism!Autonomy predict severity of depression over
time (Mazure et aI., 2000; Overholser & Freiheit, 1994; Voyer & Cappeliez, 2002),
most studies found that both personality dimensions predict severity of depression
over time (Brewin & Firth-Cozens, 1997; Fresco, Sampson, Craighead, & Koons,
200 I; Lakey & Ross, 1994; Mongrain, Lubbers, & Struthers, 2004; Mongrain &
Zuroff, 1994; Priel & Shahar, 2000; Robins, Hayes, Block, Kramer, & Villena,
1995; Shahar, Joiner, Zuroff, & Blatt, 2004; Shahar & Priel, 2003; Zuroff, Stotland,
Sweetman, Craig, & Koestner, 1995; Zuroff et al., 1999). Two other studies
reported that only Self-Critical Perfectionism!Autonomy (Shahar, Blatt, Zuroff,
Kuperminc, & Leadbeater, 2004) or Dependency/Sociotropy (Alford & Gerrity,
1995) predicted depression severity. One likely explanation for these mixed
findings is that all negative studies (with the exception of Shahar, Blatt et aI., 2004)
are based on very small samples (N's ranging between 41-75 in negative studies and
between 48-603 in positive studies), most probably resulting in inadequate
statistical power to detect significant effects (Cohen, 1988). Of particular

101
importance is a study by Brewin and Firth-Cozens (1997), which reported that both
Dependency/Sociotropy and Self-Critical Perfectionism/Autonomy predicted
severity of depression, especially in males, after 2 and 10 years, even after
controlling for initial symptoms and workload in a sample of medical students. To
address the issue of scar effects, Shahar, Blatt, Zuroff, Kuperminc and Leadbeater
(2004), investigating a pure vulnerability, a pure scar, and a reciprocal model
(including both vulnerability and scar effects) of the relationship between
Dependency/Sociotropy and Self-Critical Perfectionism!Autonomy and severity of
depression in a one-year longitudinal study of 452 adolescents, found a reciprocal
relationship between Self-Critical Perfectionism/Autonomy, but not Dependency/
Sociotropy, and depressive symptoms. However, the scar effect was small
compared to the vulnerability effect and was only found in girls. In addition, pure
vulnerability and pure scar models provided a significantly poorer fit to the data
than a reciprocal model.
Both Dependency/Sociotropy and Self-Critical Perfectionism/Autonomy have
also been shown to predict several variables related to the course of depression,
including severity of depression and worse global functioning at a 6-month follow-
up (Klein, 1989), increased time to recovery (Kasch et aI., 200 I), non-recovery
(Bothwell & Scott, 1997), and relapse (Lam, Green, Power, & Checkley, 1996;
Wilhelm, Boyce, & Brownhill, 2004).
Longitudinal studies of the relationship between Dependency/Sociotropy and
Self-Critical Perfectionism/Autonomy and the onset ofdepression are scant, mainly
because of their high cost and effort. Wilhelm et al. (2004) found that Separation
Anxiety, a core component of Dependency/Sociotropy, predicted repeated episodes
of major depression in a 5-year follow-up study of 156 community participants.
Spasojevic and Alloy (2001) found that both Dependency/Sociotropy and Self-
Critical Perfectionism/Autonomy predicted the number of future episodes of major
depression in 137 students who were followed for 2.5 years. In a 6-year prospective
study in a sample of first-degree relatives of patients with affective disorders,
Hirschfeld et a1. (1989) reported that Dependency/Sociotropy, as measured by the
IDI, predicted first onset of depression among older subjects (aged 31-41), but not
among younger (aged 17-30). The study by Sanathara et al. (2003), discussed
earlier, evaluated vulnerability, state, and scar models of the relationship between
Dependency/Sociotropy as assessed by the IDI, and major depression as measured
by the SCID, in a one-year prospective study in a community-based sample of
7,174 twins. Sanathara et al. (2003) reported, congruent with a vulnerability model,
that pre-morbid Dependency/Sociotropy predicted future onset of major depression
among those that had no prior depressive episodes (N= I,328). Congruent with the
state model, they also found that levels of Dependency/Sociotropy were influenced
by current depression. However, in contrast with the scar model, Dependency/
Sociotropy scores were not significantly elevated by a history of depression.
Interestingly, they also reported that Dependency/Sociotropy scores showed a trend
towards elevation during and after experiencing a major depressive episode. Fitting
a regression line to the change observed in Dependency/Sociotropy scores as a
function of time, they showed that Dependency/Sociotropy scores would retum to
baseline in approximately 6 months after a depressive episode.

102
Conclusions

Existing research clearly suggests that Dependency/Sociotropy and Self-Critical

Perfectionism/Autonomy are relatively stable factors influencing the onset and
course of depression as well as the severity of depression. In particular, the
relationship between these personality dimensions and depression appears to be
best described by a trait/state/scar model, with trait effects accounting for most
variance. Although state and "scar" effects are not negligible, they appear to be
small. Moreover, "scar" effects seem to "wear off' over time.
However, more well designed prospective studies of the relationship between
these personality dimensions and the onset and course of several forms of clinical
depression are needed to further validate these conclusions. For instance, whereas
both personality dimensions might predict onset of major depression, Depen-
dency/Sociotropy might actually be a protective factor against post-partum
depression (Franche, 200 I; Franche & Mikail, 1999; Priel & Besser, 1999, 2000;
however, see Boyce, Parker, Barnett, Cooney, & Smith, 1991). Future studies
should also consider the possibility of age and cohort effects because the meaning
and thus effect of Dependency/Sociotropy and Self-Critical Perfectionism/Autono-
my might change over time and with age. For instance, Hirschfeld et al. (1989)
reported that Dependency/Sociotropy predicted first onset of depression in subjects
aged 31-41, but not in young adults. Mazure and Maciejewski (2003) in tum found
in a sample of 170 subjects aged 20-91 years that the risk for clinical depression
was quite uniform across age for Dependency/Sociotropy, whereas the effect for
Self-Critical Perfectionism/Autonomy was strongest in younger adulthood and
declined with age.
Finally, it should be pointed out that the studies reviewed in this section only
provide a partial test of Blatt's and Beck's hypothesis that Dependency/Sociotropy
and Self-Critical Perfectionism/Autonomy are antecedent vulnerability factors of
depression because they only consider the diathesis (i.e., personality) component.
However, as noted in the previous chapter, Blatt's and Beck's views imply in fact a
diathesis-stress model in which these personality dimensions confer a vulnerability
(diathesis) to depression either alone or in interaction with (congruent) stressors
(stress). In the next section, we review studies concerning this congruency
hypothesis.

Proximal Precipitating Factors: The Congruency Hypothesis

The relationship between (severe) negative life events and the onset and course
of depression is well documented (Kessler, 1997; Tennant, 2002). Blatt as well as
Beck extend traditional theories of the relationship between life events and
depression in two ways. First, a key assumption of Blatt's and Beck's views is that
Dependency/Sociotropy and Self-Critical Perfectionism/Autonomy are expected to

103
interact with congruent or matching events to predict depression (e.g., interpersonal
events vs. achievement events) (Blatt & Zuroff, 1992). This "key-and-lock"
hypothesis has also been called the congruency hypothesis (Robins, 1995). A
second extension of research on the relationship between life events and depression
involves the view that personality, depression, and life stressors are :10t
independent, but dynamically interact with each other. This view implies that
dependent/sociotropic and self-critical/autonomous individuals, in part, generate
their own (congruent) life stressors. Thus, dependent/sociotropic subjects are
expected to generate interpersonal stress events (e.g., dissolution of relationships,
family quarrels), while self-critical/autonomous subjects are expected to cause
more achievement related events (e.g., failure to obtain ajob promotion). Moreover,
(congruent) life stressors may increase levels of Dependency/Sociotropy and Self-
Critical Perfectionism/Autonomy (Zuroff, 1992). For instance, dependent
individuals' fear of rejection might be increased by frequently experiencing
rejection by significant others, whereas the fear of failure and disapproval of self-
critical individuals increases their vulnerability to experiencing failure. FinaJly,
depression may influence both personality ("scar-effect") as well as life events (see
also Hammen, 1991).
Unfortunately, most research concerning the congruency hypothesis has not
investigated this dynamic interaction ism perspective, but has rather adopted a
mechanistic interaction ism view (Zuroff, 1992). That is, most studies have treated
personality and environment as distinct, static, independent entities. Life events are
thus often conceptualized as moderators (Baron & Kenny, 1986), whereas Blatt's
and Beck's views imply in fact that life events act as mediators in the relationship
between personality and depression. First, we review studies that have either
implicitly or explicitly adopted a mechanistic view of Dependency/Sociotropy and
Self-Critical Perfectionism/Autonomy and their interaction with life events. Next,
research concerning the dynamic interactionism hypothesis is examined.
Mechanistic interactionism. More than 40 empirical studies have investigated
the "mechanistic" congruency hypothesis. A considerable number of these studies
found evidence for the congruency hypothesis for both personality dimensions
(Blaney, 2002; Brown, Hammen, Craske, & Wickens, 1995; Dunkley, Zuroff,
& Blankstein, 2003; Fichman, Koestner, & Zuroff, 1997; Giordana, Wood, &
Michela, 2000; Gruen, Silva, Ehrlich, Schweitzer, & Friedhoff, 1997; Hammen &
Goodman-Brown, 1990; Hammen, Marks, Mayol, & deMayo, 1985; Hammen,
Ellicott & Gitlin, 1989; Hammen, Ellicott, Gitlin, & Jamison, 1989; Lam et aI.,
1996; Mazure & Maciejewski, 2003; Mazure et aI., 2002; Mongrain & Zuroff,
1994; Robins, 1990, Study 2; Segal et aI., 1992), other studies only found evidence
for Dependency/Sociotropy (Allen, Home, & Trinder, 1996; Bartelstone & Trull,
1995; Clark, Beck, & Brown, 1992; Dowis & Backs-Dermott, 2000; Ewart,
Jorgensen, & Kolodner, 1998; Hammen, Ellicott, & Gitlin, 1992; Lakey & Ross,
1994; Priel & Shahar, 2000; Raghavan et aI., 2002; Robins, 1990, Study I; Robins
& Block, 1988; Rude & Burnham, 1993; Segal, Shaw, & Vella, 1989; Shahar,
Joiner, Zuroff, & Blatt, 2004; Solomon et al., 1998; Voyer & Cappeliez, 2002:
Whiffen & Aube, 1999; Zuroff & Mongrain, 1987), while in one study there was
only evidence for a specific interaction between Self-Critical Perfectionism!Auto-

104
nomy and congruent life events (Clark & Oates, 1995). In addition, some studies
have found no evidence for the congruency hypothesis for neither Dependency/
Sociotropy nor Self-Critical Perfectionism/Autonomy (Burgess, Lorah, Haaga, &
Chrousos, 1996; Flett, Hewitt, Garshowitz, & Martin, 1997; Fresco et al., 2001;
Kwon & Whisman, 1998; Mazure et al., 2000; Robins et al., 1995; Santor & Zuroff,
1997; Santor, Pringle, & Israeli, 2000; Shahar & Priel, 2003; Smith et al., 1988;
Zuroff et al., 1990). 6
Studies concerning the interaction between Dependency/Sociotropy and Self-
Critical Perfectionism/Autonomy and specific types of daily hassles have yielded
more mixed results, with some studies finding support for the congruency
hypothesis (Sherry, Hewitt, Flett, & Harvey, 2003; Dunkley, Zuroff, & Blankstein,
2003), others only in depressed patients, but not in mixed psychiatric patients
(Hewitt & Flett, 1993), and still others finding none or only limited evidence for a
congruency effect (Clark & Oates, 1995; Santor & Patterson, 2003).
In sum, support for the mechanistic congruency hypothesis has been fairly
consistent for Dependency/Sociotropy, but more mixed for Self-Critical
Perfectionism/Autonomy (Blatt, 2004; Coyne & Whiffen, 1995; Nietzel & Harris,
1990; Robins, 1995), and is mainly limited to negative life events, but not daily
hassles. It is noteworthy, however, that the evidence for a congruency effect for both
dimensions is most clear when one considers the qualitatively best studies (i.e.,
using a prospective design, interview-based measures of life events and clinical
depression and/or standardized measures of the severity of depression). For
example, Hammen, Ellicott, Gitlin and Jamison (1989) found that both Depen-
dency/Sociotropy and Self-Critical Perfectionism/Autonomy in interaction with
interview-assessed congruent life events predicted both the onset of major
depressive episodes and the severity of depressive symptoms in a sample of
unipolar depressed patients during a 6 month follow-up.
Studies on the congruency hypothesis also lend further support for the status of
Dependency/Sociotropy and Self-Critical Perfectionism/Autonomy as vulnerability
factors for depression. Not only do both personality dimensions directly and
independently predict depression, but most studies have found that the interaction
between these two personality dimensions and life events is not only statistically
significant, but also accounts for a substantial, and thus clinically relevant, amount
of additional variance in predicting depression. Mazure et al. (2002), for instance,
reported that whereas a clinical diagnosis of major depression was 0.5 to 3 times
more likely when subjects had high levels of Dependency/Sociotropy and Self-
Critical Perfectionism/Autonomy, the interaction between these personality
dimensions and congruent life events made clinical depression 6 to II times more
likely (for similar results, see Raghavan et al., 2002).

"Only two studies investigated the congruency hypothesis in bipolar patients. In a 6-month follow-up study
of bipolar patients, Hammen, Ellicott, Gitlin and Jamison (1989) did not find evidence for neither
Dependency/Sociotropy nor Self-Critical Perfectionism/Autonomy, while Hammen, Ellicott and Gitlin
(1992) reported that the interaction between Dependency/Sociotropy and interpersonal events predicted
symptom severity, but not onset of symptoms, in an 18-month follow-up study of bipolar patients. For Self-
Critical Perfectionism/Autonomy, no evidence for the congruency hypothesis was found.

105
 Dynamic interactionism. Early studies of the effect of Dependency/Sociotropy
and Self-Critical Perfectionism/Autonomy on the occurrence of life events have
yielded remarkably inconsistent results, most likely because of small samples sizes
(Connor-Smith & Compas, 2002; Hammen, Marks, deMayo, & Mayol, 1985;
Mongrain & Zuroff, 1989, 1994; Robins & Block, 1988). In what is probably the
most sophisticated study of the dynamic interactionism theory to date, Simons,
Angell, Monroe and Thase (1993) investigated the influence of Dependency/
Sociotropy and Self-Critical Perfectionism/Autonomy on the definition, rating, and
generation of life events as measured by both self-report and interview in a sample
of 55 depressed patients. Self-Critical Perfectionism/Autonomy predicted the total
number as well as severity of self-reported achievement, but not interpersonal,
events. Dependency/Sociotropy was unrelated to the total number and severity of
self-reported interpersonal as well as achievement events. Thus, Self-Critical
Perfectionism, but not Dependency/Sociotropy, was related to a tendency to
overreport and overrate the importance of congruent, but not incongruent, life
events (see also Gruen et aI., 1997). Moreover, Simons et al. (1993) found that both
Dependency/Sociotropy and Self-Critical Perfectionism/Autonomy predicted the
generation of fateful (i.e., self-generated) congruent, but not incongruent, life
events, particularly in patients with no prior history of depression. Importantly, both
personality dimensions were unrelated to independent (not self-generated) events.
Daley and colleagues (1997) reported in a 2-year follow-up study of 134 late
adolescent women that Self-Critical Perfectionism/Autonomy, and to a lesser extent
Dependency/Sociotropy, predicted fateful, but not independent, life events.
Interestingly, both Dependency/Sociotropy and Self-Critical Perfectionism/Auto-
nomy predicted interpersonal conflict stress (see also Flett et al., 19(7).
Unfortunately, Daley and colleagues (1997) did not investigate whether these
personality dimensions also predicted achievement stress. In addition, in line with
the dynamic interactionism model, Daley et al. (1997) also found that depression
predicted fateful, but not independent, events. However, these fateful events
occurred mainly outside depressive episodes. Similarly, Lakey and Ross (1994)
reported that dysphoria predicted severity of depression and both interpersonal and
achievement life events in a sample of 133 students. However, controlled for initial
levels of dysphoria, Dependency/Sociotropy still interacted only with congruent
events, whereas Self-Critical Perfectionism/Autonomy interacted with both
congruent and incongruent events. In line with these findings, several studies have
reported that levels of distress do not predict subsequent life events when controlled
for Dependency/Sociotropy and Self-Critical Perfectionism/Autonomy (Priel &
Shahar, 2000; Shahar & Priel, 2003; Shahar, Blatt, Zuroff, Krupnick, & Sotsky,
2004). These results are consistent with the notion that the generation of stress
mostly occurs outside depressive episodes, and that relatively stable personality
features, such as Dependency/Sociotropy and Self-Critical Perfectionism/Autono-
my, rather than depressed mood, are responsible for stress generation (see also
Daley et al., 1997).
Recent studies by Shahar and colleagues suggest important differences between
Dependency/Sociotropy and Self-Critical Perfectionism/Autonomy and respon-
sivity to and generation of stress. Priel & Shahar (2000) reported that only Self-

106
Critical Perfectionism!Autonomy, but not Dependency/Sociotropy, was related to
life stress in a 9-week prospective study of 182 young adults. In line with these
results, they reported that a moderating model best described the relationship
between Dependency/Sociotropy, negative life events, and dysphoria, whereas a
mediating model represented the best fit to the data for the relationship between
Self-Critical Perfectionism/Autonomy, negative life events, and dysphoria.
Moreover, Dependency/Sociotropy was associated with higher levels of perceived
social support, which in turned was associated with lower levels of distress. Self-
Critical Perfectionism!Autonomy on the contrary predicted decreased social
support over time, which in turn predicted elevated distress.
Hence, self-critical, but not dependent, individuals seemed to generate their own
negative life events and further "degenerate" levels of social support. These results
were replicated in a 5-week longitudinal study of 198 university students (Shahar,
Joiner, Zuroff, & Blatt, 2004). Moreover, whereas Dependency/Sociotropy
interacted with congruent stressors related to family and friends, Self-Critical
Perfectionism!Autonomy predicted the generation of both congruent (e.g., friends-
related) and incongruent stress (e.g., school stress). In a 16-week longitudinal study,
Shahar and Priel (2003) found that both personality dimensions predicted the
generation of negative life events. One possible explanation of these divergent
findings for the effect of Dependency/Sociotropy is that the stress generation effect
of Dependency/Sociotropy may be smaller, and thus is perhaps harder to detect.
Interestingly, Shahar and Priel (2003) also reported that Dependency/Sociotropy
predicted the experience of positive events over time, which suppressed the
negative effect of Dependency/Sociotropy, whereas Self-Critical Perfectionism!
Autonomy. in contrast, was negatively associated with positive events, which in
turn predicted increased levels of distress. Hence, while both Dependency/
Sociotropy and Self-Critical Perfectionism/Autonomy were associated with the
generation of negative life events, Dependency/Sociotropy also increased the
number of positive experiences, whereas Self-Critical Perfectionism!Autonomy
decreased the number of positive events.
In summary, the vulnerability associated with Dependency/Sociotropy appears
to be somewhat more passive-reactive. The fact that Dependency/Sociotropy is, at
best, only weakly related with the generation of life stress could be explained by the
tendency of dependent/sociotropic individuals to avoid conflicts, especially in
interpersonal relationships, and to generate, at least temporarily. social support
(Blatt & Zuroff, 1992). Alternatively, because of this tendency to avoid conflicts,
they may underreport interpersonal conflicts. Evidence is more clear in suggesting
that their vulnerability for life stress is specific, in that dependent/sociotropic
individuals show increased vulnerability for depression in reaction to particular
classes (i.e., congruent) of stressors. Moreover, Dependency/Sociotropy is also
associated with the generation of protective factors such as positive life events and
perceived social support. Thus, Dependency/Sociotropy appears to include aspects
of vulnerability as well as resilience (Blatt, 2004; Bornstein, 1992; Shahar, Joiner
et aI., 2004).7 This conclusion is further substantiated by research showing that

7This might also explain its weaker relationship with distress compared to Self-Critical Perfectionism
(Shahar & Priel, 2003).

107
faced with stress, dependent/sociotropic individuals tend to use both adaptive (i.e.,
active, approach) (Besser & Priel, 2003; Fichman, Koestner, Zuroff, & Gordon,
1999) as well as maladaptive coping styles such as venting or consumption-based
self-indulgence (Fichman et aI., 1999, Haaga, Fine, Terrill, Stewart, & Beck, 1995).
Vulnerability to life stress associated with Self-Critical Perfectionism/Auto-
nomy, in contrast, appears to be more proactive but less specific. Self-criticall
autonomous individuals seem to generate a wide range of stressors. In addition,
unlike dependent/sociotropic individuals, they fail to generate positive, protective
environmental factors such as positive events or social support. Moreover, they also
tend to consistently use maladaptive (e.g., emotional, passive) coping strategies
(Chang, 2000; Chang & Rand, 2000; Connor-Smith & Compas, 2002; Besser &
Priel, 2003; Dunkley & Blankstein, 2000; Dunkley, Blankstein, Halsall, Williams,
& Winkworth, 2000; Haring, Hewitt, & Flett, 2003; O'Connor & O'Connor, 2003).
Faced with stress, they tend to withdraw and isolate themselves (Fichman et al.,
1999). In alllikelihood, this gives rise to a self-perpetuating cycle characterized by
rumination, self-criticism, further withdrawal, and thus fewer opportunities to
receive social support and to have positive experiences (Daley et al., 1997; Kasch
et aI., 2001). Finally, results from studies on the dynamic interactionism hypothesis
may explain why evidence for the "mechanistic" congruency hypothesis is .ess
equivocal for Dependency/Sociotropy than for Self-Critical Perfectionism/Auto-
nomy. First, a moderating model, which is implied by a mechanistic view, seems to
best describe the relationship between Dependency/Sociotropy and life events,
whereas a mediating model best describes this relationship for Self-Critical
Perfectionism/Autonomy. Moreover, Self-Critical-Perfectionism/Autonomy ap-
pears to be associated with a broader vulnerability to life stress than Dependen-
cy/Sociotropy.
Limitations of existing research. The above conclusions await further research
because of several methodological and theoretical caveats, especially in studies that
have adopted a mechanistic interpretation of the congruency hypothesis (see also
Coyne & Whiffen, 1995). From a methodological point of view, findings
concerning Self-Critical Perfectionism/Autonomy may be influenced by the fact
that several studies of the congruency hypothesis have employed the SAS (Beck et
aI., 1983). Because the SAS Autonomy scale is often unrelated to depression, it
should come as no surprise that some studies have found no interaction between
SAS Autonomy and (congruent) life events (Bieling et aI., 2000). Second, research
concerning the congruency hypothesis should be better informed by current life
stress research. For instance, most studies have relied on subjective, questionnaire-
based measures of life stress rather than objective, interview-based measures of
stress. Yet, studies have consistently shown that questionnaire-based measures of
life stress tend to overestimate both the frequency and severity of life stress
compared with interview-based measures (Monroe & Hadjiyannakis, 2002),
possibly leading to an inflation of life events and their impact. Furthermore,
interview-based assessment of life events also allows distinguishing between
fateful and independent life events, which is crucial in investigating the dynamic
interactionism hypothesis. In addition, many studies have typically employed small
samples. In combination with the low incidence of depression and the fact that the

108
ability to detect (personality-event) interactions typically requires a larger number
of subjects than does the detection of main effects (McClelland & Judd, 1993),
many studies in this domain suffer from inadequate statistical power. Also, often
very short time frames (i.e., often only a few days or weeks) have been used in
longitudinal studies of the congruency hypothesis. However, it is well known that
the average person will experience very few severe negative life events in such short
time spans (Tennant, 2002). Finally, it is well documented that the role of life
stressors in the onset of depression decreases as the number of previous episodes
increases, such that depression tends to run a more autonomous course, whereas
those without a history of depression may be more responsive to stressors (see also
Chapter I and Chapter 8 in this volume). The number of previous depressive
episodes, however, is rarely taken into account in research concerning the
congruency hypothesis. Hence, future studies should differentiate between
pathways toward initial and subsequent episodes of depression.
Many existing studies in this domain also show important theoretical limitations.
To begin with, the categorization of life events as either relevant for Depen-
dency/Sociotropy or Self-Critical Perfectionism!Autonomy is not only difficult, but
also to a certain extent arbitrary because dependent/sociotropic and self-
critical!autonomous individuals may interpret the same life event in different ways
(Blatt & Zuroff, 1992; Coyne & Whiffen, 1995). A dimensional approach, in which
individuals rate life events in terms of the degree of relevance for Dependency/
Sociotropy and Self-Critical Perfectionism/Autonomy is needed (Hammen,
Ellicott, & Gitlin, 1989). In addition, a more idiographic assessment, which takes
into account the meaning of a life stressor, rather than using general classes of life
events (e.g., interpersonal versus achievement), may provide a better test of the
congruency hypothesis. Assessment methods such as in vivo thought sampling or
priming techniques are likely to yield more ecologically valid tests of the
congruency hypothesis (Dozois & Backs-Dermott, 2000). Second, research
suggests that most important negative life events are in the interpersonal domain,
which could make it very hard to find a specific interaction between Self-Critical
Perfectionism!Autonomy and achievement events (Coyne & Whiffen, 1995;
Hammen, 1991; Zuroff et al., 1987). Third, some authors have argued that depres-
sion in self-critical/autonomous subjects may not require environmental activation,
but primarily results from internal factors, such as continuous self-criticism and/or
biological factors (Mongrain & Zuroff, 1994). In this context, it is important to note
that Beck (1983) links his autonomous type of depression to endogenous
depression. Fourth, very little research has explored cohort and age differences in
the relationship between Dependency/Sociotropy and Self-Critical Perfectionism/
Autonomy and life stress. In the only study addressing this issue, Mazure and
Maciejewski (2003) reported that the impact of the interaction between Self-
Critical Perfectionism/Autonomy and achievement events as a risk factor
for clinical depression was strongest in young adulthood (increasing the likelihood
of clinical depression more than 20 (!) times at age 25) and declined with
age, whereas the reverse was found for the interaction between Dependency/
Sociotropy and interpersonal events. Hence, these results suggest important
age (and perhaps also gender and cohort) differences in that self-critical!

109
autonomous individuals are particularly at risk for depression in young adulthood
when confronted with congruent stressors, whereas with age the likelihood of
depression increases for dependentJsociotropic individuals if they experience
congruent stressors.
Future research should focus more on the dynamic processes involved in the
relationship between personality and life stress. As Hewitt and Flett (2002) have
argued, personality dimensions such as Dependency/Sociotropy and Self-Critical
Perfectionism/Autonomy are likely to influence not only the perception and
generation of stress, but also the anticipation, enhancement, and perpetuation of
stress. In addition, as research on gene- and person-environment correlations and
interactions in other domains has shown (Rutter et al., 1997), a dynamic, reciprocal
model of vulnerability to depression might also play an important role in the
integration of biological and psychosocial factors in depression. Already, several
studies have shown that not only subjective, but psychophysiological and
biochemical stress as well might be contingent on the "fit" between personality and
the nature of the stressor (Allen et aI., 1996; Ewart et aI., 1998; Gruen et al., 1997;
Sauro et al., 2001). Gruen et al. (1997), for instance, found that Self-Critical
Perfectionism/Autonomy, but not Dependency/Sociotropy, was associated with
changes in plasma homovanillic acid (HVA), the primary dopamine metabolite in
humans, during exposure to an induced failure-stressor. Hence, studies concerning
the congruency hypothesis might play an important role in bridging the gap
between psychological and biological approaches in research and treatment of
depression.

Characteristic Interpersonal Style

Empirical research, especially on couples, has provided strong support for

Blatt's and Beck's descriptions of the characteristic interpersonal style associated
with Dependency/Sociotropy and Self-Critical Perfectionism/Autonomy. In
addition, these studies also provide further support for the dynamic interactionisrn
model proposed by Blatt and Zuroff (1992). Based on Buss (1987), Blatt and Zuroff
(1992) proposed that dependent/sociotropic and self-critical/autonomous
individuals select, evoke, and transform (manipulate) their social environment.
Selection refers to the ways individuals choose different environments, evocation to
the fact that individuals tend to elicit different environments, and transformation to
the tendency of individuals to transform (or manipulate) their environment. We
review evidence for each of these interactive processes in tum.
Selection. Research findings clearly indicate that dependent/sociotropic and seif-
critical/autonomous individuals prefer different (social) environments. DependentJ
sociotropic individuals are fundamentally oriented towards others in order to
receive love and support, whereas self-critical/autonomous individuals are ambi-
valent and distant towards others and mainly value achievement and success (Blatt

110
& Zuroff, 1992; Zuroff & DeLorimier, 1989; Zuroff et aI., 1995). Studies have
consistently shown that Self-Critical Perfectionism/Autonomy is associated with a
cold, distant relational style that is characterized by manifest hostility in
relationships, even in intimate relationships such as with romantic partners (Alden
& Bieling, 1996; Fichman et aI., 1994; Vettese & Mongrain, 2000; Whisman &
Friedman 1998; Zuroff & Duncan, 1999). In addition, Self-Critical Perfec-
tionism/Autonomy is associated with dissatisfaction in relationships in general,
including those with partner and children (Dimitrovsky, Levy-Schiff, & Schattner-
Zanany, 2002; Haring et aI., 2003; Lynch, Robins, & Morse, 2001; Zuroff &
Duncan, 1999; Vettese & Mongrain, 2000; Whiffen, Aube, Thompson, & Campbell,
2000; Zuroff, Koestner, & Powers, 1994), and with dissatisfaction concerning
sexuality in particular (Morrison, Urquiza, & Goodlin-Jones, 1998). Self-Critical
Perfectionism/Autonomy has also been associated with feelings of social isolation
and loneliness (Besser, Flett, & Davis, 2003; Clark et aI., 1995; Schachter &
Zlotogorski, 1995; Wiseman, 1997). For self-critical/autonomous individuals, their
worth often appears to be equated with their work. In line with theoretical
expectations, they also tend to evaluate others negatively (Shapiro, 1988a, 1988b),
show low levels of social acuity (Aube & Whiffen, 1996), are less likely to ask help
from others when confronted with problems (Mongrain, 1998), and tend to
experience various interpersonal problems. More specifically, studies have shown
that self-critical/autonomous individuals have difficulty in connecting to people and
have the feeling that they are too controlling in relationships (e.g., Fichman et aI.,
1994; Morrison et aI., 1998).
Dependent/sociotropic individuals, in contrast, are more open towards others,
report fewer problems with intimacy, and are more likely to ask for help from
others, including medical assistance (Bornstein, 1992; Mongrain, 1998). However,
studies have also shown that they consider themselves as non-assertive and
submissive, but at the same time as demanding and as overly responsible in
relationships (Alden & Bieling, 1996; Bornstein, 1992; Ewart et aI., 1998; Lynch et
aI., 2003; Morrison et aI., 1998; Whisman & Friedman, 1998). In addition,
Dependency/Sociotropy, as expected, is associated with the inhibition of aggression
in intimate relationships (Alden & Bieling, 1996; Mongrain et aI., 2004; Mongrain
& Zuroff, 1994; Mongrain, Vettese, Shuster, & Kendal, 1998; Vettese & Mongrain,
2000; Zuroff & Fitzpatrick, 1995; Zuroff et aI., 1983), with difficulty in disclosing
true feelings (Ewart et al., 1998; Mongrain & Zuroff, 1994; Zuroff et aI., 1983), and
with difficulties leaving relationships (Ewart et aI., 1998; Kurdek, 2000).
In sum, contrary to some theories that conceptualize Dependency/Sociotropy as
a maladaptive trait, but consistent with findings reviewed earlier on the relationship
between Dependency/Sociotropy and life stress, Dependency/Sociotropy has
adaptive (e.g., more openness, fewer problems with intimacy) as well as
maladaptive (e.g., submissive, demanding) aspects. Similarly, Bornstein (1992)
emphasizes that dependency should not be equated with passivity. As we will show
below, dependency is often associated with very active and even manipulative
behaviors to obtain the needed care, love and support. Self-Critical Perfectionism/
Autonomy, in contrast, again emerges as a more "pure" vulnerability factor, in that
self-critical/autonomous individuals tend to have ambivalent, distant, and cold

III
relationships, and tend to experience much conflict and dissatisfaction in their
relationships, even in close, intimate relationships (Blatt, 1995, 2004).
Studies investigating the relationship between Dependency/Sociotropy and Self-
Critical Perfectionism!Autonomy and attachment styles further add to this picture,
but at the same time also highlight important differences between Blatt's and
Beck's conceptualizations (Morrison et al., 1998; Murphy & Bates, 1997; Reis &
Grenyer, 2002; Zuroff & Fitzpatrick, 1995). Both Dependency and Sociotropy are
most closely associated with a preoccupied attachment style, characterized by a
combination of desires for love, but also fear for loss of love. Blatt's concept of
Self-Critical Perfectionism is more closely related to a fearful-avoidant attachment
style, characterized by a longing for approval. together with fear for dependency
and closeness, dissatisfaction and distrust of others. Thus, Self-Critical
Perfectionism appears to be associated with an approach-avoidance conflic: in
relationships. Beck's (1983) Autonomy concept, however, is more closely related to
a dismissive attachment style, characterized by the (defensive) avoidance of
relationships. This is in keeping with Beck's (1983) emphasis on the strong needs
for independence, solitude, and control of autonomous individuals, resulting in a
defensive separation and even downright insensitivity to the needs of others (Clark
et al., 1995).
Evocation. Dependent!sociotropic and self-critical/autonomous individuals also
tend to elicit different reactions and thus environments. Moreover, they also tend to
show marked differences in their perceptions of their social environment. In this
context, research on the relationship between Dependency/Sociotropy and Self-
Critical Perfectionism/Autonomy and social support is particularly interesting.
Research has shown that low social support is an important antecedent of
depression (Barnett & Gotlib, 1988). It is therefore generally accepted that positive
social relationships and a good social integration buffer against depression, either
directly or indirectly (Cohen & Wills, 1985). From this perspective, it should not be
surprising that Dependency/Sociotropy in general shows a weaker relationship with
depression than Self-Critical Perfectionism/Autonomy. Because of their
fundamental orientation towards others, it could be argued that dependent!socio-
tropic individuals tend to have larger social networks than self-critical/autonomous
individuals and tend to have higher levels of perceived and/or actual social support
(Mongrain, 1998). Moreover, dependent/sociotropic individuals might be less at
risk for specific (congruent) life events (e.g., divorce, social isolation, etc.) known
to be associated with depression (Brown & Harris, 1978). Conversely, it could also
be argued that the often extreme neediness and claiming attitude of dependent!
sociotropic individuals leads to resentment in (significant) others, and finally in
rejection, putting dependent!sociotropic individuals at increased risk for (con-
gruent) life stress and thus depression (Mongrain, 1998). Self-critical individuals,
in contrast, may have a more limited and less satisfying social network because of
their cold, ambivalent relational style. The combination of this restricted and
negative social network with their continuous self-criticism, places these
individuals at greater risk for depression.
Studies in this area have again shown that while Dependency/Sociotropy
contains elements of both vulnerability and resilience, Self-Critical Perfectionism/

112
Autonomy appears to be a more clear-cut vulnerability factor. Priel and Shahar
(2000), for instance, found that Dependency/Sociotropy was associated with high
levels of perceived social support, which buffer the effect of stress, while Self-
Critical Perfectionism/Autonomy was associated with low levels of social support,
which in tum predicted increased levels of distress (see also Shahar & Priel, 2003;
Shahar, Henrich, Blatt, Ryan, & Little, 2003). Moreover, whereas Self-Critical
Perfectionism/Autonomy has been consistently related to low levels of social
support, evidence for Dependency/Sociotropy is more mixed, with some studies
reporting a positive association (Mongrain, 1998; Priel & Besser, 1999; Priel &
Shahar, 2000), others a negative association (Ewart et aI., 1998), and still others no
relationship (Reynolds & Gilbert, 1991). These mixed findings are consistent with
the findings that the measures of Dependency/Sociotropy are often a complex mix
of risk and resilience components (Blatt, 2004; Rude & Burnham, 1995). In
addition, levels of social support may show strong fluctuations in dependent!
sociotropic individuals (e.g., disillusionment after initial idealization). Priel and
Shahar (2000), for instance, found in a longitudinal study that Dependency/Socio-
tropy was positively related to perceived social support at Time I, but not nine
weeks later. Such fluctuations may be especially pronounced when dependent!
sociotropic individuals experience stress. As noted in the previous section, faced
with (interpersonal) stress, dependent!sociotropic individuals tend to seek help
from others, but in an excessive, claiming way (Beck, Robbins, Taylor, & Baker,
2001; Shahar, Joiner et al., 2004), as well as using venting as a coping mechanism
(Fichman et aI., 1999), possibly leading to (further) resentment in others. In
addition, several studies suggest that the good social integration and good quality
of relationships of dependent!sociotropic individuals can be deceptive (Ewart et al.,
1998; Flett et aI., 1997; Mongrain, 1998; Whiffen & Aube, 1999; Whisman &
Friedman, 1998; Zuroff et aI., 1995). Zuroff et al. (1995), for instance, reported that
although Dependency/Sociotropy was related to more frequent and more intimate
relationships, it was also related to feelings of dissatisfaction and dysphoria in
relationships. Thus, although relationships of dependent!sociotropic individuals
may seem positive. they most likely are also characterized by underlying
dissatisfaction and exaggerated fears of rejection.
Furthermore, all of these studies have focused on perceived. but not actual social
support. This is noteworthy because several studies have shown that dependent!
sociotropic individuals perceive their relationships as positive, loving, and
supporting, while others, including their partner, perceive these same relationships
in less positive terms (Bieling & Alden, 1998; Lynch et aI., 2003; Mongrain et aI.,
1998, 2004; Shapiro, 1988). Self-critical/autonomous individuals on the other hand
not only report that they are more critical, hostile, and ambivalent towards others,
but others, including their own partner, also perceive them as more critical and less
loving and therefore tend to react more critical and hostile (Bieling & Alden, 1998;
Lynch et al., 2003; Mongrain et aI., 1998, 2004; Whiffen & Aube, 1999; Zuroff &
Duncan, 1999). In addition, self-critical/autonomous individuals appear to
exaggerate negative perceptions by others in that they think that others perceive
them more negatively than is actually the case (Fichman et al., 1996, 1997). Thus,
in sum, dependent!sociotropic individuals tend to show a positive, idealizing bias in

113
their perceptions of their relationships, risking future disillusionment, while self-
critical/autonomous individuals tend to show a negative bias in their perceptions of
their relationships and in how they are perceived by others. These findings are
congruent with Blatt's (Blatt & Shichman, 1983) proposition that Dependency is
associated with the use of avoidant defense mechanisms, such as denial, whereas
Self-Critical Perfectionism is characterized by a harsh, judgmental, internalizing
stance.
Transformation. Empirical research has also yielded considerable support for
Blatt's (Blatt & Zuroff, 1992) and Beck's (1983) prediction that dependent and
perfectionistic subjects actively influence and transform their own environment.
Santor and Zuroff (1997, 1998), for instance, found in two experimental studies that
highly dependent/sociotropic women reacted to a situation threatening interper-
sonal relatedness by trying to maintain interpersonal contact, relinquishing control
over the situation, and minimizing disagreement (e.g., praising friends even when
they disagreed), whereas highly self-critical/autonomous women reacted to threats
to status by actively trying to control the situation, even at the expense of a close
friend. These results have been consistently replicated in other experimental studies
in both nonclinical and clinical samples (Bieling & Alden, 1998,2001; Santor et al.,
2000; Vettese & Mongrain, 2000) and by research using self-report questionnaires.
Dependent/sociotropic individuals value interpersonal goals at the expense of
achievement goals, whereas self-critical/autonomous individuals mainly value
achievement goals, at the expense of interpersonal goals (Ewart et al., 1998;
Luyten, Fontaine, Soenens, Duriez, & Corveleyn, 2004; Rosenfarb et al., 1994;
Mongrain & Zuroff, 1995; Zuroff & deLorimier, 1989; Zuroff, Moskowitz, & Cote,
1999; Zuroff et al., 1983). In addition, as noted earlier, others tend to react in
critical, hostile ways to self-critical/autonomous individuals, whereas the clinging
relational style of dependent/sociotropic individuals appears to eventually lead to
irritation and resentment in others.
To summarize, both Dependency/Sociotropy and Self-Critical Perfectionism/
Autonomy appear to be associated with dysfunctional interpersonal transactional
cycles (Kiesler, 1983, see also Andrews, 1989; Safran, 1990; Wachtel, 1994). Such
a cycle implies that one's interpersonal style leads to exactly these behaviors and
reactions in others that one fears and attempts to avoid, which in tum confirms
one's expectations. Whereas dependent/sociotropic individuals may be able to
generate a positive social environment, at the same time they seem to elicit by their
claiming, clinging relational style annoyance, resentment, and eventually rejection
and abandonment, confirming their underlying fear for rejection and abandonment.
Self-critical/autonomous individuals on the other hand are ambivalent, critical and
distrustful in relationships because they constantly fear, as well as express, criticism
and disapproval. Accordingly, they are perceived by others as cold and distant and
are thus less liked by others. Thus, self-critical/autonomous individuals are likely to
have not only few, but also very ambivalent relationships, confirming their
conviction that others do not like them and criticize and disapprove of them.
Moreover, they are unable to generate positive experiences and have also been
found to continue to solicit social comparisons, particularly when these
comparisons are unfavorable (Santor & Yazbek, 2003), which is likely to further

114
confirm their negative self-views. In addition, as Whiffen and Aube (1999) have
shown, complaints and criticism from others, especially from significant others,
may lead to elevated levels of Dependency/Sociotropy and Self-Critical
Perfectionism!Autonomy, further increasing the risk for the onset, exacerbation or
relapse of depression.
The implications of these findings for clinical practice, and especially with
regard to the nature of the therapeutic relationship, are readily apparent. As both
Blatt (e.g., Blatt, 1974; Blatt et aI., 2002) and Beck (1983) emphasize,
dependent/sociotropic and self-critical/autonomous individuals not only bring very
different needs and expectations into therapy, they will also perceive the therapeutic
relationship differently. They construct the therapeutic relationship in congruent
ways, so that their initial perceptions and experiences of the therapeutic relationship
match their maladaptive beliefs and expectations. Hence, the therapeutic process
should involve the identification, articulation and working through of these
transference reactions. These and related issues are further discussed in the next
section.

Clinical Implications

Impact of Dependency/Sociotropy and Self-Critical Perfectionism/Autonomy

on Psychotherapy

Several studies have shown that Dependency/Sociotropy and Self-Critical

Perfectionism!Autonomy influence the outcome of psychotherapy for depression.
To begin with, Blatt and colleagues have reported several re-analyses of the
extensive data from the well-known National Institute of Mental Health (NIMH)-
sponsored Treatment of Depression Collaborative Research Program (TDCRP;
Elkin, Parioff, Hadley, & Autry, 1985). In this elaborate and extensive randomized
clinical trial, 239 depressed outpatients were randomly assigned to three brief
treatments for depression, i.e., cognitive-behavioral therapy (CBT), interpersonal
therapy (lPT), and medication (imipramine plus clinical management) (lMI-CM),
and a placebo control plus clinical management (PLA-CM) condition. Initial
analyses revealed few differences in the efficacy of these four treatments (Elkin et
aI., 1989). However, re-analyses by Blatt and his colleagues of these data showed
that pre-treatment Self-Critical Perfectionism!Autonomy predicted poorer outcome
in all four conditions on all primary outcome measures, including symptoms,
general clinical functioning, and social adjustment, as rated by patients, therapists,
and independent clinical evaluators, at termination (16 weeks), as well as at follow-
up after 18 months (Blatt, Quinlan, Pilkonis, & Shea, 1995; Blatt, Zuroff, Bondi,
Sanislow, & Pilkonis, 1998). Thus, these findings support Blatt's (e.g., Blatt &

115
Maroudas, 1992) contention that self-critical/autonomous patients do not respond
well to short-term, standardized or manual directed therapy.
Further analyses showed that this negative effect of Self-Critical Perfec-
tionism/ Autonomy on outcome began to manifest itself primarily in the second half
of the treatment (Blatt, Zuroff et al., 1998). Several explanations of this intriguing
finding are possible. Self-critical/autonomous patients may in the second half of
treatment begin to perceive the impending end of treatment as arbitrary and thus as
impeding their sense of control and autonomy. Alternatively, because of their high
standards, they may find their degree of therapeutic change insufficient. Yet another
possibility is that self-critical/autonomous patients have difficulty in establishing a
good therapeutic relationship and become disillusioned in their therapist and
treatment in general. Congruent with this latter suggestion, further analyses of
the TOCRP revealed that self-critical/autonomous patients showed smaller
increases in the therapeutic alliance later in the treatment than other patients,
regardless of treatment condition, which in turn predicted worse outcome
(Blatt, Zuroff, Quinlan, & Pilkonis, 1996; Zuroff, Blatt, Sotsky, Krupnick, Martin,
Sanislow, & Simmens, 2000). Additional analyses (Shahar, Blatt, Zuroff, Krupnick,
& Sotsky, 2004) showed that Self-Critical Perfectionism/Autonomy also had
a negative impact on social relationships outside treatment, which in turn pre-
dicted poorer outcome. Moreover, Self-Critical Perfectionism/Autonomy was
also associated with increased vulnerability for (chronic) stress and stress-
related increases in interview-assessed depressive symptoms during the 18-
month follow-up period (Zuroff & Blatt, 2002). These findings are consistent
with findings reviewed earlier concerning the distant, ambivalent interper-
sonal style of self-critical/autonomous individuals, which appears to further
impair their social environment, thus placing them at increased risk for
relapse/recurrence. In addition, Shahar, Blatt, Zuroff and Pilkonis (2003) showed
that Self-Critical Perfectionism/Autonomy predicted negative outcome even
controlled for personality disorders, which have been previously associated with
worse outcome in the TOCRP (Shea et aI., 1990). Moreover, Self-Critical
Perfectionism/Autonomy also predicted poor therapeutic alliance and poor social
relationship satisfaction outside treatment, whereas personality disorder features
did not.
Blatt and colleagues also tried to identify factors that could mitigate the negative
effects of Self-Critical Perfectionism/Autonomy in the TOCRP. They found that
perfectionistic patients who had an early positive view of their therapist showed
more therapeutic gain. However, this effect was only found for patients with
moderate levels of perfectionism (Blatt, Zuroff, Quinlan, & Pilkonis, 1996).
Furthermore, Blatt, Sanislow, Zuroff and Pilkonis (1996) reported that effective
therapists, irrespective of the type of treatment they provided in the TOCRP, had a
more psychological than biological orientation and used more psychotherapy than
medical treatment for depression in their general clinical practice. Interestingly,
more effective therapists also believed that change would take longer to manifest
itself than less effective therapists. Finally, effective therapists also showed less
variability in outcome, that is: they seemed to be able to work effectively with most
of their patients. These findings are particularly impressive because they occurred

116
in a relatively homogeneous group of highly experienced therapists in three
manualized treatments.
Subsequent studies have consistently replicated the finding that Self-Critical
Perfectionism!Autonomy negatively affects outcome of brief treatments. Rector,
Zuroff and Segal (2000) found that Self-Critical Perfectionism/Autonomy
negatively predicted outcome as measured by the Beck Depression Inventory (BDI;
Beck & Steer, 1993) in a sample of 51 depressed outpatients treated with CBT.
Enns, Cox and Pidlubny (2002) reported that Self-Critical Perfectionism!Auto-
nomy negatively predicted outcome of brief group CBT in a sample of 65 patients
with residual depression. Similarly, Enns, Cox and Inayatulla (2003) reported that
Self-Critical Perfectionism!Autonomy negatively predicted outcome of inpatient
treatment of 78 adolescent suicidal patients, even controlled for neuroticism. Cox,
Walker, Enns and Karpinsky (2002), in turn, found that changes in levels of Self-
Critical Perfectionism!Autonomy were significantly related to outcome in brief
group CBT of 84 patients with generalized social phobia, even controlled for
baseline severity of social phobia and depressive symptoms. Cox et al. (2002,
p. 489) conclude from these findings that self-critical/autonomous patients "might
benefit from treatment augmentation or tailoring".
This latter suggestion was partially addressed in the Riggs-Yale Project, a study
of therapeutic change in a sample of 90 seriously disturbed, treatment-resistant
psychiatric inpatients that received long-term, intensive psychoanalytically oriented
treatment (4 to 5 times a week) (Blatt, Berman, Cook, & Ford, 1998; Blatt & Ford,
1994; Blatt, Ford, Berman, Cook, & Meyer, 1988). In general, it was found that
after an average 15 months of intensive treatment, introjective patients demon-
strated greater therapeutic gain than anaclitic patients. These findings are consistent
with Blatt's suggestion that introjective patients typically need longer, insight-
oriented treatment, whereas anaclitic patients may need more structured, supportive
treatment. Moreover, both types of patients showed primarily changes in charac-
teristics that were congruent with their dominant personality functioning. For
instance, anaclitic patients showed primarily changes in the quality of interpersonal
relationships, while introjective patients showed predominantly changes in
cognitive functioning. Further evidence for patient-by-treatment interactions were
found in re-analyses of another well-known psychotherapy study, the Menninger
Psychotherapy Research Project, which compared the effect of psychoanalysis (5
times a week) and psychodynamic supportive-expressive psychotherapy (2-3 times
a week). Although in general little differences in outcome were found in this study
between psychoanalysis and psychodynamic psychotherapy, re-analyses showed
that introjective patients had greater positive change in psychoanalysis than in
psychodynamic therapy, while the reverse was found for anaclitic patients (Blatt,
1992; Blatt & Shahar, 2004).
Why would dependent/sociotropic patients benefit relatively more from
supportive-expressive psychotherapy and self-critical/autonomous from insight-
oriented psychotherapy? According to Blatt (e.g., Blatt & Ford, 1994; see also
Beck, 1983), these findings are consistent with his suggestion that dependent!
sociotropic might be particularly responsive to interpersonal factors in treatment,
whereas self-critical/autonomous may benefit most from interpretation and insight.

117
Three studies have yielded supporting evidence for these assumptions. In another
re-analysis of the Riggs-Yale Study, Fertuck, Bucci, Blatt and Ford (2004) found
that the extent to which patients were able to articulate their feelings and thoughts
predicted improvement in introjective patients, but was negatively associated with
improvement in anaclitic patients. However, increased experiences of closeness
predicted clinical improvement in anaclitic patients. Hence, anaclitic patients seem
to be able to benefit from a good therapeutic alliance, rather than from exploration
and insight, while introjective patients appeared to be especially responsive to
interpretation and insight. Blatt and Shahar (2004), in turn, sought to identify
factors that were responsible for the fact that introjective patients did better in
psychoanalysis, while anaclitic patients showed more improvement in supportive-
expressive psychotherapy in the Menninger Psychotherapy Research Project. They
found that psychoanalysis was associated with increases in associational activity,
whereas supportive-expressive psychotherapy led to decreased associational
activity. Again, these findings suggest that introjective patients appear to benefit
from insight, anaclitic patients from the therapeutic relationship. Piper, Joyce,
McCallum, Azim and Ogrodniczuk (200 I) reported similar findings. These
investigators randomized 114 mixed psychiatric outpatients to either 20 weeks of
manualized psychodynamically-oriented supportive or interpretive psychotherapy.
Patients with higher quality of object relations fared better in interpretive therapy
compared to those with lower levels of object relations. In addition, a relatively
high frequency of transference interpretations predicted poorer outcome and a more
negative therapeutic alliance in patients with low quality of object relations. These
findings are consistent with Blatt's (1974) notion that patients with relatively high
levels of object relations such as introjective patients fare better in interpretive
psychotherapy, whereas those with lower quality of object relations such as
anaclitic patients do not respond well to a highly interpretive approach, but benefit
more from a supportive relationship. Together, these findings suggest that
introjective patients, despite considerable problems in interpersonal relationships,
are able to benefit from insight in the context of a therapeutic relationship that had
sufficient time to develop. In contrast, anaclitic patients, perhaps because of the
symbiotic-like quality of their relationships and the fact that they are less able to
reflect on and articulate their feelings, might focus on the therapeutic relationship
as such, and are therefore less capable to benefit from insight in psychotherapy.
Hence, future research should investigate the relationship between the anaclitic-
introjective distinction and the capacity for mentalization (Fonagy, Gergely, Jurist,
& Target, 2002).
Research inspired by Beck's (1983) formulations adds further support for the
influence of Dependency/Sociotropy and Self-Critical Perfectionism/Autonomy on
psychosocial treatments. Zettle, Haflich and Reynolds (1992), for instance, re-
ported that highly dependent/sociotropic depressed patients responded better to
group CBT, while highly self-critical/autonomous patients showed more impro-
vement in individual CBT, which is consistent with Blatt's and Beck's predictions
that dependent/sociotropic patients benefit more from relational aspects (e.g., social
support) and self-critical/autonomous patients from an individual, insight-oriented
approach. Yet, in a similar study, Zettle and Herring (1995) found that depressed

118
patients that received "matched" treatment (e.g., individual therapy for highly self-
critical/autonomous patients) showed similar levels of severity of depression after
treatment. However, at follow-up after two months, remission rates were
significantly lower for "matched" patients. Kuyken, Kurzer, DeRubeis, Beck and
Brown (200 I), finally, reported that avoidant and paranoid beliefs (belonging to the
introjective/autonomous cluster; Beck, 1999; Blatt & Shichman, 1983), but not
dependent beliefs (that belong to the anaclitic/sociotropic cluster), negatively
predicted outcome in CBT of 162 depressed outpatients.
In summary, self-critical/autonomous patients apparently require a substantial
period of time and more intensive contact before they can establish an effective
therapeutic relationship and can begin to change their harsh, overly self-critical
views. In the long-term, however, these patients can make substantial therapeutic
progress (Blatt & Ford, 1994; Blatt & Shahar, 2004). Dependent!sociotropic
patients on the other hand, can more easily establish a therapeutic relationship and
seem to benefit most, at least in the initial phases of treatment, but perhaps also
during later stages of treatment, from interpersonal aspects of treatment and appear
not to be as responsive to interpretation and insight.

Impact of Dependency/Sociotropy and Self-Critical Perfectionism/Autonomy

on Pharmacotherapy of Depression

Few studies have systematically investigated the impact of Dependency/Socio-

tropy and Self-Critical Perfectionism! Autonomy on pharmacological treatment. In
line with Beck's (1983) predictions, Peselow, Robins, Sanfilipo, Block and Fieve
(1992) found in a study of 217 depressed outpatients that Dependency/Sociotropy
negatively predicted outcome of antidepressant treatment, whereas Self-Critical
Perfectionism! Autonomy was associated with good antidepressant response. In
addition, highly self-critical/autonomous low dependent!sociotropic patients
showed a significant better response to antidepressant drug treatment (74%) than
placebo (25%) and than did highly dependent!sociotropic low self-critical/ auto-
nomous patients (39%). Furthermore, this latter group showed no significant drug-
placebo difference (39% versus 32%). In line with Beck's (1983) predictions, 91%
of high self-critical/autonomous low dependent!sociotropic patients in their sample
met Research Diagnostic Criteria (RDC; Spitzer, Endicott, & Robins, 1978) for
endogenous depression as opposed to only 20% in the high dependent! sociotropic
low self-critical/autonomous group. Rector et al. (2000), however, found that Self-
Critical Perfectionism/Autonomy was unrelated to pharmacothera-py outcome as
measured by the BOI in a study of 58 depressed outpatients, while Blatt, Quinlan
et al. (1995) found that pre-treatment Self-Critical Perfectionism! Autonomy was a
negative predictor of pharmacological treatment in the NIMH TDRCP.
There are a number of possible explanations for these divergent findings,
including differences in the samples employed and/or different antidepressant

119
regimes. Interestingly, Scott (200 I, cited in Flett & Hewitt, 2002) reported that
Self-Critical Perfectionism!Autonomy was associated with several variables that
predicted nonadherence to medication, which is consistent with Blatt's and Beck's
views that these patients are pessimistic about professional help because of
conscious and unconscious self-destructive tendencies (Blatt, 1995). Moreover,
these patients might perceive the antidepressant treatment regime as undermining
their need for personal control. Hence, their tendency for nonadherence could in
part explain why Self-Critical Perfectionism!Autonomy could have a negative
impact on outcome of pharmacological treatment (see also Demyttenaere, 1997).
Thus, it is somewhat paradoxically that these patients respond well to
pharmacotherapy as reported by Peselow et al. (1992).
Results for Dependency/Sociotropy have been somewhat more consistent.
Bothwell and Scott (1997) reported that Dependency/Sociotropy predicted non-
recovery in a 2-year follow-up study of depressed patients treated with
antidepressant medication. Similarly, as Peselow et al. (1992), Frank, Kupfer, Jacob
and Jarrett (1987) found that dependent patients were less responsive to
pharmacotherapy. In line with these findings, several studies have found that
Reward Dependency and Harm Avoidance, two dimensions in Cloningers
(Cloninger, Svrakic, & Przbybeck, 1993) psychobiological theory of temperament
and personality, which are conceptually related to Blatt's and Beck's notions of
Dependency/Sociotropy, negatively predicted pharmacological treatment, although
some recent studies have failed to replicate these findings (Mulder, 2002).
Regarding the effect of pharmacotherapy on personality, in at least five studies
of depressed patients it was found that brief pharmacological treatment did :10t
significantly alter levels of both Dependency/Sociotropy and Self-Critical
Perfectionism!Autonomy (Bagby et aI., 1994, 200 I; Franche & Dobson, 1992;
Moore & Blackburn, 1996; Reda, Carpiniello, Secchariollo, & Blanco, 1985).
Rector et aI. (2000), however, reported that a 12-week antidepressant trial in a
sample of 58 depressed outpatients led to significant decreases in levels of
Dependency/Sociotropy and Self-Critical Perfectionism!Autonomy. Similarly,
Hellerstein, Kocsis, Chapman, Stewart and Harrison (2000) reported that pharma-
cological treatment resulted in a significant drop in levels of Harm Avoidance and
Reward Dependence. As these two latter studies lack a control group of non-
depressed and never-depressed individuals, these results are difficult to interpret.
Moreover, as previously noted, one should be careful to draw any conclusions from
such studies because measures of Dependency/Sociotropy and Self-Critical
Perfectionism!Autonomy contain both state and trait elements (Zuroff et aI., 1999).
Moreover, whereas brief treatment may result in the (partial) deactivation of
maladaptive cognitions and affects associated with these personality dimensions
(e.g., Rector et aI., 2000), as the data from the TDCRP (Zuroff et al., 1999) and
priming studies (Segal & Ingram, 1994) suggest, Dependency/Sociotropy and Self-
Critical Perfectionism!Autonomy may in fact show high levels of relative stability
and may easily be re-activated. Hence, long-term follow-up is essential to measure
the true efficacy of any treatment, pharmacological or psychosocial.
At the same time, it is clear that further research is needed concerning the
relationship between these personality dimensions and pharmacological treatment.

120
On a theoretical note, these studies could play an important role in the integration
of biological and psychological theories of depression. As noted in Chapter 3, both
Blatt and Beck have speculated on the biological origins of Dependency/Sociotropy
and Self-Critical Perfectionism/Autonomy (Beck, 1983; Blatt, Cornell, & Eshkol,
1993), and Cloninger et al. (1993) have proposed that Harm Avoidance and Reward
Dependence are associated with serotonine and norepinephrine functioning
respectively. From a clinical point of view, future studies exploring the relationship
between personality and antidepressant response thus might also enable better
matching of patients to specific pharmacological treatments (Mulder, 2002).

Treatment Implications

Several treatment strategies have been developed to treat Dependency/Socio-

tropy and/or Self-Critical Perfectionism/Autonomy based on various theoretical
perspectives, including behavioral and cognitive-behavioral (Beck, 1983; Di
Bartolo, Frost, Dixon, & Almodovar, 200 I; Ferguson & Rodway, 1994; Hirsch &
Hayward, 1998), psychodynamic (Arieti & Bemporad, 1978; Blatt, 2004; Bowlby,
1980; Fredtoft, Poulsen, Bauer, & Malrn, 1996), and integrative (Barrow & Moore,
1983; Halgin & Leahy, 1989) perspectives. Unfortunately, at present little is known
about the efficacy of these specific treatments.
Studies so far suggest the following preliminary conclusions and guidelines.
First, a growing research literature suggests that patient characteristics such as
Dependency/Sociotropy and Self-Critical Perfectionism/Autonomy are important
predictors of therapeutic outcome in the treatment of depression, regardless of
therapeutic orientation, and should therefore be an important consideration in
treatment planning. In addition, brief pharmacologic and psychotherapeutic
treatments of depression appear to be relatively ineffective in altering both
personality dimensions (Zuroff & Blatt, 2002). Self-Critical Perfectionism/
Autonomy appears to exert a particularly negative influence on outcome. This
conclusion is further supported by the fact that pre-treatment Self-Critical
Perfectionism/Autonomy in the NIMH TDCRP was associated with the opinion of
independent clinical evaluators at termination and at follow-up that patients needed
further treatment as well as with these patients's own dissatisfaction with their
treatment (Blatt, Quinlan et al., 1995). Thus, treatments of depression that do not
address personality issues might leave these vulnerabilities, which have both direct
and indirect (e.g., through life stress) effects on depression, unaltered (Daley et aI.,
1998; Blatt & Zuroff, 2004).
Moreover, whereas dependent/sociotropic patients might initially respond better
to brief (psychosocial) treatments, studies reviewed suggest that this initial
response might be based more on positive transference feelings associated with
these individuals tendency to feel relieved once they have the feeling that they are
being helped, rather than by presumed neurobiological effects of antidepressants or

121
specific psychotherapeutic techniques. This might also explain in part the high
relapse rates after brief treatment of depression (Westen & Morrison, 200 I). One
possible explanation that deserves further investigation is that dependent/socio-
tropic patients, in such brief treatments, establish a new and idealizing dependent
relationship, which initially ameliorates their symptoms (e.g., see Arieti &
Bemporad, 1978), what in psychoanalysis has been discussed as a transference
cure. However, once this therapeutic relationship ends or becomes more negative
(e.g., because of alliance ruptures), treatment effects may disappear. Future
research should investigate this suggestion, and should moreover be aimed at
identifying which patients benefit most from which forms of psychotherapy. or
which aspects of the therapeutic process. For instance, future research should
explore the possibility that dependent/sociotropic and self-critical/autonomous
patients may benefit considerably from brief interventions that are specifically
tailored to the needs and expectations of these patients. In fact, studies have found
that although Self-Critical Perfectionism!Autonomy negatively predicts outcome in
brief treatments, changes in Self-Critical Perfectionism!Autonomy predicted better
outcome (Enns et aI., 2002; Rector et al., 2000). In other words, the extent to which
these brief treatments were able to modify levels of Self-Critical Perfectionism!
Autonomy was associated with better outcome. Similarly, future studies should
identify depressed patients that typically need longer treatment. Another direction
for future research is to examine how typical transference and countertransference
issues associated with Dependency/Sociotropy and Self-Critical Perfectionism/
Autonomy emerge and are modified in the therapeutic relationship.
A second implication of the findings reviewed is that common factors such as
therapist characteristics (e.g., empathy) and relationship factors (e.g., therapeutic
alliance) are important predictors of therapeutic success in the treatment of
depression, regardless of theoretical orientation (Blatt, Zuroff et aI., 1996; Burns &
Nolen-Hoeksema, 1992; Horvath & Symonds, 1991; Krupnick et aI., 1996;
Wampold, 1997). These are not isolated findings. In fact, Lambert and Barley
(2002) estimated that specific techniques account for only approximately 15~ in
therapeutic outcome, whereas common factors explain about 30% in therapeutic
change. Findings from the NIMH TDCRP are particularly informative in this
context. While very few mode-specific effects were found for CBT and IPT, patient,
therapist and interpersonal relationship factors did predict outcome, regardless of
specific techniques (Blatt, Zuroff et a!., 1996, 1998; Imber et a!., 1990). In addition,
as hypothesized by psychodynamic (e.g., Freud, 191211958), experiential (Rogers,
1957), and cognitive-behavioral (Beck, Rush, Shaw, & Emery, 1979) theorists,
common factors such as a good therapeutic alliance appear to be an essential if not
necessary condition for specific techniques to be effective across various forms of
treatment (e.g., Division 29 Task Force on Empirically Supported Therapy
Relationships, 2002; Gaston, Thompson, Gallagher, Cournoyer, & Gagnon, 1998;
Rector et aI., 1999). A "good-enough" therapeutic relationship seems to be the
foundation on which specific techniques can exert their influence.
Thus, in our view it is time that patient and therapist characteristics are included
in the assessment, treatment planning and actual treatment of depression. To those
who are familiar with psychotherapy research, existing guidelines for the treatment

122
of depression clearly underestimate, to say the least, the importance of patient and
therapist factors, and overestimate the importance of specific techniques (Blatt &
Zuroff, 2004). In addition, existing guidelines (e.g., American Psychiatric
Association, 2000) incorrectly view patients in an undifferentiated way, expecting
them to be uniformly responsive to different pharmacological and psychosocial
treatments (Lambert & Ogles, 2004). Although the research literature just reviewed
clearly has its limitations, the almost complete neglect of patient and therapist
characteristics in these guidelines appears no longer warranted in the light of
existing research (for similar opinions, see Beutler, Clarkin, & Bongar, 2000; Blatt
& Zuroff, 2004; Division 29 Task Force on Empirically Supported Therapy
Relationships, 2002; Lambert & Ogles, 2004; Lutz, 2002; Piper et aI., 2001).

Conclusions and Directions for Future Research

Almost three decades of research have led to considerable integration among

psychodynamic and cognitive-behavioral theories of depression and have found
considerable support for these formulations. Both perspectives clearly have
mutually influenced and enriched each other (Robins, 1993). Now, the time seems
ripe to bridge the gap between these two psychosocial approaches and biological
approaches of depression. In this chapter, we suggested several possibilities for
such integration in various areas, including the biological determinants of
temperament and personality (Cloninger et aI., 1994; Woodside et aI., 2002), the
relationship between neurobiological responses to stress and personality (see also
Chapter 8, this volume), and the influence of personality on the effects of
pharmacotherapy.
To conclude, we are confident that the coming years will continue to see a
growing number of studies concerning the role of Dependency/Sociotropy and Self-
Critical Perfectionism!Autonomy in the etiology, pathogenesis, and treatment of
depression. These studies will undoubtedly lead to modifications in both Blatt's and
Beck's conceptualizations. These modifications are only natural, however, and are
the essence of the development of scientific theories and investigations that enable
us to achieve a greater understanding of depression that more closely matches the
complexities of clinical reality.

References

Alden. L. E., & Bieling, P. 1. (1996). Interpersonal convergence of personality constructs in dynamic and
cognitive models of depression. Journal of Research in Personality. 30, 60-75.
Alford, B. A., & Gerrity, D. M. (1995). The specificity of sociotropy-autonomy personality dimensions to

123
 depression vs. anxiety. Journal of Clinical Psychology, 51,190-195.
Allen, N. 8., Horne, D. J., & Trinder, J. (1996). Sociotropy, autonomy, and dysphoric emotional responses
to specific classes of stress: A psychophysiological evaluation. Journal ofAbnormal Psychology, 105,
25-33.
American Psychiatric Association (2000). Practice guidelines for the treatment of patients with major
depression. Retrieved September 3, 2004, from
https://www.psych.org/psych_pract/treatg/pglDepression2e.book.cfm
Andrews, J. D. W. (1989). Psychotherapy of depression: A self-confirmation model. Psychological Review,
96,576-607.
Arieti, S., & Bemporad, 1. (1978). Psychotherapy of severe and mild depression. Northvale/London: Jason
Aronson.
Aube, J., & Whiffen, V. E. (1996). Depressive styles and social acuity: Further evidence for dis:inct
interpersonal correlates of dependency and self-criticism. Communication Research, 23, 407-424.
Bagby, R. M., Gilchrist, E. J., Rector, N. A., Dickens, S. E., Joffe, R. T, Levitt, A, Levitan, R. D., &
Kennedy, S. H. (2001). The stability and validity of the sociotropy and autonomy personality
dimensions as measures by the Revised Personal Style Inventory. Cognitive Therapy and Research. 25,
765-779.
Bagby, R. M., Parker, J. D. A, Joffe, R. T.. & Buis, T (1994). Reconstruction and validation of the
Depressive Experiences Questionnaire. Assessment, 1, 59-68.
Bagby, R. M., Parker, 1. D. A., Joffe, R. T., Schuller, D.. & Gilchrist, E. (1998). Confirmatory factor
analysis of the revised Personal Style Inventory (PSI). Assessmentc S, 31-43.
Bagby, R. M., Schuller, D. R., Parker, J. D. A.. Levitt, A., Joffe, R. T, & Shafir, M. S. (1994). Major
depression and the self-criticism and dependency personality dimensions. American Journal of
Psychiatry, 151,597-599.
Barnett, P. A., & Gotlib, I. H. (1988). Psychosocial functioning and depression: Distinguishing among
antecedents, concomitants, and consequences. Psychological Bulletin, 104,97-126.
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological
research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social
Psychology, 41, 1173-1182.
Barrow, 1. c., & Moore, C. A. (1983). Group interventions with perfectionist thinking. The Personnel and
Guidance Journal, 61, 612-615.
Bartelstone, 1. H., & Trull, T J. (1995). Personality, life events, and depression. Journal of Personality
Assessment, 64, 279-294.
Beck, A. T. (1983). Cognitive therapy of depression: New perspectives. In P. J. Clayton & J. E. Barrett
(Eds.), Treatment of depression: Old controversies and new approaches (pp. 265-290). New York:
Raven Press.
Beck, A T (1999). Cognitive aspects of personality disorders and their relation to syndromal disorders:
A psychoevolutionary approach. In C. R. Cloninger (Ed.), Personality and psychopathology (pp. 411-
429). Washington, DClLondon: American Psychiatric Press.
Beck, A. T, Epstein, N., Harrison, R. P., & Emery, G. (1983). Development of the Sociotropy-Autonomy
Scale: A measure of personality factors ill psychopathology. Unpublished manuscript, University of
Pennsylvania, Philadelphia.
Beck, A. T., Rush, A J., Shaw, B. E, & Emery, G. (1979). Cognitive therapy of depression. New York: The
Guilford Press.
Beck, A. T, & Steer, R. A. (1993). Beck Depression Inventory Manual. San Antonio, TX: The
Psychological Corporation.
Beck, R., Robbins, M., Taylor, C., & Baker. L. (200 I). An examination of sociotropy and excessive
reassurance seeking in the prediction of depression. Journal of Psychopathology and Behavioral
Assessment, 23, 101-104.
Besser, A., Flett, G. L., & Davis, R. A. (2003). Self-Criticism, dependency, silencing the self, and
loneliness: A test of a mediational model. Personality and Individual Differences, 35, 1735-1752.
Besser, A., & Priel, B. (2003). Trait vulnerability and coping strategies in the transition to motherhood.
Current Psychology: Developmental. Learning, Personality, Social, 22, 57-72.
Beutler, L. E., Clarkin, J. E, & Bongar, B. (2000). Guidelines for the systematic treatment of the depressed
patient. New York/Oxford: Oxford University Press,
Bieling, P, 1., & Alden, L. E. (1998). Cognitive-interpersonal patterns in dysphoria: The impact of
sociotropy and autonomy. Cognitive Therapy and Research, 22,161-178.

124
Bieling. P. J., & Alden, L. E. (200 I). Sociotropy, Autonomy, and the interpersonal model of depression: An
integration. Cognitive Therapy and Research, 25, 167-184.
Bieling, P. J., Beck, A. T., & Brown, G. K. (2000). The Sociotropy-Autonomy Scale: Structure and
implications. Cognitive Therapy and Research, 24, 763-780.
Bieling, P. J., Israeli, A. L., & Antony, M. M. (2004). Is perfectionism good, bad, or both? Examining
models of the perfectionism construct. Personality and Individual Differences. 36, 1373-1385.
Blaney, P. H. (2002). Stress and depression: A personality/situation interaction approach. In S. L. Johnson,
A. M. Hayes. T. Field, P. McCabe, & N. Schneidermann (Eds.), Stress, coping, and depression (pp. 89-
116). Mahw '!. NJ: Erlbaum.
Blaney. P. H., & ",-.,'cher, G. S. (1991). Measures of depressive dimensions: Are they interchangeable?
Journal of Personality Assessment, 56, 502-512.
Blatt, S. J. (1974). Levels of object representation in anaclitic and introjective depression. The
Psychoanalytic Study of the Child, 29, 107-157.
Blatt, S. J. (1992). The differential effect of psychotherapy and psychoanalysis with anaclitic and
introjective patients: The Menninger psychotherapy research project revisited. Journal of the American
Psychoanalytic Association, 40. 691-724.
Blatt, S. J. (1995). The destructiveness of perfectionism. Implications for the treatment of depression.
American Psychologist. 50. 1003-1020.
Blatt, S. J. (1998). Contributions of psychoanalysis to the understanding and treatment of depression.
Journal of the American Psychoanalytic Association, 46, 722-752.
Blatt. S. 1. (2004). Experiences of depression: Theoretical, clinical and research perspectives. Washington,
DC: American Psychological Association.
Blatt. S. J.• Berman. W. H.• Cook, B. P., & Ford. R. Q. (1998). Effectiveness of long-term. intensive.
inpatient treatment for seriously disturbed young adults: A reply to Bein. Psychotherapy Research. 8.
42-53.
Blatt. S. J., Cornell. C. E.• & Eshkol, E. (1993). Personality style, differential vulnerability and clinical
course in immunological and cardiovascular disease. Clinical Psychology Review, 13,421-450.
Blatt. S. J., D' Afflitti. 1. P., & Quinlan, D. M. (1976). Experiences of depression in normal young adults.
Journal ofAbnormal Psychology. 85. 383-389.
Blatt. S. J.• & Ford, R. Q. (1994). Therapeutic change. An object relations perspective. New York/London:
Plenum Press.
Blatt. S. 1.. Ford, R. Q., Berman. w.. Cook. 8.. & Meyer, R. (1988). The assessment of therapeutic change
in schizophrenic and borderline young adults. Psychoanalytic Psychology, 5. 127-158.
Blatt. S. J., & Maroudas, C. (1992). Convergence among psychoanalytic and cognitive-behavioral theories
of depression. Psychoanalytic Psychology, 9, 157-190.
Blatt, S. J.• Quinlan, D. M., Chevron, E. S., McDonald, C., & Zuroff, D. (1982). Dependency and self-
criticism: Psychological dimensions of depression. Journal of Consulting and Clinical Psychology. 50,
113-124.
Blatt. S. J.• Quinlan, D. M., Pilkonis, P. A., & Shea, M. T. (1995). Impact of perfectionism and need for
approval on the brief treatment of depression: The National Institute of Mental Health Treatment of
Depression Collaborative Research Program revisited. Journal of Consulting and Clinical Psychology,
63, 125-132.
Blatt, S. J., Sanislow, C. A., Zuroff, D. C., & Pilkonis, P. A. (1996). Characteristics of effective therapists:
Further analyses of data from the NIMH TDCRP. Journal of Consulting and Clinical Psychology. 63.
125-132.
Blatt. S. J., Schaffer, C. E., Bers, S. A., & Quinlan. D. M. (1992). Psychometric properties of the
Depressive Experiences Questionnaire for adolescents. Journal of Personality Assessment, 59, 82-98.
Blatt, S. J., & Shahar, G. (2004). Psychoanalysis: With whom, for what. and how? Comparisons with
psychotherapy. Journal of the American Psychoanalytic Association, 52, 393-447.
Blatt, S. 1., Shahar, G., & Zuroff, D. C. (2002). Anaclitic/Sociotropic and Introjective/Autonomous
dimensions. In J. C. Norcross (Ed.), Psychotherapy relationships that work. Therapist contributions
and responsiveness to patients (pp. 315-333). Oxford: Oxford University Press.
Blatt, S. J., & Shichman, S. (1983). Two primary configurations of psychopathology. Psychoanalysis and
Contemporary Thought, 6, 187-254.
Blatt, S. J., & Zuroff, D. C. (1992). Interpersonal relatedness and self-definition: Two prototypes for
depression. Clinical Psychology Review, 12, 527-562.
Blatt, S. J., & Zuroff, D. C. (2004). Empirical evaluation of the assumptions in identifying evidence-based

125
 treatments in mental health. Manuscript submitted for publication.
Blatt, S. J., Zuroff, D. c, Bondi, C. M., Sanislow, C. A., & Pilkonis, P. A. (1998). When and :10W
perfectionism impedes the brief treatment of depression: Further analyses of The National Institute of
Mental Health Treatment of Depression Collaborative Research Program. Journal of Consulting and
Clinical Psychology, 66, 423-428.
Blatt, S. J., Zuroff, D. c, Quinlan, D. M., & Pilkonis, P. A. (1996). Interpersonal factors in brief treatment
of depression: Further analyses of The National Institute of Mental Health Treatment of Depression
Collaborative Research Program. Journal of Consulting and Clinical Psychology, 64,162-171.
Bornstein, R. E (1992). The dependent personality: Developmental, social, and clinical perspectives.
Psychological Bulletin, 112,3-23.
Bothwell, R., & Scott, J. (1997). The influence of cognitive variables on recovery in depressed inpatients.
Journal ofAffective Disorders, 43, 207-212.
Bowlby, J. (1980). Attachment and loss ""'I. 3 Loss: Sadness and depression. London: The Hogarth Press.
Boyce, P., Parker, G., Barnett, S., Cooney, M., & Smith, E (1991). Personality as a vulnerability factcr to
depression. British Journal of Psychiatry, 159, 106-114.
Brewin, C. R., & Firth-Cozens, J. (1997). Dependency and self-criticism as predictors of depression in
young doctors. Journal of Occupational Health Psychology, 2, 242-246.
Brown, G. P., Hammen, C. L., Craske, M. G.. & Wickens, T. D. (1995). Dimensions of dysfunctional
attitudes as vulnerabilities to depressive symptoms. Journal ofAbnormal Psychology, 104,431-435.
Brown, G. W., & Harris, T. O. (1978). Social origins of depression. Routledge: London.
Burgess, E., Lorah, D., Haaga, D. A. E, & Chrousos, G. (1996). Sociotropy, Autonomy, stress, and
depression in Cushing Syndrome. Journal of Nervous and Mental Disease, 184,362-367.
Burns, D. D., & Nolen-Hoeksema, S. (1992). Therapeutic empathy and recovery from depression in
cognitive-behavioral therapy: A structural equation model. Journal of Consulting and Clinical
Psychology, 60, 441-449.
Buss, D. M. (1987). Selection, evocation, and manipulation. Journal of Personality and Social Psychology,
53, 1214-1221.
Cane, D. B., Olinger, L. J., Gotlib, I. H., & Kuiper, N. A. (1986). Factor stucture of the Dysfunctional
Attitude Scale in a student population. Journal of Clinical Psychology, 42, 307-309.
Chang, E. C. (2000). Perfectionism as a predictor of positive and negative psychological outcomes:
Examining a mediation model in younger and older adults. Journal of Counseling Psychology, 47, 18-
26.
Chang, E. C., & Rand, K. L. (2000). Perfectionism as a predictor of subsequent adjustment: Evidence for
a specific diathesis-stress mechanism among younger and older adults. Journal of Counseling
Psychology, 47, 129-137.
Clark, D. A., & Beck, A. T. (1991). Personality factors in dysphoria: A psychometric refinement of Beck's
Sociotropy-Autonomy Scale. Journal of Psychopathology and Behavioral Assessment, 13, 369-388.
Clark, D. A., & Beck, A. T. (1999). Scientific foundations of cognitive theory and therapy of depression.
New York: John Wiley & Sons.
Clark, D. A., Beck, A. T., & Brown, G. K. (1992). Sociotropy, autonomy, and life event perceptions in
dysphoric and nondysphoric individuals. Cognitive Therapy and Research, 16, 635-652.
Clark, D. A., & Oates, T. (1995). Dailey hassles, major and minor life events and their interaction with
sociotropy and autonomy. Behaviour Research and Therapy, 33, 819-823.
Clark, D. A., Steer, R. A., Beck, A. T.. & Ross, L. (1995). Psychometric properties of revised sociotropy
and autonomy scales in college students. Behaviour Research and Therapy, 33, 325-334.
Cloninger, C. R., Svrakic, D. K., & Przbybeck, T. R. (1993). A psychobiological model of temperament
and character. Archives of General Psychiatry, 50, 975-990.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbai.m,
Cohen, S., & Wills, T. A. (1985). Stress, social support, and the buffering hypothesis. Psychological
Bulletin, 98, 310-357.
Connor-Smith, J. K., & Compas, B. E. (2002). Vulnerability to social stress: Coping as a mediator or
moderator of sociotropy and symptoms of anxiety and depression. Cognitive Therapy and Research, .26,
39-55.
Cox, B. J., & Enns, M. W. (2003). Relative stability of dimensions of perfectionism in depression.
Canadian Journal of Behavioural Science, 35, 124-132.
Cox, B. 1., McWilliams, L. A., Enns, M. W., & Clara, I. P. (2004). Broad and specific personality
dimensions associated with major depression in a nationally representative sample. Comprehensive

126
 Psychiatry, 45, 246-253.
Cox, B. J.. Walker, 1. R., Enns, M. W, & Karpinski, D. C. (2002). Self-criticism in generalized social
phobia and response to cognitive-behavioral treatment. Behavior Therapy, 33, 479-491.
Coyne, 1. C; & Whiffen, V. E. (1995). Issues in personality as diathesis for depression: The case of
sociotropy-dependency and autonomy-self-criticism. Psychological Bulletin, 118,358-378.
Daley, S. H., Hammen, c, Burge, D., Davila, J.. Paley, B., Lundberg, N., & Herzberg, D. S. (1997).
Predictors of the generation of episodic stress: A longitudinal study of late adolescent women. Journal
ofAbnormal Psychology, 106,251-259.
Demyttenaere, K. (1997). Compliance during treatment with antidepressants. Journal of Affective
Disorders, 43, 27-39.
Di Bartolo, P. M., Frost, R. 0., Dixon, A., & Almodovar, S. (200 I). Can cognitive restructuring reduce the
disruption associated with perfectionistic concerns'? Behavior Therapy, 32, 167-184.
Dimitrovsky, L., Levy-Schiff, R., & Schattner-Zanany, I. (2002). Dimensions of depression and
perfectionism in pregnant and nonpregnant women: Their levels and interrelationsships and their
relationship to marital satisfaction. Journal of Psychology, 136, 631-646.
Division 29 Task Force on Empirically Supported Therapy Relationships (2002). Conclusions and
recommendations of the Division 29 Task Force. In J. C. Norcross (Ed.), Psychotherapy relationships
that work (pp. 441-443). Oxford: Oxford University Press.
Dowis, D. J. A., & Backs-Dermott, B. J. (2000). Sociotropic personality and information processing
following imaginal priming: A test of the congruency hypothesis. Canadian Journal of Behavioural
Science, 32, 117-126.
Dunkley, D. M., & Blankstein, K. R. (2000). Self-critical perfectionism, coping, hassles, and current
distress: A structural equation modeling approach. Cognitive Therapy and Research, 24, 713-730.
Dunkley, D. M., Blankstein, K. R., Halsall, J., Williams, M., & Winkworrh, G. (2000). The relation
between perfectionism and distress: Hassles, coping, and perceived social support as mediators and
moderators. Journal of Counseling Psychology, 47, 437-453.
Dunkley, D. M., Zuroff, D. C; & Blankstein, K. R. (2003). Self-Critical perfectionism and daily affect:
Dispositional and situational influences on stress and coping. Journal of Personality and Social
Psychology, 84, 234-252.
Elkin, I., Parioff, M. B., Hadley, S. W., & Autry, J. H. (1985). NIMH Treatment of Depression
Collaborative Research Program. Archives of General Psychiatry, 42, 305-316.
Elkin, I., Shea, M. T., Watkins, J. T., Imber, S. D., Sotsky, S. M., Collins, J. E, Glass, D. R., Pilkonis, P. A.,
Leber, W R., Docherty, 1. P., Fiester, S. J., & ParIoff, M. B. (1989). NIMH treatment of depression
collaborative research program: General effectiveness of treatments. Archives of General Psychiatry,
46, 971-983.
Enns, M. W., & Cox, B. J. (1999). Perfectionism and depression symptom severity in major depressive
disorder. Behaviour Research and Therapy, 37, 783-794.
Enns, M. W, Cox, B. J., & Clara, I. (2002). Adaptive and maladaptive perfectionism: Developmental
origins and association with depression proneness. Personality and Individual Differences, 33, 921-935.
Enns, M. W., Cox, B. J., & Inayatulla, M. (2003). Personality predictors of outcome for adolescents
hospitalized for suicidal ideation. Journal (if the American Academy of Child & Adolescent Psychiatry,
42,720-727.
Enns, M. W., Cox, B. J., & Pidlubny, S. R. (2002). Group cognitive behaviour therapy for residual
depression: Effectiveness and predictors of response. Cognitive Behaviour Therapy, 31, 31-40.
Enns, M. W, Cox, B. J., Sareen, J., & Freeman, P. (2001). Adaptive and maladaptive perfectionism in
medical students: A longitudinal investigation. Medical Education, 35, 1034-1042.
Ewart, C. K., Jorgensen, R. S., & Kolodner, K. B. (1998). Sociotropic cognitions moderates blood pressure
response to interpersonal stress in high-risk adolescent girls. International Journal of
Psychophysiology, 28, 131-142.
Fairbrother, N., & Moretti, M. (1998). Sociotropy, autonomy, and self-discrepancy: Status in depressed,
remitted depressed, and control participants. Cognitive Therapy and Research, 22, 279-296.
Ferguson, K. L., & Rodway, M. R. (1994). Cognitive behavioral treatment of perfectionism: Initial
validation studies. Research on Social Work & Practice, 4, 283-308.
Fertuck, E. A., Bucci, W., Blatt, S. 1., & Ford, R. Q. (2004). Verbal representation and therapeutic change
in anaclitic and introjective patients. Psychotherapy: Theory, Research, Practice, Training, 41, 13-25.
Fichman, L., Koestner, R., & Zuroff, D. C. (1994). Depressive styles in adolescence: Assessment, relation
to social functioning, and developmental trends. Journal of Youth and Adolescence, 23, 315-330.

127
Fichrnan, L., Koestner, R., & Zuroff, D. C. (1996). Dependency, self-criticism, and perceptions of
inferiority at summer camp: I'm even worse than you think. Journal of Youth and Adolescence. 25,
113-126.
Fichman, L.. Koestner, R., & Zuroff, D. C. (1997). Dependency and distress at summer camp. Journal of
Youth and Adolescence. 26, 217-232.
Fichman, L., Koestner, R.. Zuroff, D. c.. & Gordon, L. (1999). Depressive styles and the regulation of
negative affect: A daily experiences study. Cognitive Therapy and Research. 23, 483-495.
Flett, G. L., & Hewitt, P. L. (2002). Perfectionism and maladjustment: An overview of theoretical,
definitional. and treatment issues. In G. L. Flett & P. L. Hewitt (Eds.), Perfectionism. Theory. research.
and treatment (pp. 5-31). New YorkILondon: Guilford Press.
Flett, G. L., Hewitt, P. L.. Garshowitz, M., & Martin, T. R. (1997). Personality. negative social interactions.
and depressive symptoms. Canadian Journal of Behavioural Science. 29. 28-37.
Fonagy, P., Gergely, G.. Jurist, E, & Target, M. (2002). Affect regulation. mentalization, and the
development of the self. New York: Other Press.
Franche, R-L. (200 I). Psychologic and obstetric predictors of couples' grief during pregnancy after
miscarriage or perinatal death. Obstetrics & Gynecology, 97, 597-602.
Franche, R.-L.. & Dobson, K. (1992). Self-criticism and interpersonal dependency as vulnerability factors
to depression. Cognitive Therapy and Research, 16 (4), 419-435.
Franche, R-L., & Mikail, S. E (1999). The impact of perinatal loss on adjustment to subsequent pregnancy.
Social Science & Medicine, 48. 1613-1623.
Frank, E., Kupfer, D. 1., Jacob. M., & Jarrett. D. (1987). Personality features and response to acute
treatment in recurrent depression. Journal of Personality Disorders, 1. 14-26.
Frank, S. J., Van Egeren, L. A.. Paul, J. S., Poorman, M. 0., Sanford, K.. Williams, O. 8., & Field, D. T.
(1997). Measuring self-critical and interpersonal preoccupations in an adolescent inpatient sample.
Psychological Assessment. 9. 185-195.
Fredhoft, T., Poulsen, S., Bauer. M.. & Malm, M. (1996). Dependency and perfectionism: Short term
dynamic group psychotherapy for university students. Psychodynamic Counselling, 24, 476-497.
Fresco, D. M., Sampson, W. S.. Craighead, L. w.. & Koons. A. N. (2001). The relationship of sociotropy
and autonomy to symptoms of depression and anxiety. Journal of Cognitive Psychotherapy, 15,
17-31.
Freud, S. (1958). The dynamics of transference. In J. Strachey (Ed.), The standard edition of the complete
psychological works of Sigmund Freud (Vol. 12, pp. 99-108). London: Hogarth. (Original V10rk
published 1912)
Frost. R. 0., Marten, P.. Lahart, C. M., & Rosenblate, R. (1991). The dimensions of perfectionism.
Cognitive Therapy and Research. 14.449-468.
Gaston, L., Thompson, L., Gallagher, D., Cournoyer, L.-G., & Gagnon, R. (1998). Alliance, technique, and
their interactions in predicting outcome of behavioral. cognitive, and brief dynamic therapy.
Psychotherapy Research, 8, 190-209.
Giordani, C; Wood, J. v., & Michela, J. L. (2000). Depressive personality styles. dysphoria, and social
comparisons in everyday life. Journal of Personality and Social Psychology, 79, 438-451.
Goldberg, J. 0., Segal, Z. V.. Vella. D. D., & Shaw. B. E (1989). Depressive personality: Millon clinical
multiaxial inventory profiles of sociotropic and autonomous subtypes. Journal of Personality
Disorders, 3, 193-198.
Gruen, R. 1., Silva, R, Ehrlich. J.. Schweitzer, 1. w., & Friedhoff. A. J. (1997). Vulnerability to stress: Self-
Criticism and stress-induced changes in biochemistry. Journal of Personality, 65, 33-47.
Gudleski, G. D., & Shean, G. D. (2000). Depressed and nondepressed students: Differences in
interpersonal perceptions. The Journal of Psychology. 134, 56-62.
Haaga, D. A. E, Fine. 1. A.. Terrill. D. R., Stewart, B. L.. & Beck, A. T. (1995). Social problem solving
deficits, dependency, and depressive symptoms. Cognitive Therapy and Research. 16,409-418.
Halgin. R P., & Leahy, P. M. (1989). Understanding and treating perfectionistic college students. Journal
of Counseling & Development, 68, 222-225.
Hammen, C. (1991). Generation of stress in the course of unipolar depression. Journal of Abnormal
Psychology, 100,555-561.
Hammen, C; Ellicott. A.. & Gitlin, M. (1989). Vulnerability to specific life events and prediction of course
of disorder in unipolar depressed patients. Canadian Journal of Behavioural Science. 21, 377-388.
Hammen, C., Ellicott, A., & Gitlin, M. (1992). Stressors and sociotropy/autonomy: A longitudinal study of
their relationship to the course of bipolar disorder. Cognitive Therapy and Research. 16, 409-418.

128
Hammen, C; Ellicott, A., Gitlin, M., & Jamison (1989). Sociotropy/autonomy and vulnerability to specific
life events in patients with unipolar depression and bipolar disorders. Journal ofAbnormal Psychology,
98, 154-160.
Hammen, C., & Goodman-Brown, T. (1990). Self-schemas and vulnerability to specific life stress in
children at risk for depression. Cognitive Therapy and Research, 14,215-227.
Hammen, C, Marks, T, deMayo, R., & Mayol, A. (1985). Self-schemas and risk for depression: A
prospective study. Journal of Personality and Social Psychology, 49, 1147-1159.
Hammen, C., Marks, T, Mayol, A., & deMayo, R. (1985). Depressive self-schemas, life stress, and
vulnerability to depression. Journal ofAbnormal Psychology, 94, 308-319.
Haring, M., Hewitt, P. L., & Flett, G. L. (2003). Perfectionism, coping, and quality of intimate
relationships. Journal of Marriage and Family, 65, 143-158.
Hartlage, S., Arduino, K., & Alloy, L. B. (1998). Depressive personality characteristics: State dependent
concomitants of depressive disorder and traits independent of current depression. Journal ofAbnormal
Psychology, 107, 349-354.
Hellerstein, D. J., Kocsis, J. H., Chapman, D., Stewart, J. W, & Harrison, W (2000). Double-blind
comparision of sertraline, imipramine, and placebo in the treatment of dysthymia: Effects on
personality. American Journal of Psychiatry, 157, 1436-1444.
Hewitt, P. L., & Flett, G. L. (1991). Perfectionism in the self and social contexts: Conceptualization,
assessment, and association with psychopathology. Journal of Personality and Social Psychology, 60,
456-470.
Hewitt, P. L., & Flett, G. L. (1993). Dimensions of perfectionism, daily stress, and depression: A test of a
specific vulnerability hypothesis. Journal ofAbnormal Psychology, 102, 58-65.
Hewitt, P. L., & Flett, G. L. (2002). Perfectionism and stress processes in psychopathology. In G. L. Flett
& P. L. Hewitt (Eds.), Perfectionism. Theory, research, and treatment (pp. 255-284). New York/London:
Guilford Press.
Hewitt, P. L., Flett, G. L., Besser, A., Sherry, S. B., & McGee, B. (2003). Perfectionism is multi-
dimensional: A reply to Shafran, Cooper and Fairburn (2002). Behaviour Research and Therapy, 41,
1221-1236.
Hirsch, C. R., & Hayward, P. (1998). The perfect patient: Cognitive behavioral therapy for perfectionism.
Behavioural and Cognitive Psychotherapy, 26, 359-364.
Hirschfeld, R. M. A., Klerman, G. L., Gough, H. G., Barrett, J., Korchin, S. J., & Chodoff, P. (1977). A
measure of interpersonal dependency. Journal of Personality Assessment, 41, 610-618.
Hirschfeld, R. M. A.. Klerman, G. L., Lavori, P., Keller, M. 8., Griftith, P., & Coryell, W (1989).
Premorbid personality assessment of first onset of major depression. Archives of General Psychiatry,
46, 345-350.
Hong, S., & Lee, M. K. (2001). Hierarchical confirmatory factor analysis of the revised Personal Style
Inventory: Evidence for the multidimensionality problem of perfectionism. Educational and
Psychological Measurement, 61,421-432.
Horvath, A. 0., & Symonds, B. D. (1991). Relation between working alliance and outcome in
psychotherapy: A meta-analysis. Journal of Counseling Psychology, 38, 139-149.
Imber, S. D., Pilkonis, P. A., Sotsky, S. M., Elkin, I., Watkins, 1. T., Collins, J. E, Shea, M. T, Leber, W.
R., & Glass, D. R. (1990). Mode-specific effects among three treatments for depression. Journal of
Consulting and Clinical Psychology, 58, 352-359.
Kash, K. L., Klein, D. N., & Lara, M. E. (2001). A construct validation study of the Response Styles
Questionnaire Rumination Scale in participants with a recent-onset major depressive episode.
Psychological Assessment. 13, 375-383.
Kessler, R. C. (1997). The effects of stressful life events on depression. Annual Review of Psychology, 48,
191-214.
Kiesler, D. J. (1983). The 1982 Interpersonal Circle: A taxonomy for complementarity in human
transactions. Psychological Review, 90,185-214.
Klein, D. N. (1989). The Depressive Experiences Questionnaire: A further evaluation. Journal of
Personality Assessment, 53, 703-715.
Klein, D. N., Harding, K., Taylor, E. B., & Dickstein, S. (1988). Dependency and self-criticism in
depression: Evaluation in a clinical population. Journal ofAbnormal Psychology, 97, 399-404.
Klein, D. N., Taylor, E. B., Dickstein, S., & Harding, K. (1988). Primary early-onset dysthymia:
Comparison with primary nonbipolar nonchronic major depression on demographic, clinical, familial,
personality, and socioenvironmental characteristics and short-term outcome. Journal of Abnormal

129
 Psychology, 97, 387-398.
Koestner, R., Zuroff, D. C; & Powers, T. A. (1991). Family origins of adolescent criticism and its
continuity into adulthood. Journal ofAbnormal Psychology, /00, 191-197.
Krupnick, J. L., Sotsky, S. M., Simmens, S., Moyer, J., Elkin, I., Watkins, J. T., & Pilkonis, P. A. (1996).
The role of therapeutic alliance in psychotherapy and pharmacotherapy outcome: Findings in the
National Institute of Mental Health Treatment of Depression Collaborative Research Program. Journal
of Consulting and Clinical Psychology, 64, 532-539.
Kurdek, L. A. (2000). The link between sociotropy/autonomy and dimensions ofrelationship commitment:
Evidence from gay and lesbian couples. Personal Relationships, 7, 153-164.
Kuyken, w., Kurzer, N., DeRubeis, R. J., Beck, A. T., & Brown, G. K. (2001). Response to cognitive
therapy in depression: The role of maladaptive beliefs and personality disorders. Journal of Consuiting
and Clinical Psychology, 69, 560-566.
Kwon, P., Campbell, D. G., Williams, M. G. (2001). Sociotropy and autonomy: Preliminary evidence for
construct validity using TAT narratives. Journal of Personality Assessment, 77, 128-138.
Kwon, P., & Whisman, M. A. (1998). Sociotropy and autonomy as vulnerabilities to specific life events:
Issues in life event categorization. Cognitive Therapy and Research, 22, 353-362.
Lakey, 8., & Ross, L. T. (1994). Dependency and self-criticism as moderators of interpersonal and
achievement stress: The role of initial dysphoria. Cognitive Therapy and Research, /8,581-599.
Lam, D. H.. Green, B., Power. M. J., & Checkley, S. (1996). Dependency, matching adversities, length of
survival and relapse in major depression. Journal ofAffective Disorders, 37, 81-90.
Lambert, M, 1., & Barley, D. E. (2002). Research summary on the therapeutic relationship and
psychotherapy outcome. In J. C. Norcross (Ed.), Psychotherapy relationships that work (pp. 17-32).
Oxford: Oxford University Press.
Lambert, M, 1., & Ogles, B. M, (2004). The efficacy and effectiveness of psychotherapy. In M. J. Lambert
(Ed.), Bergin and Garfield's handbook ofpsychotherapy and behavior change (5th ed.) (pp. 139-193).
New York: John Wiley & Sons.
Lehman, A. K., Ellis, B., Becker, J., Rosenfarb, L Devine, R., Khan, A.. & Reichler, R. (1997). Personality
and depression: A validation study of the Depressive Experiences Questionnaire. Journal of Personality
Assessment, 68, 197-210.
Little, S. A., & Garber, J. (2000). Interpersonal and achievement orientations and specific stressors
predicting depressive and aggressive symptoms in children. Cognitive Therapy and Research, 24, 651-
670.
Lutz, W. (2002). Patient-focused psychotherapy research and individual treatment progress as a scientific
groundwork for an empirically based clinical practice. Psychotherapy Research, 12, 251-272.
Luyten, P. (2002). Normbesef ell depressie. Aanzet tot een integratief theoretisch kader en een empirisch
onderzoek aan de hand vall de depressietheorie vall S. J. Blatt [Personal standards and depression: A
theoretical framework and an empirical investigation of S. J. Blatt's theory of depression], Unpublisied
doctoral dissertation. University of Leuven, Leuven, Belgium.
Luyten, P., Fontaine, J. R. 1., Soenens, B., Duriez, B., & Corveleyn, J. (2004). What you risk reveals what
you value. Depressive personality styles and value priorities. Manuscript submitted for publication.
Luyten, P., Fontaine, J. R. J., Soenens, B., Meganck, S., Jansen, B., De Grave, C; Corveleyn, J., Maes, F.,
& Sabbe, B. (2004). Factor structure. reliability, and content analysis of the Depressive Experiences
Questionnaire (DEQ) in students, adults and psychiatric inpatients. Manuscript submitted for
publication.
Luyten, P., Van Houdenhove, B., Cosyns, N., & Vandenbroecke, A. (2004). Are patients with Chronic
Fatigue Syndrome perfectionistic - or were they? A case-control study. Manuscript submitted for
publication.
Lynch, T. R., Robins, C. J., & Morse, J. Q. (2001). Couple functioning in depression: The roles of
sociotropy and autonomy. Journal of Clinical Psychology, 57, 93-103.
Mazure, C. M., Bruce, M. L., Maciejewski, P. K" & Jacobs, S. C. (2000). Adverse life events and cognitive-
personality characteristics in the prediction of major depression and antidepressant response. American
Journal of Psychiatry, /57, 896-903.
Mazure, C. M., & Maciejewski, P. K. (2003). A model of risk for major depression: Effects of life stress
and cognitive style vary by age. Depression and Anxiety, 17,26-33.
Mazure, C. M., Maciejewski, P. K., Jacobs, S. C, & Bruce, M. L. (2002). Stressful life events interactng
with cognitive/personality styles to predict late-onset major depression. American Journal of Geriatric
Psychiatry, 10,297-304,

130
Mazure, C. M., Raghavan, C; Maciejewski, P. K., Jacobs, S. C; & Bruce. M. L. (200 I). Cognitive
personality characteristics as direct predictors of unipolar major depression. Cognitive Therapy and
Research, 25, 215-225.
McClelland. G., & Judd, C. (1993). Statistical difficulties of detecting interaction and moderator effects.
Psychological Bulletin, 114. 376-390.
Mongrain, M. (1998). Parental representations and support-seeking behaviors related to dependency and
self-criticism. Journal of Personality, 66, 151-173.
Mongrain, Moo Lubbers. R., & Struthers. W. (2004). The power of love: Mediation of rejection in roommate
relationships of dependents and self-critics. Personality and Social Psychology Bulletin, 30. 94-105.
Mongrain, M., Vettese, L. C; Shuster, B.. & Kendal, N. (1998). Perceptual biases, affect, and behavior in
the relationships of dependents and self-critics. Journal of Personality and Social Psychology, 75, 230-
241.
Mongrain, M.• & Zuroff, D. C. (1989). Cognitive vulnerability to depressed affect in dependent and self-
critical college women. Journal of Personality Disorders, 3, 240-251.
Mongrain, M., & Zuroff, D. C. (1994). Ambivalence over emotional expression and negative life events:
Mediators of depressive symptoms in dependent and self-critical individuals. Personality and
Individual Differences. 16.447-458.
Mongrain, Moo & Zuroff, D. C. (1995). Motivational and affective correlates of dependency and self-
criticism. Personality and Individual Differences, 18,347-354.
Monroe, S. M., & Hadjiyannakis, K. (2002). The social environment and depression: Focusing on severe
life stress. In I. H. Gotlib & C. L. Hammen (Eds.), Handbook of depression (pp. 314-340). New
York/London: The Guilford Press.
Moore. R. G., & Blackburn, I-M. (1996). The stability of sociotropy and autonomy in depressed patients
undergoing treatment. Cognitive Therapy and Research, 20, 69-80.
Morrison, T. L., Urquiza. A. L, & Goodlin-Jones, B. L. (1998). Depressive experiences and romantic
relationships in young adulthood. Psychological Reports, 82, 339-349.
Morse, J. Q., Robins, C. J., & Gittes-Fox, M. (2002). Sociotropy, autonomy. and personality disorder
criteria in psychiatric patients. Journal of Personality Disorders, 16, 549-560.
Mulder, R. T. (2002). Personality pathology and treatment outcome in major depression: A review.
American Journal ofPsychiutry. 159,359-371.
Murphy. B., & Bates. G. W. (1997). Adult attachment style and vulnerability to depression. Personality and
Individual Differences, 22, 835-844.
Nelson. D. Roo Hammen. Coo Daley. S. E.. Burge. D., & Davila, J. (2001). Sociotropic and autonomous
personality styles: Contributions to chronic life stress. Cognitive Therapy and Research. 25, 6 I-76.
Nietzel. M. T.. & Harris. M. J. (1990). Relationship of dependency and achievement/autonomy to
depression. Clinical Psychology Review. 10.279-297.
Nordahl. H. M., & Stiles, T. C. (2000). The specificity of cognitive personality dimensions in cluster C
personality disorders. Behavioural and Cognitive Psychotherapy. 28, 235-246.
O'Connor. R. c., & O'Connor. D. B. (2003). Predicting hopelessness and psychological distress: The role
of perfectionism and coping. Journal (if Counseling Psychology, 50, 362-372.
Ouimette. P. c.. & Klein. D. N. (1993). Convergence of psychoanalytic and cognitive-behavioral theories
of depression: An empirical review and new data on Blatt's and Beck's models. In J. M. Masling & R.
F. Bornstein (Eds.), Psychoanalytic perspectives on psychopathology (Empirical studies of
psychoanalytic theories Vol. 4) (pp, 191-223). Washington, DC: American Psychological Association.
Ouimette. P. C; Klein. D. N; Clark. D. C, & Margolis, E. T. (1992). Personality traits in offspring of
parents with unipolar affective disorder: An exploratory study. Journal "f Personalitv Disorders. 6,
91-98.
Ouimette, P. C; Klein, D. N., & Pepper, C. M. (1996). Personality traits in the first degree relatives of
outpatients with depressive disorders. Journal ofAffective Disorders, 39, 43-53.
Overholser, J. C; & Freiheit, S. R. (1994). Assessment of interpersonal dependency using the Millon
Clinical Multiaxiallnventory-II (MCMI-II) and the Depressive Experiences Questionnaire. Personality
and Individual Differences, 17,71-78.
Peselow, E. D., Robins, C. L Sanfilipo, M. P., Block, P. & Fieve, R. R. (1992). Sociotropy and autonomy:
Relationship to antidepressant drug treatment response and endogenous-nonendogenous dichotomy.
Journal ofAbnormal Psychology. 101,479-486.
Pincus. A. L.. & Gurtman. M. B. (1995). The three faces of interpersonal dependency: Structural analyses
of self-report dependency measures. Journal of Personality and Social Psychology, 69, 744-758.

131
Piper, W. E., Joyce, A. S., McCallum, M.. Azim, H. E, & Ogrodniczuk, J. S. (2001). Interpretive and
supportive psychotherapies: Matching therapy and patient personality. Washington, DC: American
Psychological Association.
Priel, B., & Besser, A. (1999). Vulnerability to postpartum depressive symptomatology: Dependency, self-
criticism and the moderating role of antenatal attachment. Journal of Social and Clinical Psychotogy,
18, 240-253.
Priel, B.. & Besser, A. (2000). Dependency and self-criticism among first-time mothers: The role of global
and specific support. Journal of Social and Clinical Psychology, /9,437-450.
Priel, B., & Shahar, G. (2000). Dependency, self-criticism, social context and distress: Comparing
moderating and mediating models. Personality and Individual Differences, 28, 515-525.
Raghavan, C; Le, H. N., & Berenbaum, H. (2002). Predicting dysphoria and hostility using the diathesis-
stress model of sociotropy and autonomy in a contextualized stress setting. Cognitive Therapy and
Research, 26, 231-244.
Rector, N. A., Bagby, R. M., Segal, Z. v., Joffe, R. T., & Levitt. A. (2000). Self-criticism and dependency
in depressed patients treated with cognitive therapy or pharmacotherapy. Cognitive Therapy and
Research, 24, 571-584.
Rector, N. A., Zuroff, D. C; & Segal, Z. V. (1999). Cognitive change and the therapeutic alliance: The role
of technical and nontechnical factors in cognitive therapy. Psychotherapy, 36, 320-328.
Reda, M. A., Carpiniello. B.. Seccharioli. L., & Blanco, S. (1985). Thinking, depression, and
antidepressants: Modified and unmodified depressive beliefs during treatment with amitripty .ine.
Cognitive Therapy and Research. 9. 135-143.
Reis, S., & Grenyer, B. E S. (2002). Pathways to anaclitic and introjective depression. Psychology and
Psychotherapy: Theory. Research and Practice. 75,445-459.
Reynolds, S., & Gilbert, P. (1991). Psychological impact of unemployment: Interactive effects of
vulnerability and protective factors on depression. Journal of Counseling Psychology, 38, 76-84.
Robins, C. J. (1990). Congruence of personality and life events in depression. Journal of Abnormal
Psychology, 99, 393-397.
Robins, C. J. (1993). Implications of research in the psychopathology of depression for psychotherapy
integration. Journal of Psychotherapy Integration, 3, 313-330.
Robins, C. J. (1995). Personality-event interaction models of depression. European Journal of Personality,
9,367-378.
Robins, C. J.. & Block, P. (1988). Personal vulnerability. life events, and depressive symptoms: A test of a
specific interactional model. Journal of Personality and Social Psychology, 54, 847-852.
Robins, C. J., Hayes, A. M.. Block. P.. Kramer. R. J.. & Villena. M. (1995). Interpersonal and achievement
concerns and the depressive vulnerability and symptom specificity hypotheses: A prospective study.
Cognitive Therapy and Research. /9. 1-20.
Robins, C. J., Ladd, J., Welkowitz, 1.. Blaney. P. H.. Diaz, R., & Kutcher, G. (1994). The Personal Style
Inventory: Preliminary validation studies of new measures of sociotropy and autonomy. Journal of
Psychopathology and Behavioral Assessment, 16.277-300.
Rogers, C. R. (1957). The necessary and sufficient conditions of therapeutic personality change. Journal
of Consulting Psychology. 22. 95-103.
Rosenfarb, I. S., Becker. J.. Khan. A.. & Mintz. J. ( 1994). Dependency. self-criticism, and perception, of
socialization experiences. Journal ofAbnormal Psychology, /03.669-675.
Rosenfarb, I. S., Becker. J.. Khan. A.. & Mintz. J. (1998). Dependency and self-criticism in bipolar and
unipolar depressed women. British Journal of Clinical Psychology. 37, 409-414.
Rude, S. S., & Burnham, B. L. (1993). Do interpersonal and achievement vulnerabilities interact with
congruent events to predict depression'! Comparison of DEQ, SAS, DAS, and combined scales.
Cognitive Therapy and Research. /7. 531-548.
Rude, S. S., & Burnham, B. L. (1995). Connectedness and neediness: Factors of the DEQ and SAS
dependency scales. Cognitive Therapy and Research, /9, 323-340.
Rutter, M., Dunn, J., Plomin, R., Simonoff, E., Pickles, A., Maughan, B., Ormel, J., Meyer, J., & Eaves,
L. (1997). Integrating nature and nurture: Implications of person-environment correlations
and interactions for developmental psychopathology. Development and Psychopathology, 9, 335-
364.
Saboonchi, G., & Lundh, L.-G. (2003). Perfectionism, anger, somatic health, and positive affect.
Personality and Individual Differences, 35, 1585-1599.
Safran, J. D. (1990). Towards a refinement of cognitive therapy in light of interpersonal theory: I. Theory.

132
 Clinical Psychology Review, 10,87-105.
Sahin, N., Ulusoy, M., & Sahin, N. (1993). Exploring the sociotropy-autonomy dimensions in a sample of
Turkish psychiatric inpatients. Journal of Clinical Psychology, 49, 751-763.
Sakado, K, Sato, T.. Uehara, T, Sakado. M., Kuwabara, H., & Someya, T. (1999). The association between
the high interpersonal sensitivity type of personality and a life time history of depression in a sample
of employed Japanese adults. Psychological Medicine, 29, 1243-1248.
Sanathara, V. A.• Gardner, C. 0 .. Prescott, C. A., & Kendler, K. S. (2003). Interpersonal dependence and
major depression: Aetiological inter-relationships and gender differences. Psychological Medicine. 33,
927-931.
Santor. D. A., & Patterson. R. L. (2003). Frequency and duration of mood fluctuations: Effects of
dependency, self-criticism. and negative events. Manuscript submitted for publication.
Santor, D. A., Pringle. J. D., & Israeli. A. L. (2000). Enhancing and disrupting cooperative behavior in
couples: Effects of dependency and self-criticism following favorable and unfavorable performance
feedback. Cognitive Therapy and Research. 24, 379-397.
Santor, D. A.• & Yazbeck, A. A. (2003). Soliciting unfavorable social comparisons: Effects of self-criticism.
Manuscript submitted for publication.
Santor. D. A., & Zuroff, D. C. (1997). Interpersonal responses to threats to status and interpersonal
relatedness: Effects of dependency and self-criticism. British Journal of Clinical Psychology, 36. 521-
541.
Santor. D. A., & Zuroff, D. C. (1998). Controlling shared resources: Effects of dependency, self-criticism.
and threats to self-worth. Personality and Individual Differences, 24, 237-252.
Sato, T, & McCann, D. (1997). Vulnerability factors in depression: The facets of sociotropy and autonomy.
Journal of Psychopathology and Behavioral Assessment, 19,41-62.
Sauro, M. D., Jorgensen, R. S.. Larson, C. A.. Frankowski, J. J., Ewart. C. K., & White, J. (2001).
Sociotropic cognition moderates stress-induced cardiovascular responsiveness in college women.
Journal (If Behavioral Medicine. 24. 423-439.
Schachter, E. P., & Zlotogorski, Z. (1995). Self-critical and dependent aspects of loneliness. Israel Journal
of Psychiatry and Related Sciences, 32. 205-211.
Segal. Z. v., & Ingram. R. E. (1994). Mood priming and construct activation in tests of cognitive
vulnerability to unipolar depression. Clinical Psychology Review, 14, 663-695.
Segal. Z. V.. Shaw, B. E. & Vella. D. D. (1989). Life stress and depression: A test of the congruency
hypothesis for life event and depressive subtype. Canadian Journal of Behavioural Science, 21. 389-
400.
Segal, Z. V.. Shaw. B. E, Vella, D. 0 .. & Katz, R. (1992). Cognitive and life stress predictors of relapse in
remitted unipolar depressed patients: Test of the congruency hypothesis. Journal of Abnormal
Psychology, /01,26-36.
Shafran, R.. & Mansell. W. (2001). Perfectionism and psychopathology: A review of research and
treatment. Clinical Psvchology Review, 21. 879-906.
Shahar. G., Blatt. S. J.. Zuroff, D. C .. & Pilkonis, P. A. (2003). Role of perfectionism and personality
disorder features in response to brief treatment for depression. Journal of Consulting and Clinical
Psychology. 71, 629-633.
Shahar, G .. Blatt. S. J., Zuroff, D. c.. Krupnick, J. L., & Sotsky. S. M. (2004). Perfectionism impedes social
relations and response to brief treatment for depression. Journal ofSocial and Clinical Psychology. 23,
140-154.
Shahar, G.. Blatt, S. J., Zuroff. D. C .. Kuperrninc, G. P.. & Leadbeater, B. J. (2004). Reciprocal relations
between depressive symptoms and self-criticism (but not dependency) among early adolescent girls
(but not boys). Cognitive Therapy and Research, 28, 85-103.
Shahar, G.. Henrich, c.. Blatt, S. 1., Ryan, R.. & Little, T. D. (2003). Interpersonal relatedness, self-
definition, and their motivational orientation during adolescence: A theoretical and empirical
integration. Developmental Psychology, 39, 470-483.
Shahar, G., Joiner. T E.. Zuroff, D. C.. & Blatt, S. 1. (2004). Personality. interpersonal behavior, and
depression: Co-existence of stress-specific moderating and mediating effects. Personality and
Individual Differences, 36, 1583-1596.
Shahar, G.. & Priel, B. (2003). Active vulnerability, adolescent distress. and the mediating/suppressing role
of life events. Personality and Individual Differences, 35, 199-218.
Shapiro, J. P. (1988a). Relationships between dimensions of depressive experience and perceptions of the
lives of people in general. Journal of Personality Assessment, 52, 297-308.

133
Shapiro, J. P. (l988b). Relationships between dimensions of depressive experience and evaluative beliefs
about people in general. Personality and Social Psychology Bulletin, 14, 388-400.
Shea, M. T., Pilkonis, P. A., Beckham, E.• Collins, J. E, Elkin, I., Sotsky, S. M .• & Docherty, 1. P. (1990).
Personality disorders and treatment outcome in the NIMH Treatment of Depression Collaborative
Research Program. American Journal of Psychiatry. 147,711-718.
Sherry, S. B., Hewitt. P. L., Flett, G. L.. & Harvey, M. (2003). Perfectionism dimensions. perfectionistic
attitudes, dependent attitudes. and depression in psychiatric patients and university students. Journal of
Counseling Psychology, 50, 373-386.
Simons, A. D., Angell, K. L., Monroe, S. M., & Thase, M. E. (1993). Cognition and life stress in
depression: Cognitive factors and the definition, rating, and generation of negative life events. Journal
ofAbnormal Psychology, 102.584-591.
Smith. T. w., O'Keeffe, J. L.. & Jenkins, M. (1988). Dependency and self-criticism: Correlates of
depression or moderators of the effects of stressful events"! Journal of Personality Disorders, 2, 160-
169.
Solomon, A., Arnow, B. A.• Gotlib, I. H. & Wind, B. (2003). Individualized measurement of irrational
beliefs in remitted depressives. Journal of Clinical Psychology. 59, 439-455.
Solomon, A., Haaga, D. A. E. Brody. C; Kirk. L., & Friedman, D. G. (1998). Priming irrational beliefs in
recovered-depressed people. Journal (if Abnormal Psychology, /07,440-449.
Spasojevic, J., & Alloy, L. B. (2001). Rumination as a common mechanism relating depressive risk factors
to depression. Emotion. 1.25-37.
Spitzer. R., Endicott, J.• & Robins. C. J. (1978). Research Diagnostic Criteria: Rationality and reliability.
Archives of General Psychiatry, 35, 773-782.
Tennant, C. (2002). Life events, stress and depression: A review of recent findings. Australian and New
Zealand Journal of Psychiatry, 36, 173-182.
Vettese, L. C., & Mongrain, M. (2000). Communication about the self and partner in relationships of
dependents and self-critics. Cognitive Therapy and Research, 24, 609-626.
Vogel, P. A.• Stiles, T. C., & Nordahl, H. M. (2000). Cognitive personality styles in OCD outpat.ents
compared to depressed outpatients and healthy controls. Behavioural and Cognitive Psychotherapy, 28,
247-258.
Voyer. M.• & Cappeliez, P. (2002). Congruency between depressogenic schemas and life events for the
prediction of depressive relapse in remitted older patients. Behavioural and Cognitive Psychotherapy,
30, 165-177.
Wachtel, P. L. (1994). Cyclical processes in personality and psychopathology. Journal of Abnormal
Psychology. 103, 51-66.
Wampold, B. E. (1997). Methodological problems in identifying efficacious psychotherapies.
Psychotherapy Research, 7. 21-43.
Westen, D., & Morrison. K. (200 I). A multidimensional meta-analysis of treatments for depression, panic,
and generalized anxiety disorder: An empirical examination of the status of empirically supported
therapies. Journal of Consulting and Clinical Psychologv, 69, 875-899.
Whiffen, V. E.• & Aube, J. A. ( 1999). Personality, interpersonal context and depression in couples. Journal
of Social and Personal Relationships. /6, 369-383.
Whiffen, V. E.• Aube, J. A.. Thompson. J. Moo & Campbell, T. L. (2000). Attachment beliefs and inter-
personal contexts associated with Dependency and Self-Criticism. Journal (if Social and Clinical
Psychology, 19, 184-205.
Whisman, M. A., & Friedman. M. A. (1998). Interpersonal problem behaviors associated with lys-
functional attitudes. Cognitive Therapy and Research, 22, 149-160.
Wilhelm, K., Boyce, P. & Brownhill, S. (2004). The relationship between interpersonal sensitivity, anxiety
disorders and major depression. Journal ofAffective Disorders, 79, 33-41.
Wiseman, H. (1997). Interpersonal relatedness and self-definition of loneliness during the transition to
university. Personal Relationships, 4. 285-299.
Woodside, D. 8., Bullik, C. Moo Halmi, K. A.. Fichter, M. Moo Kaplan. A.. Berrettini, W. H., Strober, M.,
Treasure. 1., Lilenfeld, L.. Klump. K., & Kaye, W. H. (2002). Personality, perfectionism, and attitudes
toward eating in parents of individuals with eating disorders. International Journal of Eating Disorders,
31.290-299.
Zettle, R. D., Haflich, J. L.. & Reynolds, R. A. ( 1992). Responsivity to cognitive therapy as a function of
treatment format and client personality dimensions. Journal of Clinical Psychology, 48, 787-797.
Zettle, R. D., & Herring. E. L. (1995). Treatment utility of the sociotropy/autonomy distinction:

134
 Implications for cognitive therapy. Journal of Clinical Psychology, 51,280-289.
Zuroff, D. C (1992). New directions for cognitive models of depression. Psychological Inquiry, 3, 274-
277.
Zuroff, D. C, & Blatt, S. 1. (2002). Vicissitudes of life after the short-term treatment of depression: Roles
of stress, social support, and personality. Journal of Social and Clinical Psychology, 21, 473-496.
Zuroff, D. C. Blatt, S. 1., Sanislow III, C A, Bondi, eM., & Pilkonis, P. A. (1999). Vulnerability to
depression: Reexamining state dependence and relative stability. Journal ofAbnormal Psychology, 108,
76-89.
Zuroff, D. C., Blatt, S. J., Sotsky, S. M., Krupnick, J. L., Martin, D. J., Sanislow III, CA., & Simmens, S.
(2000). Relation of therapeutic alliance and perfectionism to outcome in brief outpatient treatment of
depression. Journal of Consulting and Clinical Psychology, 68, 1[4-124.
Zuroff, D. C, & de Lorimier, S. (1989). Ideal and actual romantic partners of women varying in
dependency and self-criticism. Journal of Personality, 57, 825-846.
Zuroff, D. C, & Duncan, N. (1999). Self-criticism and conflict resolution in romantic couples. Canadian
Journal of Behavioural Science, 31,137-149.
Zuroff, D. C, & Fitzpatrick, D. K. (1995). Depressive personality styles: Implications for adult attachment.
Personality and Individual Differences, 18,253-265.
Zuroff, D. C., lgreja, I., & Mongrain, M. (1990). Dysfunctional attitudes, dependency, and self-criticism as
predictors of depressive mood states: A 12-month longitudinal study. Cognitive Therapy and Research,
14,315-326.
Zuroff, D. C., Koestner, R., & Powers, T. A. (1994). Self-criticism at age 12: A longitudinal study of
adjustment. Cognitive Therapy and Research, 18,367-385.
Zuroff, D. C., & Mongrain, M. (1987). Dependency and self-criticism: Vulnerability factors for depressive
affective states. Journal ofAbnormal Psychology, 96, 14-22.
Zuroff, D. C., Moskowitz, D. S., & Cote, S. (1999). Dependency, self-criticism, interpersonal behaviour
and affect: Evolutionary perspectives. British Journal of Clinical Psychology. 38. 231-250.
Zuroff, D. C, Moskowitz, D. S.. Wielgus, M. S., Powers, T. A., & Franko. D. L. (1983). Construct
validation of the Dependency and Self-Criticism scales of the Depressive Experiences Questionnaire.
Journal of Research in Personality, 17, 226-241.
Zuroff, D. C, Stotland, S.. Sweetman, E., Craig, J.-A., & Koestner, R. (1995). Dependency, self-criticism
and social interactions. British Journal of Clinical Psychology, 34, 543-553.

135
 Chapter 5

A Dialectic Model of Personality Development

and Psychopathology: Recent Contributions

to Understanding and Treating Depression

Sidney J. Blatt and Golan Shahar

Personality Development and Psychopathology

Psychological development is often portrayed as a linear, monotonic function,

with each developmental phase following and deriving from a prior phase. We
believe that psychological development, however, is complex and more aptly
described as a dialectic process in which development occurs reciprocally across
various developmental lines, such that development in one domain facilitates
development in parallel but contrasting domains that then, in turn, facilitate further
development in the original domain. The development of concepts of self and of
significant others is an example of this reciprocal development, such that a more
differentiated relationship with an other contributes to further differentiation within
the self and, conversely, further differentiation within the self leads to further
differentiation in relationships with others. Self and other are reciprocal constructs
that develop in a mutually facilitating dialectic transaction.
 Blatt and colleagues (Blatt & Shichman, 1983; Blatt & Blass, 1990, 1996) have
articulated this fundamental dialectic between the development of the self and
relatedness to others. Based on an elaboration and extension of Erikson's (1950)
epigenetic psychosocial developmental model, Blatt and colleagues (e.g., Blatt &
Shichman, 1983) identified three levels in the development of interpersonal
relatedness: basic trust (vs. mistrust), cooperation (vs. alienation), and intimacy (vs.
isolation) which they coordinate with aspects of Erikson's model that they
identified as pertinent to the development of the self: autonomy (vs. shame),
initiative (vs. guilt), industry (vs. inferiority), identity (vs. role diffusion), genera-
tivity (vs. stagnation), and integrity (vs. despair). Blatt and Blass (1990, 1996), in
an elaboration of this dialectic developmental model, noted that phases in the
development of relatedness (trust, cooperation, and intimacy) are juxtaposed in
Erikson's model with their polar attributes (mistrust, alienation, and isolation,
respectively). The negative counterparts of each phase in the development of self-
definition (e.g., shame, guilt, inferiority), however, are not polar attributes; rather
they reflect a more complex developmental process involving both behavioral and
experiential components. Erikson identified the negative experiential or affective
component of each phase in the development of self-definition (or identity) such
that, for example, the counterpart of a failure to develop autonomy is not a lack of
autonomy, but shame. Likewise, the experiential or affective component of
difficulty in the development of initiative is not a lack of initiative, but guilt. Thus,
difficulty in the development of aspects of self-definition can be expressed in
affective (experiential) dimensions (e.g., shame, guilt, as specified by Erikson) as
well as in behavioral dimensions such as a lack of autonomy, a lack of initiative, or
a lack of industry (Blatt & Blass, 1990, 1996).
For development to proceed in the relatedness dimension from a sense of trust to
a capacity for cooperation, first in the family and later with peers, the child has to
develop a behavioral capacity for autonomy (or separateness) with a sense of pride
(or a lack of shame), as well as a behavioral capacity for initiation without feelings
of guilt, and a capacity for industry free of feelings of inferiority. Behavioral and
affective achievements in these early phases in the development of self enable the
child to establish relationships with others in new ways. The process of moving
from a sense of basic trust with a caregiving other to an emergent capacity for
reciprocal relatedness - to a sense of cooperation (Blatt & Blass, 1992, 1996) -
depends on the child's ability to establish a sense of self as autonomous and capable
of initiation and industry that facilitates the child's beginning to be able to
participate in cooperative relationships (e.g., first in parallel, then interactive, and,
eventually, in reciprocal play). And the development of a capacity to participate in
cooperative and collaborative relationships, in tum, leads to further refinement of
the self. Eventually, with the development of a sense of identity, one has the
capacity to form an intimate reciprocal relationship based on the recognition of
what one wants to share in a relationship with an appreciation of what one brings
to and what one wants from an enduring intimate relationship. Thus, the sense: of
self and the quality of interpersonal relatedness develop in a reciprocal, mutually
facilitating, dialectic transaction. These two dimensions normally develop in a
reciprocal dialectic process, such that development in one dimension facilitates

138
development in the other. Throughout life, from infancy to senescence, develop-
ment in the sense of self leads to refinements in the capacity for interpersonal
relatedness, and development in the quality of interpersonal relationships enriches
the sense of self (Blatt & Blass, 1990, 1996; Blatt & Shichman, 1983).
As noted in Chapter 3, these processes define two fundamental dimensions in
personality development:
(a) development of a differentiated, integrated, realistic, essentially positive sense
of self or an identity, and
(b) development of the capacity to form mutually satisfying, reciprocal, inter-
personal relationships.

These two dimensions are central in theoretical formulations in psychoanalytic

theory (e.g., Freud, 1914/1958, 1926/1959, 1930/1961; Loewald, 1962), and in
general conceptual (e.g., Angyal, 1941, 1951; Bakan, 1966), and empirical ap-
proaches (e.g., Benjamin, 1974; Wiggins, 1991) to personality theory.
Relatedness and self-definition are central dimensions not only in normal
personality development, but they are also central issues in many forms of
psychopathology. Most forms of psychopathology emerge from disruptions, at
different developmental levels, of this normal developmental dialectic process in
the development of self and of a capacity for interpersonal relatedness. Disruptions
at different points in this normal dialectic developmental process are expressed in
different ways in various forms of psychopathology (e.g., Blatt, 1990; Blatt &
Shichman, 1983), including depression. Thus, most forms of psychopathology
involve either a one-sided, distorted, intense preoccupation, at different develop-
mental levels, with preserving a sense of self or a one-sided, distorted preoccupa-
tion, at different developmental levels, with maintaining interpersonal relatedness.
Some individuals, more often women, deal with developmental disruptions by
becoming preoccupied with interpersonal relationships at the expense of the
development of a sense of self. Other individuals, more often men, deal with
developmental disruptions by becoming preoccupied with the issues of self-
definition at the expense of relatedness. Thus, two basic configurations of
psychopathology can be identified: I) an anaclitic or relatedness configuration of
psychopathology that includes the dependent or infantile personality disorder,
anaclitic depression, and the hysterical personality disorder - disorders preoccu-
pied with issues of interpersonal relatedness (e.g., trust, dependability, affec-
tion, and love) at various developmental levels; and 2) an introjective or
self-definitional configuration of psychopathology that includes the para-
noid, obsessive-compulsive, depressive, and phallic narcissistic personality
disorders - disorders preoccupied with issues of self-definition (e.g., autonomy,
self-control, and self-worth) at different developmental levels (Blatt & Shichman,
1983).
In these formulations, various forms of psychopathology, including depression,
are not viewed as separate diseases that derive from presumed, but unspecified and
undocumented, biological disturbances, but rather maladaptive modes that

139
individuals develop to deal with severe and persistent disruptions in psychological
development. This approach assumes that most biological dimensions create
nonspecific vulnerabilities that can gain expression in multiple ways in the
vicissitudes of the interpersonal care-giving relationships between parents and
child. Extensive research of normal psychological development provides a well-
delineated developmental framework for identifying deviations or distortions of
normal developmental processes that derive from the interaction of biological
vulnerabilities with disrupted parent-child interaction and lead to a variety of
psychological disturbances (e.g., Blatt, 1995a; Blatt & Homann, 1992; Goodman &
Gotlib, 2002). Freud's insights into melancholia (Freud, 1917/1957), for example,
were derived from his contrast of aspects of clinical depression with normal loss
and grieving (mourning). Freud's contrast of pathology with normality enabled him
to identify central elements in clinical depression. This approach of defining
psychopathology as deriving from disruptions of normal psychological
development has important potential advantages including avoiding categoncal
distinctions of psychopathology based on manifest symptoms, maintaining
continuity between normal psychological development and various forms of
psychopathology, establishing relationships among various forms of psy-
chopathology to deal with complex issues of cornorbidity, and of identifying
dimensions relevant to various disorders beyond their manifest symptomatic
expressions.

The Understanding and Treatment of Depression

Dissatisfaction with symptom-based diagnostic classifications of depression has

led several investigators to differentiate types of depression based on a
differentiation among the fundamental developmental concerns that contribute to
individuals becoming depressed. Our research team (e.g., Blatt, 1974; Blatt,
D' Afflitti, & Quinlan, 1976) identified two major foci in depression: (I) disruptions
of gratifying interpersonal relationships (e.g., object loss or neglect) and (2)
disruptions of an effective and essentially positive sense of self (e.g., feelings of
failure or guilt). Extensive clinical and empirical evidence indicates that depression
is organized around these two issues - around interpersonal issues such as feelings
of abandonment and loneliness, or around an impaired sense of self expressed in
preoccupations with personal failure, inadequacy, or transgression (Blatt, 2004;
Blatt & Zuroff, 1992). These two themes (abandonment and failure) differentiate
two major foci around which depression can be organized: (I) depressive
experiences often precipitated by object loss that we have called anaclitic (Blatt,
1974) or dependent (Blatt et al., 1976, Blatt, Quinlan, Chevron, McDonald, &
Zuroff, 1982) and (2) introjective (Blatt, 1974) or self-critical (Blatt et aI., 1976,
1982) depressive experiences precipitated by feelings of failure, guilt, and other
impairments of self-worth.

140
 Based on the assumption that psychopathology is most effectively considered,
not as separate diseases deriving from unspecified and undocumented biological
disturbances (e.g., such as a clinically undocumented chemical imbalance in
depression), but as disruptions of normal psychological development, we (Blatt et
aI., 1976) reviewed the classic psychoanalytic literature on depression (e.g., Freud's
Mourning and Melancholia [1917/1957], Edward Bibring's Mechanisms of
Depression [1953), Mabel Blake Cohen and colleagues' An Intensive Study of
Twelve Cases of Manic Depression Psychosis [1954 D, and gleaned from these
reports, not symptoms of depression, but examples of the everyday life experiences
of depressed individuals and their families. We eventually identified 66 items from
this literature - items that tapped issues like a distorted or depreciated sense of self
and others, dependency, helplessness, egocentricity, fear of loss, ambivalence,
difficulty dealing with anger, self-blame, guilt, loss of autonomy, and distortions of
family relations. These items were not selected from any particular theoretical
position on depression, but because they represented a wide range of experiences
that were characteristic of the lives of depressed individuals. These 66 items were
initially administered to a large sample of college students with the instructions to
rate, on a 7-point scale, the degree to which each item characterized their
experiences. Based on Principal Components Analysis with a varimax rotation, we
found that this questionnaire, which we called the Depressive Experiences
Questionnaire (DEQ), was composed of three primary factors (Blatt et al., 1976): a
factor containing items primarily focused on issues of loneliness and abandonment
(e.g., "I often think about the danger of losing someone who is close to me"),
labeled Dependency, 2) a factor comprised of items focused on issues of self-
definition and self-worth (e.g., "There is considerable difference between how I
am now and how I would like to be"), labeled Self-Criticism, and an Efficacy
factor comprised of items reflecting feelings of resilience, competence and
personal strength (e. g., "I have many inner resources"). Thus, we serendipitously
developed an empirical method (the DEQ; Blatt et al., 1976) that reliably and
systematically assesses the two primary sources or foci of depression that Freud
(1917/1957) had tried to integrate into a unified conceptualization of depression in
Mourning and Melancholia: a traditional source of depression derived from
superego issues of self-worth, self-blame and guilt, and a source of depression
derived from interpersonal concerns of loneliness and fears of abandonment (Blatt,
1974).
A wide range of empirical studies over the past three decades have demonstrated
that the three factors of the DEQ are highly stable; they have been replicated
in clinical and non-clinical samples of both adolescents and adults in a number
of different cultures (e.g., Beutel et al., 2004; Compas, 2002; Frank et aI., 1997;
Jae Irn, 1998; Luyten, 2002; Priel, Besser, & Shahar, 1998; Zuroff, Quinlan,
& Blatt, 1990). These studies consistently indicated that the 66 items of the
DEQ cluster into three primary factors. The first factor, Dependency, reflects
wishes to be cared for, loved and protected, and fear of being abandoned. The
second factor, Self-Criticism or Self-Critical Perfectionism, taps preoccupation
with achievement and feelings of inferiority and guilt in the face of perceived
failure to meet standards. The third factor, Efficacy, represents personal resilience

141
and inner strength. A wide range of research (see summary in Blatt, 2004) has
demonstrated that the two primary factors of the DEQ (the Interpersonal or Depen-
dency and the Self-Criticism factors) are relatively independent dimensions of
depression that provide understanding of some of the distal as well as proximal
factors that contribute to the onset of depression as well as of various expressions
of depression in a wide range of clinical and non-clinical samples in a number of
cultures.
The empirical confirmation of these two primary dimensions of depressive
experiences is consistent with clinical-theoretical formulations about the nature of
depression from psychoanalytic (e.g., Arieti & Bemporad, 1978, 1980; Blatt, 1974;
Bowlby, 1980) and cognitive-behavioral (Beck, 1983) perspectives. Arieti and
Bemporad (1978, 1980) differentiated a depression associated with a preoccupation
with the loss of a "dominant other" from a depression provoked by a failure to
achieve a "dominant goal," Bowlby (1980) discussed depression as associated with
either an anxious attachment or a compulsively self-reliant attachment style. From
a cognitive-behavioral perspective, Aaron Beck (1983) articulated a similar
distinction in his differentiation of a sociotropic depression, a depression focused
around disruptions of interpersonal relationships, from an autonomous depress:ion
focused around disruptions of autonomy and self-definition. Thus several groups of
psychoanalytic theorists and a preeminent cognitive-behavioral theorist agree about
the importance of differentiating between a depression focused on interpersonal
issues and a depression focused on issues of self-worth and self-definition (Blatt &
Maroudas, 1992).
Several other procedures, in addition to the DEQ, have been developed that
assess these two dimensions of depression. Aaron Beck (1983) developed an
assessment method similar to the DEQ, the Sociotropy-Autonomy Scale (SAS;
Beck, 1983), and several other investigators (e.g., Cane, Olinger, Gotlib, & Kuiper,
1986) found that the Dysfunctional Attitudes Scale (DAS; Weissman & Beck,
1978) also assesses these two dimensions of depression. Robins and Luten (1991)
combined elements from the DEQ and the SAS to create the Personal Style
Inventory (PSI) to assess these two dimensions of depression in an attempt to
develop a scale that might have improved psychometric properties (Robins et al.,
1994). Research with all of these scales (e.g., DEQ, SAS, DAS, PSI) has resulted
in an impressive extensive research literature that has led to a fuller understanding
of some of the early and contemporary life experiences that contribute to the
occurrence of depression and of important differences in the clinical expression and
treatment of anaclitic (dependent) and introjective (self-critical) types of depression
(e.g., Blatt, 2004; Blatt & Homann, 1992; Blatt & Zuroff, 1992; Clark & Beck,
1999; see also Chapter 4, this volume).
In addition to contributing to clarifying aspects of the etiology, clinical
expressions, and treatment of depression, the development of the DEQ has
contributed to differentiating several developmental levels in the evolution of self-
definition and of the capacity for interpersonal relatedness. As discussed earlier,
interpersonal relatedness evolves from feelings of trust, to a capacity for
cooperation and collaboration, to a capacity for intimacy. Likewise, self-definition
or identity evolves from reactive feelings of autonomy, to proactive capacities for

142
initiative and industry, to the formation of a consolidated identity and a capacity for
generativity, and ultimately to feelings of integrity.

Developmental Levels of Interpersonal Relatedness and Self-Definition

As noted earlier, interpersonal relatedness and self-definition are central to

personality development. Personality development evolves out of a dynamic
dialectic tension between these two fundamental developmental processes. An
increasingly differentiated, integrated, and mature sense of self is contingent on
establishing satisfying interpersonal relationships and, conversely, the continued
development of increasingly mature and satisfying interpersonal relationships is
contingent on the development of a mature self-concept or identity. In normal
personality development, these two developmental processes evolve in an
interactive, reciprocally balanced, mutually facilitating fashion from birth through
senescence (Blatt, 1990, 1995a; Blatt & Blass, 1990, 1992, 1996; Blatt &
Shichman, 1983). If development does not proceed well, however, interpersonal
relatedness can be characterized by feelings of mistrust about the dependability of
others, by an alienation from others, or by feelings of loneliness and isolation.
Likewise, disruptions in the development of a sense of self can involve fears of
annihilation; concerns about a loss of control; and feelings of shame, guilt and
inferiority; and of a lack of integrity.
Recent research with the DEQ (e.g., Blatt, Zohar, Quinlan, Zuroff, & Mongrain,
1995; Blatt, Zohar, Quinlan, Luthar, & Hart, 1996; Blatt & Shahar, 2003) has
differentiated empirically two levels of interpersonal relatedness: (I) a maladaptive
desperate neediness, in which one feels terrified about abandonment and threatened
by feelings of helplessness when alone, and (2) a more adaptive relatedness, in
which one has meaningful involvement with particular individuals and feels a sense
of sadness and loss if these relationships are disrupted. Other recent research with
the DEQ (Kuperminc, Blatt, & Leadbeater, 1997) has also identified empirically
two levels in the development of the self - a maladaptive, reflective, ruminative,
self-critical, evaluative sense of self and an adaptive, proactive, efficacious sense of
self. Self-evaluative ruminative concerns about self-worth include feelings of a loss
of autonomy, initiative, and industry, and feelings of shame, guilt, and inferiority.
These concerns and doubts about self-worth are orthogonal to, and independent of,
the second level in the development of the self - feelings of efficacy, identity,
integrity, and purpose.

143
Two Levels of Interpersonal Relatedness

As research with the DEQ has progressed, it has become evident that Factor 2,
the Self-Criticism factor, was more effective than Factor I, the Dependency factor,
in predicting depressive symptoms and a host of adaptive and maladaptive
outcomes (Blatt et al., 1982; Nietzel & Harris, 1990; Priel & Shahar, 2000). We
(Blatt, Zohar, et al., 1995, 1996) examined Factor I of the DEQ with the hypothesis
that the Dependency factor of the DEQ may actually include two facets or
subfactors, one that assesses adaptive interpersonal relatedness and the other that
assesses a more maladaptive dependence or neediness. Using Small Space Analysis
(SSA), a theory-driven clustering procedure developed by the research
methodologist Louis Guttman, we identified these two facets within the
Dependency factor (Factor I) in both adults and adolescents (Blatt, Zohar et .11.,
1995, 1996; Henrich, Blatt, Kuperminc, Zohar, & Leadbeater, 200 I; see also Rude
& Burnham, 1995).
One facet, labeled relatedness, includes items that assess feelings of loss,
sadness, and loneliness in reaction to disruption of relationships with a particular
person (e.g., "I would feel like I'd be losing an important part of myself if I lost a
very close friend"). These feelings are not undifferentiated and non-specific; rather,
they reflect concerns about the loss of a special person to whom one feels attached.
These items do not reflect feeling helpless without this relationship, but rather
feeling that this particular relationship is valued and therefore loss is accompanied
by feelings of sadness and loss. The second facet, labeled dependence or neediness,
includes items expressing feelings of helplessness, fears and apprehension about
separation and rejection. and intense and broad-ranging concerns about a general
loss of contact with others, unrelated to a particular relationship (e.g., "I become
frightened when I am alone"). These items reflect a desperate need for others but
with little differentiation or specification of any particular person or relationship.
The primary theme of these items is an intense fear of abandonment and feelings of
helplessness. Hence, the results of Small Space Analyses with adolescent and adults
indicate that the Dependency or Interpersonal Factor of the DEQ is comprised of
two sets of items that assess both adaptive and maladaptive manifestations of
interpersonal relatedness.
When the two facets of dependency (i.e., relatedness and dependence or
neediness) were used to predict depressive symptoms and psychological well-
being, two interesting patterns emerged. First, the dependence or neediness facet
had significantly higher correlations with measures of depression while the
relatedness facet had significantly higher correlations with measures of
psychological well-being (Blatt, et al. 1995, 1996; Henrich et al., 200 I). Second.
the intercorrelations between the four DEQ variables (i.e., self-criticism, efficacy.
dependence or neediness, and relatedness) no longer reflect the orthogonality found
among the three original DEQ factors (i.e., Dependency, Self-criticism and
Efficacy). Specifically, Self-criticism correlated positively with dependence or
neediness and negatively with relatedness (e.g., r =.19 and -.10; respectively; Blatt,

144
et aI., 1996). That is, self-criticism, a maladaptive sense of self, correlated
positively with the maladaptive dimension of interpersonal relatedness (neediness),
and negatively with the adaptive dimension of interpersonal relations (relatedness).
In contrast, Efficacy, an adaptive sense of self, correlated positively with
relatedness and negatively with dependence or neediness (e.g., r = .16 and -.23;
respectively) (Blatt, et al. 1996).

Two Levels of Self-Definition

The differential correlations of the DEQ Self-Criticism and Efficacy factors with
the maladaptive and adaptive dimensions of interpersonal relationships
(i.e., neediness and relatedness, respectively) suggested that the Self-Criticism
and Efficacy factors of the DEQ assess different levels of the self-concept -
a maladaptive and an adaptive view of the self respectively. Thus the DEQ
has become more than a measure of two sources of depression, it has also be-
come an effective research instrument for measuring adaptive as well as
maladaptive aspects of both interpersonal relations and self-definition. The
significant correlations between maladaptive aspects of self-definition and
interpersonal relations (self-criticism and neediness) and between adaptive
aspects of self-definition and interpersonal relations (efficacy and relatedness)
also suggest support for the formulations (Blatt & Blass, 1990, 1996; Blatt &
Shichman, 1983) of the parallel dialectic development of self-definition and
relatedness.

Self-Criticism: Maladaptive aspects of the sell Extensive research docu-

ments the maladaptive aspects of self-criticism (e.g., Blatt, 1974, 2004; Blatt
et aI., 1976, 1982). Individuals with elevated scores on DEQ Self-Criticism
are sensitive to ridicule and are uncomfortable in interpersonal relationships;
they tend to be interpersonally isolated and insensitive, formal, ambivalent,
reserved, and distant, and they often try to manipulate others through deception
and flattery (e.g., Dunkley, Blankstein, & Flett, 1997). They are prone to
feelings of guilt, sadness, hopelessness, and depression (e.g., Mongrain, 1998),
and at times they can be seriously suicidal (Blatt, 1974, 1995b; Blatt et aI., 1982;
Beck, 1983; Enns, Cox, & Inayatulla, 2003; Fazaa, 2001; Fazaa & Page,
2003; Shahar, 200 I). These feelings of self-criticism are relative resistant to
brief therapeutic interventions (Blatt, Quinlan, Pilkonis, & Shea, 1995; Blatt,
Zuroff, Bondi, Sanislow, & Pilkonis, 1998) and several studies (e.g., Alden
& Sieling, 1966; Zuroff & Fitzpatrick, 1995) demonstrate the avoidant quali-
ties of the individuals with elevated scores on the Self-Criticism factor of the
DEQ.

145
 Efficacy: Adaptive aspects of the Self. In contrast to the extensive research on
self-criticism as a maladaptive dimension of the self, research on the Efficacy factor
of the DEQ has been minimal. Investigators who use the DEQ usually report
findings with the Dependency and Self-Criticism factors, mentioning only in
passing that the Efficacy factor was not used in the study (cf. Priel & Shahar, 20(0).
Authors who mention the Efficacy factor usually do so in a few lines, rarely
elaborating its negative correlation with depression (Blatt et al., 1976; Blatt, et al.,
1982; Klein, 1989). Even when Efficacy was found to be associated with higher
levels of functioning over time (Klein, 1989), it was still mentioned only briefly.
Our recent increased interest in the Efficacy factor is primarily the result of our
study of risk and resilience in early adolescence (Leadbeater, Kuperminc, Blatt &
Herzog, 1999). A large and diverse sample of early adolescents (230 girls and 230
boys) in an urban middle school (6th to 8th grade), were evaluated over a one-year
interval. Although most of the reports from this research project focused on
Dependency and Self-Criticism, some interesting findings were obtained with the
Efficacy factor. While Self-Criticism was related to maladaptive indicators of social
and academic functioning, Efficacy was related to indicators of adaptation
(Kuperminc, Blatt, & Leadbeater, 1997), a finding that suggests that Self-Criticism
taps ruminative, self-reflective preoccupations with past and current deficits and
deficiencies, whereas Efficacy assesses proactive feelings and behavior - positive
self-attitudes. Self-Criticism, for example, predicted increases in internalizing and
externalizing symptoms over a one-year period whereas Efficacy predicted a
reduction in these symptoms over the same time period (Kuperminc, Leadbeater, &
Blatt, 2001). Further analyses of the data from the study of these adolescent
students (Shahar, Gallagher, Blatt, Kuperrninc, & Leadbeater, in press) indicate that
Efficacy moderates, or buffers, the adverse effects of Dependency and Self-
Criticism on depressive symptoms. Adolescents who had greater levels of
vulnerability, as reflected in elevated Dependency and Self-criticism, were even
more prone to depression and impaired functioning, especially if they also had
reduced levels of personal resilience, as reflected in lower Efficacy. These
longitudinal analyses indicated that Efficacy moderated the combined effect of
Dependency and Self-Criticism in the development of depression. Specifically,
adolescents who had elevated levels of both Dependency and Self-Criticism tended
to be more depressed if they also have low, rather than high, levels of Efficacy (see
also Blatt et aI., 1982).

Adaptive and Maladaptive Levels of Interpersonal Relatedness and Self-

Definition and the Social Context

Extensive research has examined the interaction between the four DEQ variables
(Neediness, Relatedness, Self-criticism, and Efficacy) and aspects of the social
context like stressful life events, social support, and close interpersonal

146
relationships. As noted in Chapter 4, earlier studies that examined the effect of the
DEQ variables on depressive symptoms usually treated the social context as a
moderator (Blatt & Zuroff, 1992). Contextual variables like stressful events were
expected to augment the maladaptive tendencies of self and interpersonal
relatedness. According to what has come to be called the "congruency hypothesis"
(Blatt & Zuroff, 1992; Coyne & Whiffen, 1995; Robins, 1995), individuals with
elevated levels of dependency would become depressed in response to stressful
interpersonal events like rejection and loss. Similarly, individuals with elevated
levels of self-criticism would become depressed in response to events threatening
the self, such as failure. Empirical support has usually been found for the
congruency hypothesis with respect to dependency, but only on occasion for self-
criticism (for reviews, see Blatt, 2004; Blatt & Zuroff, 1992; Coyne & Whiffen,
1995; Robins, 1995; Shahar, 2001; Chapter 4, this volume).
It is important to note that these various attempts to study the congruency
hypothesis usually assumed that individuals are passive in relation to the contextual
factors that seem to precipitate their distress (Priel & Shahar, 2000). More recent
formulations, however, have come to emphasize that individuals may actively
generate the contextual conditions implicated in their distress (Buss, 1987; Coyne,
1976, 1998; Hammen, 1991, 1998; Joiner, 1994; Plomin & Caspi, 1999). Conse-
quently, investigators have now begun to focus on the interactions of the DEQ
variables with contextual variables like stressful events, social support, and the
quality of close relationships (Blatt & Shahar, 2003).
Research findings indicate that self-criticism is a primary instigator of depressive
symptoms because it generates a risk-related social environment (Dunkley, Zuroff,
& Blankstein, 2003; Shahar, 2001). Specifically, elevated self-criticism inter-
feres with close relations: it predicts interpersonal ruptures and tensions
(Mongrain, Vettese, Shuster, & Kendal, 1998; Vettese & Mongrain, 2000; Priel &
Besser, 2000; Priel & Shahar, 2000; Zuroff & Duncan, 1999) and other
interpersonal stressful events (Priel & Shahar, 2000; Shahar, Joiner, Zuroff, & Blatt,
2004; Shahar & Priel, 2003). Self-criticism also predicts reduced levels of social
support (Mongrain, 1998; Priel & Shahar, 2000), and fewer positive life events
(Shahar & Priel, 2003). Thus, individuals with a maladaptive sense of self appear
to generate contextual conditions that render them vulnerable to depression and
emotional distress.
Dunkley, Zuroff and Blankstein (2003), studying college men and women,
recently examined both personal dispositional and situational factors that contribute
to high negative and low positive affect in self-critical individuals. Over a 7-day
period, they assessed daily reports of hassles, stress, social support, and coping
styles. Using structural equation modeling (SEM), they found that Self-Criticism
influenced emotional experiences each day through a number of maladaptive
tendencies. Self-Criticism contributed to experiences of negative affect through an
increase in daily hassles and the use of an avoidant coping style, and to a reduction
of positive affect through a failure to maintain social support. Also, self-critical
individuals were particularly reactive to stressors that implied personal failure, loss
of control, and criticism from others. They also were relatively ineffective in using
more adaptive coping strategies like problem-focused coping.

147
 Another compelling example of the tendency of self-critical individuals to
generate a negative social context was obtained in analyses we conducted on data
from the Treatment of Depression Collaborative Research Project (TDCRP), a
randomized clinical trial that compared three outpatient treatments for major
depression: Cognitive-behavioral Therapy (CBT), Interpersonal Therapy (IPT), and
Imipramine plus clinical management (IMI-CM), sponsored by the National
Institute of Mental Health. These three active treatments were also compared to an
inactive placebo plus clinical management (PLA-CM). Original analyses of these
data indicated few substantial differences in clinical outcome among the three
active treatment groups (Elkin et aI., 1989; Imber et al., 1990). Additional analyses
that our research team conducted on this data set, however, demonstrated that
patients' pretreatment self-criticism or perfectionism had a highly significant
negative impact on therapeutic outcome in all three treatments conditions (Blatt,
Quinlan et aI., 1995; Blatt, Zuroff et aI., 1998). Pretreatment Self-Critical
Perfectionism impeded the improvement of two-thirds of the patients in this
sample, and this impeding of therapeutic progress occurred primarily during the
second half of the treatment, between the 9th and the 16th session (Blatt et aI.,
1998). Further analyses (Zuroff, Blatt, Sotsky, Krupnick, Martin, Sanislow, &
Simmens, 2000) indicated that at least part of this adverse effect of pretreatment
self-criticism (perfectionism) on treatment outcome was mediated through patients'
impaired participation in the therapeutic alliance. Pretreatment Self-Critical
Perfectionism predicted lower levels of patients' constructive contribution to the
therapeutic alliance. This interference in the formation and maintenance of the
therapeutic alliance in turn predicted poorer therapeutic outcome. We (Shahar,
Blatt, Zuroff, Krupnick, & Sotsky, 2004) also found that the poorer outcome of self-
critical, perfectionist patients resulted not only from their failure to participate in
the therapeutic alliance but also from their difficulties in maintaining close
interpersonal relationships outside therapy. Specifically, pretreatment Self-Critical
Perfectionism predicted reduced interpersonal involvement with the therapist in
treatment as well as in the patients' general social network outside of treatment.
Both of these, in turn, predicted poorer therapeutic outcome. These adverse effects
of pretreatment Self-Critical Perfectionism on the therapeutic alliance and on close
relations explained much of the variance of the adverse effect of Self-Critical
Perfectionism on therapeutic outcome (Shahar, Blatt et aI., 2004; see also Shahar,
Blatt, Zuroff, & Pilkonis, 2002). Thus, Self-Critical Perfectionism not only
contributes to considerable emotional distress but it also disrupts therapeutic
attempts to alleviate this distress.
We (Blatt & Shahar, 2003) believe that self-critical individuals generate a risk-
related social context because they have negative representations of self and
significant others. Several investigators (e.g., Blatt, 1974; Blatt, Wein, Chevron, &
Quinlan, 1979; Mongrain, 1998) have demonstrated that self-critical young adults
hold particularly negative representations of parental figures. These negative
representations seem to organize self-critical individuals' social exchange, making
it difficult for them to respond to positive interpersonal cues (Aube & Whiffen,
1996), forcing them to avoid intimacy and self-disclosure (Zuroff & Fitzpatrick,
1995) and to act in a hostile manner in close relations (Zuroff & Duncan, 1999),

148
thereby creating conflicts, confrontations, and other stressful events (Priel &
Shahar, 2000; Shahar & Priel, 2003). It seems that self-critical individuals project
their own self-criticism onto others and therefore expect the condemnation from
others that they inflict upon themselves. Ironically, to the extent that these negative
representations generate a risk-related environment, this negative interpersonal
environment is likely to consolidate and even exacerbate their negative
representations of themselves and others, thus contributing to a reciprocal, vicious,
interpersonal loop or circle that is frequently observed by clinicians treating
depressed patients (Andrews, 1989; Blatt & Zuroff, 1992; Wachtel, 1994; Zuroff,
1992).
In contrast to the disruptive effects of self-criticism on social relationships, the
impact of the dependency (or the interpersonal) dimension of the DEQ on social
relationships is more complex suggesting that this personality construct contains
both elements of risk and resilience (Blatt, Zohar et aI., 1995, 1996; Bornstein,
1998; Shahar, 200 I; Shahar & Priel, 2003). Though dependency predicts
interpersonal problems that contribute to depression, it also predicts a capacity for
intimacy (Fichman, Koestner, & Zuroff, 1994) and being able to establish and
maintain elevated levels of social support (Mongrain, 1998; Priel & Shahar, 2000).
Dependency, like self-criticism, predicts elevated levels of negative events that lead
to depression and anxiety; but different than self-criticism, dependency also
predicts positive events which partly explains why dependent individuals report
lower levels of distress than self-critical individuals (Shahar & Priel, 2002).
Dependent women are interested in closeness and intimacy and experience greater
feelings of affection and love in their romantic relationships (Zuroff & deLorimier,
1989; Zuroff & Fitzpatrick, 1995). They are more positive about their same-sex
relationships - they perceive these relationships as more friendly and tend to more
frequently use positive expressions in their same-sex interactions (Zuroff & Franko,
1986). Dependent individuals go to remarkable length to preserve interpersonal
harmony (Santor, Pringle, & Israeli, 2000), are uncomfortable with feelings of
hostility (Zuroff, Moskowitz, Wieglus, Powers, & Franko, 1983), and have
difficulty being assertive (Fichman et aI., 1994). Thus, they tend to use compromise
in dealing with interpersonal conflicts (Zuroff & Fitzpatrick, 1995). In summary,
dependency is correlated with investment in interpersonal relationships. In
nonclinical samples it is related to valuing emotional closeness and an active
interest in maintaining good interpersonal relationships. In clinical samples,
dependency is associated with apprehensions and resentments about loss, neglect,
deprivation, and abandonment by parents, spouse, and friends (Blatt et aI., 1982).
Studies have only recently begun to investigate the impact of Efficacy, the
adaptive dimension of self, on the social context. To address this question, we
examined the role of the adaptive and maladaptive sense of self in adolescent
development. We examined the role of the Self-Criticism and Efficacy factors of
the DEQ in predicting a positive social context in young adolescents (Blatt &
Shahar, 2003). The results of this study, derived from a Structural Equation
Modeling (SEM; Hoyle & Smith, 1994) analysis examination of the longitudinal
effects of Self-Criticism and Efficacy on the social context, indicate that Self-
criticism and Efficacy present opposite patterns. Self-Criticism significantly pre-

149
dieted an increasingly negative social context while Efficacy significantly predicted
an increasingly positive social context. Also, it was noteworthy that social context
at Time 1 strongly predicted Time 2 social context, indicating that the social context
of adolescents tends to be stable over time. Thus, the variations in the social context
produced by Self-Criticism and Efficacy are impressive.
In summary, empirical research with the Depressive Experiences Questionnaire
(DEQ) demonstrates that Self-criticism and Efficacy, which respectively assess
maladaptive and adaptive dimensions of the self, are intimately tied to social
relations, and in fact generate contextual circumstances in predictable ways. Self-
criticism generates a negative, risk-related, social context whereas Efficacy gene-
rates a positive, resilience-related context. Thus, the DEQ is an instrument not only
for assessing two major foci of depression, but also a method for assessing several
levels in the development of interpersonal relations and in the sense of self. This
further differentiation of developmental levels within the DEQ should facilitate
further research on depression as well as provide a method for systematically
studying different phases of personality development. Not only has the devel-
opment of the DEQ facilitated the differentiation of two primary sources of
depression and the systematic investigation of some of the etiological, clinical, and
therapeutic issues in these two types of depression (e.g., Blatt, 2004; Blatt &
Zuroff, 1992) - the distal and proximal antecedents, as well as the clinical
implications of these different types of depression - it now has the potential to
contribute to the study of personality development.

Anaclitic and Introjective Configurations of Psychopathology

In addition to the research using the DEQ to assess anaclitic and introjective
dimensions in depression and different levels of interpersonal relatedness and self-
definition, a fair amount of clinical research has used the anaclitic-introjective
distinction categorically to study the differential response of anaclitic and
introjective patients in both brief and long term intensive treatment in outpatient
and inpatient settings. Judges in several studies (Blatt, 1992; Blatt & Ford, 1994;
Blatt & Shahar, 2004b) were able to reliably distinguish between anaclitic and
introjective patients. The results of these studies indicate that anaclitic and
introjective patients are differentially responsive to different types of therapeutic
interventions and express their therapeutic gains in different ways. Introjective
patients are relatively unresponsive to brief treatment (e.g., Blatt et aI., 1995), but
they respond effectively to long-term intensive, psychodynamically oriented
treatment (Blatt, 1992; Blatt & Ford, 1994; Blatt & Shahar, 2004a; Fonagy et al.,
1996). Further analyses (Blatt 1992; Blatt & Shahar, 2004a) of data from the
Menninger Psychotherapy Research Project (MPRP) revealed that introjective
patients had a significantly more constructive response to psychoanalysis than to
Supportive-Expressive Psychotherapy (SEP). In contrast, anaclitic patients were

150
significantly more responsive to SEP than they were to psychoanalysis (Blatt,
1992). Other studies (Blatt & Ford, 1994) indicate that anaclitic and introjective
patients tend to express their therapeutic gains in different ways. Therapeutic
progress in introjective patients is usually expressed in change in their cognitive
functioning and in the intensity of their manifest symptoms, while therapeutic gain
in anaclitic patients is usually expressed in changes in their interpersonal
relationships and in their representation of the human form on the Rorschach. Thus,
anaclitic and introjective patients seem to change in different dimensions - the
dimensions most salient to their character type.
While the anaclitic-introjective distinction has lead to the development of
dimensional measures that have been quite productive in research, the question
remains about the possibility of a mixed type of individual with both predominant
anaclitic and introjective features. We have begun to explore a "mixed group" of
individuals, both within clinical and non-clinical samples, who cannot be easily
classified within the anaclitic and introjective distinction. We have begun to
explore, for example, differences among patients who clearly seem to have
primarily an intense primary preoccupation with issues of relatedness or with issues
of self-definition (clearly differentiated anaclitic and introjective patients) with a
mixed group of patients who are not easily classified within the anaclitic-
introjective distinction and who seem to have intense concerns with both
dimensions - with issues of both relatedness and self-definition.
Research findings in both clinical and non-clinical samples, using a wide-range
of criteria, clearly indicate the validity of the anaclitic and introjective distinction
both categorically and dimensionally. Theoretically, it is possible, however, that
some individuals can have predominant aspects of both dimensions. In fact, one
criterion of normality or high level psychological functioning, such as in self-
actualizing individuals (e.g., Maslow, 1954), would be the ability to integrate, at
high developmental levels, aspects of both an investment in interpersonal
relatedness and in expressions of self-definition - in Eriksonian terms, investment
in intimacy and in generativity, or in Freud's terms, in love and work. Thus, high
levels of psychological functioning can be defined as a capacity to establish and
maintain an effective integration of high levels of both anaclitic and introjective
dimensions - to maintain a capacity of reciprocal relatedness with a clear and
effective identity. As discussed earlier, the capacity for mutuality and interpersonal
reciprocity requires a clear and effective sense of self - an identity in which one
clearly recognizes what one can contribute to, as well as gain from, a relationship.

Mixed Anaclitic and lntrojective Features

Some patients are less amenable to the anaclitic-introjective categorization

because they demonstrate intense and extreme preoccupations in both domains -
with issues of both self-definition and with interpersonal relatedness. Although the

151
number of mixed-type patients is usually relatively small, they raise important
clinical and theoretical issues. Further analyses (Shahar, Blatt, & Ford, 2003) of the
data from the study of long-term, intensive psychodynamically oriented inpatient
treatment of ninety seriously disturbed, treatment resistant patients (the Riggs-Yale
Project [R- YP]; Blatt & Ford, 1994) revealed important differences between the
"mixed group" of patients and patients clearly and confidently classified as either
anaclitic or introjective. Two judges, in rating these 90 patients as either anaclitic or
introjective, also indicated, on a 100-point scale, their certainty or confidence in
making these ratings. We assumed that patients received lower certainty scores
because they had both anaclitic and introjective features. Using these clarity/
confidence ratings, we were able to divide the 90 patients in the Riggs-Yale sample
into three groups. The mixed-type group consisted of the 13 anaclitic patients and
the 14 introjective patients with the lowest scores of clarity/conviction. The mean
certainty of the 13 "previously anaclitics", now mixed-group, patients was M ==
57.53 (SD == 3.07, range == 52-60) and the mean certainty of the 14 "previously
introjectives", now mixed-group, patients was M == 57.53 (SD == 2.49, range == 55-
60). These two groups did not differ in terms of the level of their clarity/conviction
ratings «([25] == .16, ns). Neither did the remaining 29 anaclitic and 34 introjective
patients differ in the average of clarity/conviction assigned to them (M's == 73.13 vs.
76.70, SD's == 7.85 and 10.99, ranges == 63-90 and 63-98; for anaclitics and
introjectives, respectively; ([61] == 1.45, ns).
The mixed-type patients were significantly more distressed and vulnerable at
admission than their anaclitic and introjective counterparts. They had significantly
more psychiatric symptoms, greater thought disorder, lower Performance and Full
Scale IQ, less accurate object representations (among men), and had grea.ter
utilization of maladaptive defense mechanisms (projection [among men], and
identification [among women)). The findings that the mixed group of patients 'With
both anaclitic and introjective features had significantly higher indications of
psychopathology, is consistent with earlier findings (Blatt et al., 1982) based 011 a
comparison of three groups of depressed inpatients - patients with elevated scores
on DEQ Dependency, a group of patients with elevated scores of DEQ self-
Criticism, and a third group of patients with elevated scores on both DEQ
dimensions - that found that the mixed group had significantly higher levels of
depression. Contrary to expectations, however, the mixed-group evidenced
significantly greater therapeutic gain over the course of long-term intensive,
inpatient treatment. Mixed group men improved significantly in terms of
psychiatric symptoms and mixed group men and women had significantly greater
increase in performance IQ and less frequent utilization of projection as a defense
mechanism (Shahar et al., 2003).
These findings raise several important theoretical and clinical issues. These
results suggest that the intense focus of anaclitic and introjective individuals on
issues of interpersonal relatedness or self-definition, respectively, indicates their
capacity to construct a relatively focal mode of adaptation that enables them to
function more effectively at admission than less clearly organized patients. The
mixed-type patients appear not to have achieved the same level of personality
consolidation as either of the "pure" type of individuals. But this consolidated mode

152
of adaptation of clearly defined anaclitic and introjective inpatients appears to
impair significantly their accessibility to treatment and to limit the extent of their
therapeutic gain. The lack of consolidation of a well-articulated defensive orga-
nization in mixed group inpatients, at least in the seriously disturbed, treatment-
resistant inpatients evaluated in this study, appears to make them more accessible
to therapeutic intervention (Shahar et aI., 2003).

Anaclitic and Introjective Dimensions in Dynamic Interaction

Theoretical formulations and research investigations have generally viewed

anaclitic and introjective features as a basic character or personality trait that
usually occurs as a predominant quality or, somewhat less frequently, in com-
bination. This assumption of an anaclitic and an introjective personality orga-
nization has been supported by the results of a wide-range of empirical research
(see summary in Blatt, 2004; see also Chapter 4, this volume). Recent research
(Shahar et aI., 2003) indicates, as discussed above, that the mixed group also
defines another type of personality organization. But recent clinical experience also
suggests that on occasion these character or personality qualities can also co-occur
in a dynamic constellation in which one set of qualities serves as a defense against
recognizing and experiencing the other set of qualities. Thus, an exaggerated
emphasis on introjective issues of self-worth, power, and agency can serve as a
defense against recognizing and experiencing intense and painful interpersonal
longings, or, conversely, intense preoccupation with issues of interpersonal
relatedness and always seeking to be with others can serve as a defense against self-
reflection and the painful recognition of intense feelings of dissatisfaction with
oneself because of profound feelings of failure or guilt.
A clinical example of an impressively powerful, agentic woman demonstrates
how her exaggerated emphasis on these introjective qualities served as a defense
against recognizing and acknowledging the intensity of her depressive longings for
closeness with her mother, a relationship she had long-sought but had never been
able to achieve. These desperate longings for a close relationship with her very
distant, aloof, and unavailable mother unconsciously dominated her psychological
organization and were part of her marked vulnerability to depression and possibly
to suicide.
Madeline, a highly successful 35 year-old attorney, volunteered to be the subject
in a study of the contributions of different approaches to psychological assessment,
conducted by Jerry Wiggins (2003). As part of this project, Madeline visited with
several senior psychologists, each of whom had expertise in a particular approach
to psychological assessment. Assessment of Madeline from a psychodynamic
perspective was conducted by Behrends and Blatt (2003) and material from that
assessment demonstrated how her intense and highly effective introjective
personality organization also served as a defense against profoundly painful, rela-

153
tively unconscious, anaclitic needs. The recent analysis of Madeline's psy-
chological test protocols (Behrends & Blatt, 2003) as part of a book on para-
digms of psychological assessment (Wiggins, 2003) provides an excellent example
of this more dynamic interaction of anaclitic and introjective personality
dimensions.
Madeline, a 35-year old, single, Native American woman, grew up in a highly
disrupted environment, raised by seriously abusive, alcoholic parents. She left
home at the age of 12, lived in a series of foster placements, and in her late
adolescence and young adulthood spent several years in prison. During her
incarceration, she was placed in solitary confinement after seriously assaulting a
fellow female prisoner. After discharge from prison, Madeline changed her life
profoundly. She received a high school diploma through equivalency examination
and eventually graduated law school. Despite her criminal record, she successfully
petitioned the state bar association to sit for the bar examination, and, after much
dispute, was eventually admitted to the bar in her state. She established an extensive
legal practice, successfully defending over 50 clients in criminal proceedings. In
her community, she was considered a most impressive and powerful individual who
had overcome considerable adversity to establish a very successful career. She was
much admired by friends and colleagues, so much so, that Wiggins (2003) selected
her to be the subject of his book on psychological assessment.
Psychodynamic psychological assessment with the Rorschach, Thematic
Apperception Test (TAT), Wechsler Intelligence Test, and the Object Relations
Inventory (ORI; e.g., Blatt, Stayner, Auerbach, & Behrends, 1996), consistent with
findings independently reported from other assessment procedures (e.g., MMPI,
NEO-FFI, etc), noted Madeline's remarkable agentic strengths which were vividly
portrayed in her opening response to the Rorschach on Card I and in a response to
Card V. Equally impressive in Madeline's Rorschach protocol were responses that
indicated her intense vulnerability to threat as expressed in her response to Card IV.
The intensity of these responses suggested that Madeline's agentic capacities also
served as a counter phobic defense to deal with intense fears and apprehensions.
But even more basic was the function of these intense agentic qualities to also ward
off vulnerability to largely unconscious, profoundly painful, feelings of loneliness
and loss as expressed in her continued longing to establish a relationship with her
mother. Thus in many ways, Madeline's impressive introjective qualities and
powerful agentic strengths were the basis not only for her developing very
important adaptive capacities, but they also provide important defensive functions
to protect her from introjective fears and apprehensions about attack and assault,
and, more importantly, from recognizing and experiencing powerful unfilled, but
deeply repressed, anaclitic longings for a close, need-gratifying relationship with a
maternal object.
Madeline, according to Behrends and Blatt (2003), presented an extremely
complex diagnostic picture of being head-strong and fiercely independent as well
as achingly tender and vulnerable. During psychological assessment she was
provocatively oppositional and defiant as well as selfless and generous with a
capacity for empathy and mutuality. In her psychological test protocols (WAIS,
TAT, and Rorschach), Madeline communicated a sense of power and strength that

154
was expressed in her functioning as a highly competent and successful attorney
who has made remarkable achievements despite an extraordinarily painful and
difficult, even traumatic, childhood and adolescence. Her power, strength, and
accomplishment were conveyed in Rorschach responses like the well-perceived
response of a woman in the center of Card I who was seen as "holding her hands
up, got great big wings. Like she's professing! Very powerful! I like that. .. her back
is to you. She's facing the crowd. She'd have to be giving them information ....
Someone important in front of all these people! Like she'd have something
important to say," or in Madeline's response to Card V of the Rorschach, "A
butterfly in flight, quite majestic, out for an afternoon flight."
These responses of power, strength, independence, beauty, and majesty were
juxtaposed with responses indicating intense vulnerability as well as a profound
sense of loneliness and emptiness. Madeline's intense vulnerability was expressed
in her response to Card IV of the Rorschach of a "Scary monster. Great big monster,
getting sick! Huge feet, small head, claws. Oh, it's like fire, burning this little
person. Poor bugger. Very imposing figure! Tiny head, not very smart, dangerous!"
She elaborated the response further by noting that "top is his head, looking down,
spraying from the mouth. First looks like he's getting sick. Then looks like fire,
very dark ....Fire from guy's mouth. Burning him on purpose! Little bugger didn't
stand a chance! .... Like it wasn't accidental. His back is to us. He's inside of the
fire. Little arms hanging down there .... Don't you see that? God, I hope so! It's so
obvious! I need to put some dancing pandas in that picture!" (Referring back to one
of her more positive, playful responses to Card II of the Rorschach).
These Rorschach responses express the polarity of Madeline's experience from
a sense of personal strength and power, to feelings that the world is dangerous and
destructive in which a poor vulnerable little person can be tortured and destroyed.
And Madeline's powerful agentic (introjective) qualities clearly offered her a
considerable sense of strength in this highly destructive and dangerous world. But
it is important to note that Madeline was not only vulnerable to feelings of danger
in a hostile destructive world, but also to a profound sense of loneliness, emptiness,
and abandonment - an active yearning for her mother that was expressed in her
comments when asked to describe herself that "there was never a baby in our
family. The first time I ever kissed my mother was when she was lying in the
hospital, having just slit her wrists on the kitchen table. I saw my cousin brushing
her hair, and I was green with envy." And in describing her mother, Madeline notes
that "She wouldn't let you know her, so your questions remain questions. If you ask
her about her childhood, she just gets up and leaves! So I'm left with just blanks."
Madeline's unfulfilled longings and needs for a close relationship and a primary
attachment with her mother was vividly expressed in several of her responses to
Card X of the Rorschach, "It looks like a party in a psychedelic aquarium. A party!
They're having fun. Everybody's smiling. They live in an ecosystem, all in
someway attached. Fine, so no one's trying to get away. All are enjoying themselves
at the party. All so very, very, very, different. Having a great time." Madeline then
notes that there is a blue crab on the side, "This blue crab. Guy's forlorn, defeated.
There is the eye and a big ole nose. Not so much sad as hopeless. This (on the other
side) is not a mirror image. Otherwise, an underwater circus! A great thing going

155
on!" (The examiner noted feeling sad and tearful herself in reaction to these
responses).
The response of pleasure and excitement in the comfort, stability, and security in
being "attached" to others within a containing ecosystem seems to express a deep
longing for a primary reunion with her mother. Thus, Madeline's exaggerated
introjective assertion of autonomy, freedom, independence, and power seem to have
both important adaptive and defensive functions. These introjective strivings and
expressions of agency are not only expressed in her successful career and her
remarkable resilience in coping with severe adversity in a potentially dangerous and
destructive world, but they also serve to defend Madeline from the painful recog-
nition of the intensity of her profound depressive anaclitic longings for a primary
relationship with her mother. This complex diagnostic picture illustrates how, in
some individuals, exaggerated emphasis on one configuration (in this case, on
qualities of the introjective configuration) can serve to defend against recognizing
and experiencing profound involvement with the warded-off elements of more
painful and threatening anaclitic issues.
Future clinical and empirical research needs to identify ways to differentiate
when anaclitic and introjective qualities define basic personality or character
features and when these two sets of qualities have a more dynamic relationship in
which intense emphasis on one set of attributes serves to defend against recognizing
and experiencing the intensity of concerns in the other domain. This dynamic
defensive interplay between introjective and anaclitic issues can be identified in
clinical evaluations. But subsequent research on this phenomenon will depend on
our ability to develop methods that might enable us to differentiate between
anaclitic and introjective character or personality organization from when these two
types of personality organization occur in a more dynamic context.
The findings with Madeline also suggest that the primary differentiation of
anaclitic and introjective dimensions on the Rorschach, and possibly on other
unstructured assessment procedures like the Thematic Apperception Test (T..:\T;
Morgan & Murray, 1935), may occur primarily in the thematic content rather than
the structural organization of the responses (see also, for example, the contributions
of McAdams, 1993). This would explain why Blatt (1992) and Blatt and Ford
(1994) found so few significant pretreatment differences between anaclitic and
introjective patients on the Rorschach variables they used to empirically assess
therapeutic change. The structural variables on the Rorschach, like F+% (i.e., the
degree of reality testing), degree and type of thought disorder (Allison, Blatt &
Zimet, 1988), and the differentiation, articulation, and integration of human
responses (Blatt, Brenneis, Schimek, & Glick, 1976), may not be effective in
differentiating anaclitic from introjective personality organization, but they may be
very useful in defining the developmental level of the anaclitic and introjective
personality organization. The content of responses to the Rorschach and the Tf\T
(whether they focus on issues of relatedness or self-definition), however, may
provide a reliable basis for differentiating between anaclitic and introjective
personality organization, while the structural variables, like the type and degree of
cognitive organization, may define the developmental level at which the different
behavioral and symptomatic expressions of an anaclitic or introjective personality

156
organization are expressed. Thus, less differentiated and integrated anaclitic
individuals would have more infantile and dependent features - in DEQ terms, have
greater neediness; while anaclitic patients with greater differentiation and
integration would have more oedipal features - in DEQ terms, greater relatedness
with greater adaptive potential. And less differentiated and integrated introjective
individuals would have more paranoid, obsessive, and depressive features - in DEQ
terms, a more self-critical focus, while more differentiated and integrated
introjective mdividuals would have, in DEQ terms, a greater sense of personal
efficacy with greater adaptive potential. These formulations await systematic
investigation.

Conclusions

In summary, our recent empirical research and clinical observations indicate that
refinements of the anaclitic-introjective distinction have facilitated the study of
normal development as well as aspects of the clinical process. The formulations of
the importance of issues of interpersonal relatedness and self-definition as two
fundamental dimensions of psychological development appears to be applicable to
understanding normal development, disruptions in development that are expressed
in different forms of psychopathology, and in the processes of therapeutic change,
thus supporting the fundamental assumption of this work that much can be gained
by considering psychological disturbances, not as separate diseases that may have
some as yet unspecified and undocumented biological causes, but as expressions of
severe disruptions of normal psychological development.

References

Alden, L. E., & Bieling, P. J. (1996). Interpersonal convergence of personality constructs in dynamic and
cognitive models of depression. Journal of Research in Personality, 30, 60-75.
Allison, J., Blatt, S. J., & Zimet, C. N. (1968). The Interpretation ofPsychological Tests. New York: Harper
and Row. Second edition: reprinted in 1988, Hemisphere Publishing Company.
Andrews, J. D. W. (1989). Psychotherapy of depression: A self-confirmation model. Psychological Review,
96,576-607.
Angyal, A. (1941). Foundations for a science of personality. New York: Viking Press.
Angyal, A. (1951). Neurosis and treatment: A holistic theory. New York: Wiley.
Arieti, S., & Bemporad, 1. R. (1978). Severe and mild depression: The therapeutic approach. New York:
Basic Books.
Arieti, S., & Bemporad, J. R. (1980). The psychological organization of depression. American Journal of
Psychiatry, 137, 1360-1365.
Aube, J., & Whiffen, V. E. (1996). Depressive styles and social acuity: Further evidence for distinct
interpersonal correlates of dependency and self-criticism. Communication Research, 23, 407-424.

157
Bakan, D. (1966). The duality ofhuman existence: An essay on psychology and religion. Chicago, IL: Rand
McNally.
Beck, A. T. (1983). Cognitive therapy of depression: New perspectives. In P. J. Clayton & J. E. Barrett
(Eds.), Treatment of depression: Old controversies and new approaches (pp. 265-290). New York:
Raven.
Behrends, R. S., & Blatt, S. J. (2003). Psychodynamic assessment. In 1. Wiggins & K. Trobst (Eds.),
Paradigms of Personality Assessment (pp. 226-342). New York: Guilford Press.
Benjamin, L. S. (1974). Structural analysis of social behavior. Psychological Review. 81, 392-425.
Beutel, M. E., Wiltink, 1., Hafner, c. Reiner, I., Bleichner, E, & Blatt, S. J. (2004). Abhangigkeit und
Selbstkritik OIls psychologische dimensionen der depression - validierung der deutschsprachgen
version des Depressive Experiences Questionnaire (DEQ) [Dependency and Self-Criticisrr as
psychological dimensions in depression: Validation of the German version of the Depressive
Experiences Questionnaire (DEQ)]. Zeitschrift fiir Klinische Psychologie, Psychiatrie und
Psychotherapie, I. 1-14.
Bibring, E. (1953). The mechanism of depression. In P. Greenacre (Ed.), Affective disorders (pp. 13-48).
New York: International Universities Press.
Blatt, S. J. (1974). Levels of object representation in anaclitic and introjective depression. Psychoanaiytic
Study of the Child. 29. 107-157.
Blatt, S. J. (1990). Interpersonal relatedness and self-definition: Two personality configurations and their
implications for psychopathology and psychotherapy. In J. L. Singer (Ed.), Repression and
dissociation: Implications for personality theory. psychopathology & health (pp. 299-335). Chicago:
University of Chicago Press.
Blatt, S. J. (1992). The differential effect of psychotherapy and psychoanalysis on anaclitic and introjective
patients: The Menninger Psychotherapy Research Project revisited. Journal of the American
Psychoanalytic Association. 40, 691-724.
Blatt, S. 1. (\99501). Representational structures in psychopathology. In D. Cicchetti & S. Toth (Eds.),
Rochester Symposium on Developmental Psychopathology: Vol. 6. Emotion. Cognition. and
Representation (pp. 1-33). Rochester, NY: University of Rochester Press.
Blatt, S. J. (I995b). The destructiveness of perfectionism: Implications for the treatment of depression.
American Psychologist, 50. 1003-1020.
Blatt, S. J. (2004). Experiences ofdepression: Theoretical. clinical and research perspectives. Washington,
DC: American Psychological Association.
Blatt, S. J., & Blass, R. (1996). Relatedness and self-definition: A dialectic model of personality
development. In G. G. Noam & K. W. Fischer (Eds.), Development and vulnerabilities in doole
relationships (pp, 309-338). Hillsdale, NJ: Lawrence Erlbaum Associates.
Blatt, S. 1., & Blass. R. B. (1990). Attachment and separateness: A dialectic model of the products and
processes of psychological development. Psychoanalytic Study of the Child,45. 107-127.
Blatt, S. J., & Blass, R. B. (1992). Relatedness and self-definition: Two primary dimensions in personality
development, psychopathology. and psychotherapy. In J. Barron, M. Eagle. & D. Wolitsky (Eds.), The
interface of psychoanalysis and psychology (pp. 399-428). Washington, DC: American Psychological
Association.
Blatt, S. J., Brenneis, C. B.. Schimek, J. G., & Glick, M. (1976). The normal development and
psychopathological impairment of the concept of the object on Rorschach. Journal of Abnormal
Psychology, 85, 364-373.
Blatt, S. J., D' Afflitti, 1. P., & Quinlan, D. M. (1976). Experiences of depression in normal young adults.
Journal ofAbnormal Psychology. 85, 383-389.
Blatt, S. J., & Ford, R. (1994). Therapeutic Change: An object relations perspective. New York: Plenum.
Blatt, S. J., & Homann. E. (1992). Parent-child interaction in the etiology of dependent and self-crit cal
depression. Clinical Psychology Review, 12,47-91.
Blatt, S. J., & Maroudas, C. (1992). Convergence among psychoanalytic and cognitive-behavioral theories
of depression. Psychoanalytic Psychology, 9, 157-190.
Blatt, S. 1., Quinlan, D. M .. Chevron, E. S., McDonald, C., & Zuroff, D. (1982). Dependency and self-
criticism: Psychological dimensions of depression. Journal of Consulting and Clinical Psychology, 50,
113-124.
Blatt, S. J., Quinlan, D. M., Pilkonis, P. A., & Shea, T. (1995). Impact of perfectionism and need for
approval on the brief treatment of depression: The National Institute of Mental Health Treatment of
Depression Collaborative Research Program Revisited. Journal of Consulting and Clinical Psychology,

158
 63, 125-132.
Blatt, S. J., & Shahar, G. (2003). Das dialogische Selbst: Adaptive und maladaptive Dimensionen. In P.
Giarnpieri-Deutsch (Ed.), Psychoanalyse im Dialog der Wissenschaften. Band 2, Anglo-Amerikanische
Perspektiven (pp. 285-309). Stuttgart: Kohlhammer.
Blatt. S. J.•
Blatt, J., & Shahar, G. (2004a). Psychoanalysis: With whom, for what, and how? Comparisons with
psychotherapy. Journal of the American Psychoanalytic Association, 52, 393-447.
Blatt, S. J., & Shahar, G. (2004b). Stability of the patient by treatment interactions in the Menninger
Psychotherapy Research Project. Bulletin of Menninger Clinic, 68, 23-36.
Blatt, S. J., & Shichman, S. (1983). Two primary configurations of psychopathology. Psychoanalysis and
Contemporary Thought, 6, 187-254.
Blatt, S. 1., Stayner, D., Auerbach, J., & Behrends, R. S. (1996). Change in object and selfrepresentations
in long-term, intensive, inpatient treatment of seriously disturbed adolescents and young adults.
Psychiatry: Interpersonal and Biological Processes, 59, 82-107
Blatt, S. J., Wein, S. J., Chevron, E. S.. & Quinlan, D. M. (1979). Parental representations and depression
in normal young adults. Journal ofAbnormal Psychology, 88, 388-397.
Blatt, S. J., Zohar, A. H., Quinlan, D. M., Zuroff, D. C; & Mongrain, M. (1995). Subscales within the
dependency factor of the Depressive Experiences Questionnaire. Journal ofPersonality Assessment, 64,
319-339.
Blatt, S. J., Zohar, A., Quinlan, D. M., Luthar, S. S., & Hart, B. (1996). Levels of relatedness within the
dependency factor of the Depressive Experiences Questionnaire for adolescents. Journal of Personality
Assessment, 67, 52-71.
Blatt, S. 1., & Zuroff, D. C. (1992). Interpersonal relatedness and self-definition: Two prototypes for
depression. Clinical Psychology Review, 12, 527-562.
Blatt, S. J., Zuroff, D. C., Bondi, C. M., Sanislow, C; & PiIkon is, P. (1998). When and how perfectionism
impedes the brief treatment of depression: Further analyses of the N[MH TDCRP. Journal of
Consulting and Clinical Psychology, 66, 423-428.
Bornstein, R. E (1998). Depathologizing dependency. Journal of Nervous and Mental Diseases, 186,
67-73.
Bowlby, J. (1980). Attachment and loss: Volume 3. Loss: Sadness and depression. New York: Basic Books.
Buss, D. M. (1987). Selection, evocation, and manipulation. Journal of Personality and Social Psychology,
53, 1214-1221.
Cane, D. B., Olinger, L. J., Gotlib, 1. H., & Kuiper, N. A. (1986). Factor structure of the Dysfunctional
Attitude Scale in a student population. Journal of Clinical Psychology, 42, 307-309.
Clark, D. A., & Beck, A. T. (1999). Scientific foundations of cognitive theory and therapy of depression.
New York: John Wiley & Sons.
Cohen, M. 8., Baker, G., Cohen, R. A., Fromm-Reichman, E, & Weigert, E. V. (1954). An intensive study
of twelve cases of manic depressive psychosis. Psychiatry, 17, 103-137.
Compas, R. C. (2002). Manifestations of dependent and self-critical personality styles in Rorschach: An
exploratory study. Journal of Projective Psychology and Mental Health, 9, 93-104.
Coyne, J. C. (1976). Toward an interactional description of depression. Psychiatry, 39, 28-40.
Coyne, J. C. (1998). Thinking interactionally about depresssion: A radical restatement. In T. Joiner & 1. C.
Coyne (Eds.), The interactional nature of depression. Washington, DC: American Psychological
Association.
Coyne, J. C; & Whiffen, V. E. (1995). Issues in personality as diathesis for depression: The case of
sociotropy-dependency and autonomy self-criticism. Psychological Bulletin, 118,358-378.
Dunkley, D. M., Blankstein, K. R., & Flett, G. L. (1997). Specific cognitive-personality vulnerability styles
in depression and the five-factor model of personality. Personality and Individual Differences, 23,
1041-1053.
Dunkley, D. M., Zuroff, D. C; & Blankstein, K. R. (2003). Self-Critical perfectionism and daily affect:
Dispositional and situational influences on stress and coping. Journal of Personality and Social
Psychology, 84, 234-252.
Elkin, 1., Shea, M. T., Watkins, J. T., Imber, S. D., Sotsky, S. M., Collins, 1. E, Glass, D. R., Pilkonis, P. A.,
Leber, W. R., Docherty, J. P., Fiester, S. 1., & Parioff, M. B. (1989). NIMH Treatment of Depression
Collaborative Research Program: General effectiveness of treatments. Archives of General Psychiatry,
46, 971-983.
Enns, M. w., Cox, B., & Inayatulla, M. (2003). Personality predictors of outcome for adolescents
hospitalized for suicidal ideation. Journal of the American Academy of Child & Adolescent Psychiatry,

159
 42,720-727.
Erikson, E. H. (1950). Childhood and society (2nd ed.). New York: Norton.
Fazaa, N. (2001). Dependency, Self-criticism and suicidal behavior. Unpublished thesis, Universi:y of
Windsor, Windsor, Ontario.
Pazaa, N., & Page, S. (2003). Dependency and self-criticism as predictors of suicidal behavior. Suicide &
Life-Threatening Behavior. 33,172-185.
Fichman, L., Koestner, R., & Zuroff, D. C. (1994). Depressive styles in adolescence: Assessment, relation
to social functioning, and developmental trends. Journal of Youth and Adolescence, 23, 315-330.
Fonagy, P., Leigh. T., Steele, M.. Steele, H.. Kennedy, R., Mattoon, G., Target, M., & Gerber, A. (1996).
The relation of attachment status, psychiatric classification, and response to psychotherapy. Journal of
Consulting and Clinical Psychology, 64, 22-31.
Frank, S. J., Van Egeren, L. A., Paul, J. S.. Poorman, M. 0., Sanford, K., Williams. O. 8., & Field, D. T.
(1997). Measuring self-critical and interpersonal preoccupations in an adolescent inpatient sample.
Psychological Assessment. 9. 185-195.
Freud, S. (1957). Mourning and melancholia. In J. Strachey (Ed. and Trans.), The standard edition ot the
complete psychological works of Sigmund Freud (Vol. 14.243-258). London: Hogarth Press. (Original
work published 1917)
Freud, S. (1958). On narcissism: An introduction. In J. Strachey (Ed. and Trans.), The standard edition of
the complete psychological works of Sigmund Freud (Vol. 14, 73-102). London: Hogarth Press,
(Original work published 1914)
Freud, S. (1959). Inhibitions, symptoms and anxiety. In J. Strachey (Ed. and Trans.), The standard edition
of the complete psychological works of Sigmund Freud (Vol. 20, 87-174). London: Hogarth Press.
(Original work published 1926)
Freud, S. (1961). Civilization and its discontents. In J. Strachey (Ed. and Trans.), The standard edition of
the complete psychological works of Sigmund Freud (Vol. 21, 64-145). London: Hogarth Press.
(Original work published 1930)
Goodman, S., & Gotlib, I. (2002). Children of depressed parents. Mechanisms of risk and implications for
treatment. Washington, DC: American Psychological Association.
Hammen, C. (1991). Generation of stress in the course of unipolar depression. Journal of Abnormal
Psychology, 100,555-561.
Hammen, C. (1998). The emergence of an interpersonal approach to depression. In T. E. Joiner & J. C.
Coyne (Eds.), The interpersonal nature of depression (pp. 21-35). Washington, DC: American
Psychological Association.
Henrich, C; Blatt. S. J., Kuperminc, G. P.• Zohar, A., & Leadbeater, B. J. (2001). Levels of interpersonal
concerns and social functioning in early adolescent boys and girls. Journal of Personality Assessment,
76,48-67.
Hoyle, R.H., & Smith, G.T. (1994). Formulating clinical research hypotheses as structural equal ion
models: A conceptual overview. Journal of Consulting and Clinical Psychology, 3, 429-440.
Imber, S. D., Pilkonis, P. A., Sotsky, S. M., Elkin, I., Watkins, J. T., Collins. J. E. Shea, M. T., Leber, W.
R., & Glass, D. R. (1990). Mode-specific effects among three treatments for depression. Journat of
Consulting and Clinical Psychology. 58. 352-359.
Jae 1m, C. (1996). The characteristics of two depressive dimensions. Unpublished master's thesis.
University of Korea.
Joiner, Jr., T. E. (1994). Contagious depression: Existence, specificity to depressive symptoms, and the IDle
of reassurance-seeking. Journal of Personality and Social Psychology, 67, 287-296.
Klein. D. E (1989). The revised DEQ: A further evaluation. Journal of Personality Assessment, 53, 7D3-
715.
Kuperrninc, G. P., Blatt. S. J., & Leadbeater, B. J. (1997). Relatedness, self-detinition and early adolescent
adjustment. Cognitive Therapy and Research, 21. 301-320.
Kuperminc, G. P., Leadbeater, B. J, & Blatt, S. J. (2001). School social climate and individual differences
in vulnerability to psychopathology among middle school students. Journal of School Psychology, 39,
141-159.
Leadbeater, B. J., Kuperminc, G. P., Blatt, S. J., & Hertzog, C. (1999). A multivariate mode of gender
differences in adolescent's internalizing and externalizing problems. Developmental Psychology, 35,
1268-1282.
Loewald, H. W. (1962). Internalization, separation, mourning, and the superego. Psychoanalytic
Quarterly, 31, 483-504.

160
Luyten, P. (2002). Normbesef en depressie: Aanzet tot een integratief theoretiscn kader en een empirisch
onderzoek aan de hand van de depressietheorie van S. J. Blatt [Personal standards and depression: An
integrative psychodynamic framework, and an empirical investigation of S. J. Blatt's theory of
depression]. Unpublished doctoral dissertation, University of Leuven, Leuven, Belgium.
Maslow, H. (1954). Motivation and personality. New York: Harper & Row.
McAdams, D. P. (1993). The stories we live by: Personal myths and the making ofthe self. New York:
Morrow
Mongrain. M. (1998). Parental representations and support-seeking behavior related to dependency and
self-criticism. Journal of Personality. 66, 151-173.
Mongrain, M., Vettese, L. C., Shuster, B., & Kendal, N. (1998). Perceptual biases, affect, and behavior in
the relationships of dependents and self-critics. Journal of Personality, 75, 230-241.
Morgan, C. D., & Murray, H. A. (1935). A method for investigating fantasies: The Thematic Apperception
Test. Archives of Neurology and Psychiatry, 34, 289-306.
Nietzel, M. T., & Harris, M. J. (1990). Relationship of dependency and achievement/autonomy to
depression. Clinical Psychology Review, 10, 279-297
Plornin, R., & Caspi, A. (1999). Behavioral genetics and personality. In L. A. Pervin & O. P. John (Eds.),
Handbook ofpersonality theory and research (2nd ed., pp. 251-276). New York: Guilford.
Priel, B., & Besser, A. (2000). Dependency and self-criticism among first-time mothers: The role of global
and specific support. Journal of Social and Clinical Psychology, 19,437-450.
Priel, B., & Shahar, G. (2000). Dependency, self-criticism, social context and distress: Comparing
moderating and mediating models. Personality and Individual Differences, 28, 515-525.
Priel, B., Besser, A., & Shahar, G. (1998). Israeli Adaptation of the DEQ: Psychometric properties.
Unpublished Manuscript. Behavioral Sciences Department, Ben-Gurion University of the Negev, Beer-
Sheva, Israel.
Robins, C. J. (1995). Personality-event interaction models of depression. European Journal of Personality,
9.367-378.
Robins, C. 1., & Luten, A. G. (1991). Sociotropy and autonomy: Differential patterns of clinical
presentation in unipolar depression. Journal ofAbnormal Psychology, 100,74-77.
Robins, C. J., Ladd, 1., Welkowitz, 1., Blaney, P. H., Diaz, R., & Kutcher. G. (1994). The Personal Style
Inventory: Preliminary validation studies of new measures of sociotropy and autonomy. Journal of
Psychopathology and Behavioral Assessment. 16,277-300.
Rude, S. S., & Burnham, B. L. (1995). Connectedness and neediness: Factors of the DEQ and SAS
dependency scales. Cognitive Therapy and Research, 19, 323-340.
Santor, D. A., Pringle, J. D., & Israeli, A. L. (2000). Enhancing and disrupting cooperative behavior in
couples: Effects of dependency and self-criticism following favorable and unfavorable performance
feedback. Cognitive Therapy and Research, 24, 379-397.
Shahar, G. (2001). Commentary on "Shame and community: Social components of depression."
Personality, shame, and the breakdown of social ties: The voice of quantitative depression research.
Psychiatry, 64, 228-239.
Shahar, G., Blatt, S. J., & Ford, R. Q. (2003). Mixed anaclitic-introjective psychopathology in treatment
resistant inpatients undergoing psychoanalytic psychotherapy. Psychoanalytic Psychology, 20, 84-102.
Shahar, G., Blatt, S. J.. Zuroff, D. C; Krupnick, J. L., & Sotsky, S. M. (2004). Perfectionism impedes social
relations and response to brief treatment for depression. Journal of Social and Clinical Psychology, 23,
140-154.
Shahar, G., Blatt, S. 1., Zuroff, D. C, & Pilkonis, P. A. (2002). Role of perfectionism and personality
disorder features in response to brief treatment for depression. Journal of Consulting and Clinical
Psychology. 71,229-233.
Shahar, G., Gallagher, L. E, Blatt, S. J., Kuperminc, G. P., & Leadbeater, B. J. (in press). An interactive-
synergetic approach to the assessment of personality vulnerability to depression: Illustration with the
adolescent version of the Depressive Experiences Questionnaire. Journal of Clinical Psychology.
Shahar, G., Joiner, T. E. Jr., Zuroff, D. C., & Blatt, S. 1. (2004). Personality, interpersonal behavior, and
depression. Co-existence of stress-specific moderating and mediating effects. Personality and
Individual Differences, 36, 1583-1596.
Shahar, G., & Priel, B. (2003). Active vulnerability, adolescent distress, and the mediating/suppressing role
of life events. Personality and Individual Differences, 35, 199-218.
Vettese, L. C; & Mongrain, M. (2000). Communication about the self and partner in the relationships of
dependents and self-critics. Cognitive Therapy and Research, 24, 609-626.

161
Wachtel, P. L. (1994). Cyclical processes in personality and psychopathology. Journal of Abnormal
Psychology, 103,51-66.
Weissman, A. N., & Beck, A. T. (1978. August-September). Development and validation of the
Dysfunctional Attitudes Scale: A preliminary investigation. Paper presented at the 86th Annual
Convention of the American Psychological Association. Toronto.
Wiggins, 1. S. (1991). Agency and communion as conceptual coordinates for the understanding and
measurement of interpersonal behavior. In W. W. Grove & D. Cicchetti (Eds.), Thinking clearly about
psychology, Vo!' 2: Personality and psychotherapy (pp. 89-1(3). Minneapolis: University of Minnesota
Press.
Wiggins, 1. S. (Ed.), (2003). Paradigms of personality assessment. New York: Guilford Press.
Zuroff, D. C. (1992). New directions for cognitive models of depression. Psychological Inquiry, 3,
274-277.
Zuroff, D. c, Blatt, S. 1.. Sotsky, S. M., Krupnick, J. L., Martin, D. 1., Sanislow, CA., & Simrnens, S.
(2000). Relation of therapeutic alliance and perfectionism to outcome in brief outpatient treatment of
depression. Journal of Consulting and Clinical Psychology. 68, 114-124.
Zuroff, D. C, & deLorimier, S. (1989). Ideal and actual romantic partners of women varying in dependency
and self-criticism. Journal of Personality, 57. 825-846.
Zuroff, D. C; & Duncan, N. (1999). Self-criticism and conflict resolution in romantic couples. Canadian
Journal of Behavioral Science. 31, 137-149.
Zuroff, D. Coo & Fitzpatrick. D. (1995). Depressive personality styles: Implications for adult attachment.
Personality and Individual Differences. 18,253-265.
Zuroff, D. C., & Franko, D. L. (1986, April). Depressed and test anxious subjects' interactions with friends:
Effects of dependency and self-criticism. Paper presented at the meeting of the Eastern Psychological
Association, New York.
Zuroff, D. C, Moskowitz, D. S.. Wielgus, M. S.• Powers, T. A., & Franko. D. L. (1983). Construct
validation of the Dependency and Self-Criticism scales of the Depressive Experiences Questionna ire.
Journal of Research in Personality, /7, 226-241.
Zuroff, D. C; Quinlan, D. M., & Blatt. S. 1. (1990). Psychometric properties of the Depressive Experiences
Questionnaire. Journal of Personality Assessment, 55. 65-72.

162
 Chapter 6

'Closed Doors and Landscapes in the Mist' 1.

Childhood and Adolescent Depression

in Developmental Psychopathology

Nicole Yliegen, Patrick Meurs, & Gaston Cluckers

In the following two chapters, we will focus on depression in childhood and

adolescence. Chapter 6 presents theoretical perspectives and empirical research
from the domain of developmental psychopathology in general, summarizing the
knowledge that has been acquired since childhood depression became recognized
as an important clinical problem, some 25 years ago. We will describe the major
developments in this area and we discuss recent integrative theories of childhood
depression. Chapter 7 focuses more specifically on psychodynamic developmental
psychopathology. It treats, from a historical point of view, the contribution
psychoanalysis can make to the domain of developmental psychopathology today,
and it reflects on how psychoanalytic developmental psychopathology can benefit
from the broader current of developmental psychopathology.
Both chapters refer in their title to "closed doors and landscapes in the mist", an
image from a play therapy of a depressive child. We will introduce this image in
chapter 6 and work it out at the end of chapter 7.
 From the first sessions on, eight-year-old Jan' establishes a positive and
differentiated contact with his therapist. He says that he likes to come to the
playroom and is able to symbolize his thoughts and affects in different ways: by
talking, playing, drawing, etc. Once the therapeutic alliance is well formed, he
starts to bring in more depressive contents, signalling an important element he will
have to work through in the course of the therapy. In session 8, Jan makes a very
significant drawing about "silhouettes in the mist". These silhouettes have no alms
and cannot reach each other. Their faces lack any reference to senses. Because of
this, the facial expressions of these figures are rather flat.
In the same session, Jan also plays a boy, longing to go to a house, but the house
is really far away. When the boy reaches the house at last, the door is closed. The
boy waits on the doorstep. He stays there, waiting and waiting for the door to open.
But the boy is unsure whether or when others will come to let him in, and, whether
they will like his initiative at all.

This is the kind of material we can meet in play therapy with depressed children.
Elements of the developmental psychopathological perspective constructed in the
next two chapters, will be used to think about certain vulnerabilities and signs of
resilience in this schoolboy named Jan. In addition to this, we will extensively
return to the introjective and anaclitic elements in Jan's play material and try to
understand them from a "developmental pathways perspective". Some elements of
intervention, sensitive to the specificities of introjective and anaclitic themes, will
be shown.
But firstly, in this chapter, we will summarize some of the main results from
mainstream developmental psychopathological research on depression. We disCJSS
the specificities of the symptomatology of depression in children and its age-related
manifestations; we treat classification issues, prevalence, course and comorbidity.
We discuss epidemiological findings and consider some major theoretical
formulations about childhood depression and development. By the end, we will
come to an integration of perspectives with authors such as Cicchetti and Toth
(1998) and Goodman and Gotlib (1999); we also point out the importance of Blatt's
contribution (Blatt, 2004; Blatt & Homann, 1992) on the developmental origins of
depression. By doing so, we will have worked out a broad framework against
which the addition of a psychodynamic developmental psychopathology
will become clear in the next chapter. In the following chapter, we sketch
a developmental line of normal and maladaptive depressive phenomena in
childhood, bringing together some anchor points of the most important
psychoanalytic theories of depression. At the end of the next chapter, we will then
return to the playroom, in the presence of eight-year-old Jan. Some illustrations
from later parts of his therapy will highlight how one can work with depressed

I The therapist in this case is Patrick Meurs. Jan's process has been described extensively by Meurs and

Cluckers (1999). Some aspects that are illustrative of our clinical work with childhood depression will be
described in the next chapter of this book.

164
children in different phases of treatment and with different kinds of depressive
phenomena.

Depression in Developmental Psychopathology: Research Findings

From Denial to Recognition

For the past twenty-five years, childhood depression has been recognized as an
important problem. Cees de Wit (1997) describes the important change that has
taken place in the overall attitude on this topic. A firm denial gave way into a
general and strong conviction about the importance of depressive syndromes in
childhood and of the implications throughout the life course. This recognition made
it possible for therapeutic interventions and prevention programs to be developed
and set up for depressive children, and to have these programs financed and
evaluated on a scientific basis. However, only recently, a multidisciplinary
workgroup preparing the Strategic Plan for Mood Disorders Research of the
National Institute of Mental Health (NIMH), still noted important gaps in our
understanding of the onset, course, and recurrence of mood disorders in children
and adolescents (Costello, et aI., 2002).
Several factors have facilitated this evolution concerning the recognition of
childhood depression. Firstly, society is confronted with the enormous cost of
untreated childhood depression later on in life. The few follow-up studies that exist
show that prepubertal-onset depressive children have more substance abuse/
dependence and conduct disorder/antisocial personality disorder in adulthood as
well as impaired functioning in important roles (Weissman, et aI., I 999a, b;
Hammen & Garber, 2001). Major depressive disorder (MOD) in particular shows
remarkable continuity from childhood to adulthood and is associated with high
morbidity and potential mortality through suicide (Pine, Cohen, Gurley, Brook, &
Ma, 1998; Weissman, et al., I999a, b).
Secondly, there is increased recognition of the fact that depression cannot only
occur in puberty and adolescence, but also in grade-school children, and even in
toddlers and infants (de Wit, 1997; Garber & Flynn, 2001; Kolvin & Sadovski,
2001; Lazarus, 2002; Luby, 2000; Shaffer & Waslick 2002). One can only be struck
by the tragic suicides of depressed children, reported in our newspapers from time
to time:
A twelve-year-old boy ended his life Thursday night at his school in the
neighborhood of Liege. He hanged himself in the playground and left a note about
being disappointed in love. Suicide at this age is rather exceptional, but depression
in young children is increasing.
Belga, De Standaard, March 2003

165
Symptomatology

Is childhood depression the same as or comparable to depression in adulthood?

Are the symptoms and the course identical? To answer these questions, we will first
of all differentiate between core symptoms and age-specific symptoms of childhood
depression.
Core Symptoms. The DSM-IV criteria (American Psychiatric Association, 1994)
for depressive disorders are essentially identical for all developmental levels or
phases. There are only two minor variations for children and adolescents, with
regard to the duration of the symptoms and the understanding of irritability in
children with depressive symptoms (see also below).
Core symptoms that are typical of depression across one's life span are
mentioned in the DSM-IV, as well as by several important researchers; they include
loss of the capacity to enjoy, decreased interest and motivation, and enduring
dysthymia (Harrington, 1993; Poznanski & Mokros, 1994; De Wit, 1997).
Age-specific manifestations. Developmental psychopathologists (Cicchetti &
Toth, 1998; de Wit, 1997) suggest that manifestations of depression might depend
considerably on the level of cognitive, social, emotional, moral and physiological
development. This means that some of the symptoms of depression are different
during childhood and throughout one's lifespan.
Weiss and Garber (2003) suggested two explanations for developmental
differences in depressive symptoms. First, children and adults differ in how they
express particular symptoms of depression, although the very nature of these
symptoms would be similar at all ages. For example, dysphoric mood might be
expressed by excessive crying in very young children, by nonverbal sadness in
grade-school children, by irritability in adolescents and by depressed mood in
adults, but the core mood symptom is essentially the same across these age-specific
expressions (Garber & Flynn, 200 I).
Another possibility is that the symptoms of the depressive illness really differ in
nature, due to the developmental level. Some depressive symptoms require a certain
level of cognitive, emotional, moral or social development. Under the age of eight,
for example, suicidal ideation is not really possible; one will only observe these
thoughts later in childhood. Another example is that of guilt, which becomes
observable only in older toddlers (de Wit, 1997).
Based upon clinical observation and developmentally based research in 3 to 6
year olds, de Wit (1997) has proposed a model that includes both core symptoms of
depression that are invariable across the life span and age or 'developmental phase
specific' depressive symptoms. This model, which subsequently underwent a series
of revisions based on later research findings (Lous, de Wit, De Bruynn & Riksen-
Walraven, 2002; Lous, de Wit, De Bruyn, Riksen- Walraven, & Rost, 2000) is
presented in Table I.

166
 TABLE I
Differential age-specific symptoms of depression (Based on de Wit, 1997, pp. 169-170; Lous, de Wit,
De Bruyn & Riksen-Walraven, 2002; Lous, de Wit, De Bruyn, Riksen-Walraven, & Rost, 2000).

Infant Toddler/preschooler Grade-school children Adolescent

Anaclitic depression Sadfacial expression Sad mood Sad mood

- Crying and posture
- Searching concerning play and
- Protest playfulness: Concerning activity Concerning activity
- Despair - Inability to - Inability to - Inability to
- Withdrawal experience pleasure experience pleasure experience pleasure
- Apathy - Lack of playfulness - Apathy - Apathy
- Absence of symbolic play - Decreased activity - Decreased activity
- More nonplay behaviors - Withdrawal
- Boredom
Concerning emotionality - Decreased achieve-
Infant depression Concerning growth - Feelings of guilt ment motivation
- Inconsolable and - Failure to thrive - Feelings of being bad - Lack of appetite
frequent crying - Retarded growth - Negative self-esteem - Weariness
- Failure to thrive - Negative thoughts about - Sleep disorders
- Sleep and eating Concerning emotionality the future
disorders - Separation anxiety - Preoccupation with Concerning
- Retarded growth - 'magic' feelings of guilt illness and death emotionality
(from 4 years on) - Preliminary suicidal - Feelings of guilt
- Absence of other thoughts - Feelings of being bad
cognitive symptoms of - Negative self-esteem
depression Secondary - Feeling uneasy and
- Social problems uncomfortable about
- Problems with aggression one's own body,
- Problems with regard to sexuality, relationships
school adjustment with peers
- Somatic problems - Suicidal plans and/or
attempts

Sometimes
- Drug/alcohol abuse
- Conduct disorders
- Truancy! absenteeism

Based on research on children of depressed mothers, Goodman and Gotlib

(1999) offer a view that is even more developmentally guided. These authors not
only take into account symptoms of depression, but they also look at what happens
in the further development of children of depressed mothers. Although these
findings may perhaps not be generalizable to all depressed children, they are
nevertheless very interesting.' Compared with children of non-depressed controls:
• Infants of depressed mothers have been found to be more fussy, to obtain lower
scores on measures of mental and motor development, and to have more difficult
temperaments and less secure attachments to their mothers.

, Research on children of depressed mothers is discussed in more detail below.

167
• Toddlers of depressed mothers have been found to react more negatively to stress
and to be delayed in their acquisition of effective self-regulation strategies.
• School-aged children and adolescents of depressed mothers experience more
school problems, are less socially competent and have lower levels of self-
esteem and more behavior problems (Goodman & Gotlib, 1999).
Blatt (2004, p. 174) states that behavioral disorders in adolescence are often
symptomatic expressions of an enduring subclinical depression that has not reached
a level of severity intense enough to warrant being diagnosed as depression.
Goodman and Gotlib (2002) hypothesize that these developmental disturbances are
part of the vulnerability for depression. We will return to this issue later on in this
chapter.

Classification

In the DSM-IV, depression in childhood and adolescence is not mentioned in the

section on specific childhood disorders, but is discussed in the section on mood
disorders in adults. According to DSM-IV, criteria for depression in childhood and
adolescence are essentially the same as for diagnosing major depression or
dysthymic disorder in adulthood, with two minor differences:
• First, irritability is mentioned as an important manifestation of dysphoric mood.
• Second, in order to diagnose dysthymic disorder in children and adolescents, the
minimal duration of a depressive mood is one year instead of two years for
adults.
Further indications of age-specific manifestations are not mentioned in the
DSM-IY. This is considered highly problematic by many authors in the field of
developmental psychopathology. There remains an important gap between
descriptions of childhood and adolescent depression in DSM-IV and clinical and
research knowledge concerning age-specific expressions of depression. One
important attempt to fill this gap is the DC: 0-3, a diagnostic classification manual
of the zero to three ages (Zero to three's Diagnostic Classification Task Force,
1999). This classification is intended to complement the DSM-IV categorization,
particularly in clinical settings where interventions are set up for (very) young
children and their families because the DSM-IV lacks any criteria for diagnosing
depression in these early developmental phases. The DC describes three types of
mood disorders in infants and toddlers: (l) prolonged bereavement/grief reaction;
(2) depression of infancy and early childhood, and (3) mixed disorder of emotional
expressiveness:
• Prolonged bereavement/grief reaction is typically seen after the loss of a
primary caregiver. Symptoms include:
- Crying, calling, searching, refusing the attempts of others to offer comfort
- Emotional withdrawal, lethargy, sad facial expression, lack of interest in
age-appropriate activities

168
 - Eating and sleep disorders
- Regression/loss of previously achieved developmental milestones
- Constricted range of affect
- Detachment/seeming indifference toward reminders of the lost caregiver
versus extreme sensitivity to any reminder of the caregiver
• Depression ofinfancy and early childhood is a type of depression in infants and
young children who exhibit a pattern of sadness or irritable mood with
diminished interest and/or pleasure in developmentally appropriate activities,
diminished capacity to protest, excessive whining, and a diminished repertory of
social interactions and initiative. These symptoms may be accompanied by
disturbances in sleep or eating behavior, including weight loss. The symptoms
must be present for a period of at least two weeks.
• Mixed disorder of emotional expressiveness should be used for infants and
young children who have an ongoing difficulty expressing developmentally
appropriate emotions. The difficulties reflect problems in their affective
development and experiences.

Prevalence

Traditionally, prevalence studies provide different prevalence estimates.

According to de Wit (1997) and Garber and Flynn (200 I) depression is quite
rare in young children, less common during childhood, but its frequency in-
creases clearly in adolescence. Among children under five, about 0.9% are
depressed (Poznanski & Mokros, 1994). Among grade-school children,
approximately 2% are depressed (Anderson, Williams, McGee, & Silva, 1987;
Kashani & Simonds, 1979). In adolescence, the prevalence of depression
ranges from 3% to 18%. Heuves (1990) found in a study in the Netherlands
that 3% to 8% were depressed. Birmaher et al. (1996) reviewed ten years of Ameri-
can research and found comparable rates ranging between 0.4% and 8.3%.
Lewinsohn, Hops, Roberts, Seeley and Andrews (1993) found a range from
8.3% to 18.5%. Cicchetti and Toth (1998) argue that some developmental
constraints on cognition, language, memory and self-understanding may
compromise the accuracy of an assessment of depression in children. Hankin
and colleagues (1998) state that it is likely that most mood disorders in adults
began during middle to late adolescence, which might be a vulnerable period for the
first depressive episode.
From adolescence on, results of research also show a significant gender
difference (Hankin & Abramson, 1999). From the age of 13 on, girls are about
twice as vulnerable to depression as boys, specifically when one considers "pure"
childhood depression, with no co-morbidity with other diagnoses. When co-
morbidity with other internalizing or externalizing disorders is taken into account,
sex differences are less clear. In pre-puberty, the rate of depression is about equal

169
in boys and girls (Garber & Flynn, 2001), but in some studies - especially those on
early childhood depression - the prevalence of depression is higher among boys (de
Wit, 1997). This raises the question whether depression in pubertal and adolescent
boys is often unrecognized and may express itself in other symptoms such as
irritability, difficult behavior, hyperactivity, learning problems and even conduct
disorders (Blatt, 2004).

Course

When we look in detail at clinical psychological research on depression in

children, we find that major depressive disorders in childhood last on average
about nine months before recovery, while dysthymic disorders (DD) endure for
a period of approximately four years (Kovacs et aI., 1984b). This is very similar
to the duration of depressive episodes in adults (Coryell et aI., 1994). Because
of the important developmental tasks that children face during childhood
and adolescence, these are considerable periods of time. Taking into account
the amount of new developmental capacities a child has to master in several
domains at the same time, we should bear in mind that nine months is almost the
length of time a school year lasts and that four years is a big part of a child's school
life.
With regard to prognosis, most children with major depression recover but the
probability of relapse is considerable. The cumulative probability of a new
depressive episode within two years is 40%, within five years about 70% (Birmaher
et aI., 1996; Kovacs, 1996; Kovacs, Feinberg, Crouse-Novak, Paulauskas, &
Finkelstein, 1984a; Kovacs et aI., 1984b ). The course of dysthymic disorder in
children is different and less spectacular than in childhood major depression, but
still exhibits plenty of pitfalls. Seventy percent of the children that recovered from
dysthymic disorder develop a major depressive disorder within a time span of two
to three years (Kovacs et al. 1984a; Kovacs et al. 1984b ). This subgroup - depressed
children with the longer but less intense course of depression - is also at increased
risk of relapse within two years after recovery: 50% to 70% show relapse if not
treated, and have a less positive clinical picture and outcome when treated.

Comorbidity

Comorbidity is a very common phenomenon and a particular problem for

depressed children and adolescents (de Wit, 1997; Garber & Flynn, 2001). Co-
morbidity rates between depression and other disorders in childhood are estimated

170
at 40% to 90%, which is very high. The most frequent comorbid disorders include
dysthymic disorder, anxiety disorders, oppositional defiant disorder, antisocial
conduct disorders, ADHD, eating disorders, obsessive-compulsive disorders and, in
adolescence, substance abuse.

Of special interest from a clinical point of view is the comorbidity of depression

and antisocial behavior (Harrington, Fudge, Rutter, Pickles, & Hill, 1991). The
problems facing depressed children with antisocial behavior clearly differ from
those of children with depression only. Running away from home, setting fires, and
destroying other people's property are symptoms that occur more frequently in
antisocial children with depression than in antisocial children without depression.
Blatt (2004) argues that many of these behavioral disturbances may be expressions
of depression or depressive equivalents. Depressed children with comorbid anti-
social behavior typically also show more severe depression. From a clinical and
therapeutic point of view, it is very important to keep this potential comorbidity in
mind in the assessment and treatment of children, especially in those children with
antisocial features. Antisocial behavior is often the most obvious and visible
problem, and thus underlying depressive problems are not recognized and therefore
left untreated. In addition, antisocial behavioral problems such as impulse control
problems and the regulation of aggression clearly can have a different meaning in
the case of an underlying depression.
In general, child psychiatric literature considers comorbidity as an important
phenomenon, one that is associated with more severe psychopathology and a worse
prognosis (de Wit, 1997; Harrington, Rutter, & Fombonne, 1996)

Etiology

It is quite difficult to describe the complexity of etiological factors and processes

involved in childhood and adolescent depression. Most research is restricted to one
of these etiological factors, such as:
• genetic factors
• neurobiological factors
• environmental and interpersonal factors
• stressful life events
• temperament
• cognitive distortions

Garber and Flynn (2001) offer an exhaustive overview of research into these
etiological factors that contribute to the vulnerability of depression in childhood

171
and adolescence. In what follows, we briefly highlight some of these research
findings.

Children of depressed parents are three times more likely to experience an epi-
sode of depression than children of a reference group without parental depression.
Among children of depressed parents, 40% experience a depressive episode of at
least six months' duration before they are 18, compared to only 12% of the children
in the control group (Beardslee et al., 1998). Furthermore, the onset of depression
is earlier in these children than in a control group (Weissman et al., 1987).
Research findings provide evidence for both genetic and environmental effects
(Plomin, 1990). Children of depressed parents are genetically more vulnerable but
are also exposed to higher levels of stressful life events, including marital and
family conflict and financial problems (Blatt & Homann, 1992; Hammen et al.,
1987). Furthermore, children of depressed parents seem to be less competent in
dealing with stressors (Garber, Braafladt, & Zeman, 1991).
Phares, Duhig and Watkins (2002) identified 19 studies that explored paternal
depression in relation to their children's functioning. The differences between the
effects of paternal and maternal depression are relatively few. Children of depressed
fathers and those of depressed mothers have about the same risk of developing
emotional and behavioral problems. Research has also shown that the father can
have either a protective or a risk-enhancing role when the mother is depressed,
depending upon his abilities to offer the child interaction and communication that
is more healthy. playful and vital.

Children who, from very early on, grow up in an atmosphere of parental mood
changes that not have been diagnosed have a greater probability of suffering from
dysthymic disorder (DD) and major depressive disorder (MDD) throughout
childhood. These unrecognized recurring mood changes in one or both parents
usually pass, but they can have a serious influence on several domains of child
development, especially when the other parent also suffers from psychological
problems or is little involved in raising the child. The younger the child is when
faced with parental mood disorders, the more it is at risk of developing depressive
symptoms later in childhood or in adult life. This means that treating depression in
adults also can prevent later depression in their offspring by decreasing the
probability of the intergenerational transmission of mood disorders. Furthermore,
this implies that clinicians engaged in the treatment of depressed adults have to take
into account the children of these patients. This can be done in various ways,
ranging from inviting the children to talk about what is happening to their mother
or father, to bringing the family in contact with child guidance workers who can
support the family.

172
 Interactions of depressed mothers with their children are more negative, more
controlling and less responsive or affectively involved. From their exhaustive
review of the literature, Garber and Flynn (2001) conclude that non-supportive and
negative family interactions increase the child's vulnerability for depression in two
ways. Firstly, through modelling and direct feedback from the depressed parent, the
child learns inadequate patterns of social interaction, problem solving and affect
regulation. For instance, a mother's flat and apathic response to her toddler's
proudly showing that he can handle his spoon may lead to feelings of unworthiness
and guilt, and/or a comparable flat expressiveness in the child, through processes of
imitation and identification, especially when this is the mother's habitual way of
responding to the child's moments of pride. Radke-Yarrow and Zahn- Wachsler
(1990) noted that depressed mothers are more hostile, less affectionate or less
consistent in their affection. They are less communicative, less skilful in managing
their children, and try to avoid discipline and punishment, but are at the same time
more negative and critical towards their children.
Second, these negative interactions between child and parent have negative
consequences for the development of the self. To begin with, they tend to
undermine the development of a sense of basic trust in the child. The experience
that others are there or will be available for sensitive parenting or as "a secure home
base" is missing. This also makes the child more vulnerable in interaction with
others in general. The child develops a poor capacity to count upon reliable others,
becomes less communicative about what is difficult for him or her, and does not
develop the skills that can help him find other supportive adults in school or other
contexts.

Stressful life events contribute to the onset of childhood depression and to an

increase in depressive symptoms. There is a significant association between
stressful life events and depression in children (Compas, Grant, & Ey, 1994).
Although there is not one single specific type of stressful life event that causes
depression, research findings indicate that depressed children have experienced
more loss, separation and interpersonal conflict (e.g., Monroe, Rohde, Seeley, &
Lewinsohn, 1999; Rueter, Scaramella, Wallace, & Conger, 1999).
For example, in life histories of depressed adolescents, the death of a parent or
the divorce of parents are more often mentioned than in the life histories of
nondepressed adolescents. But this is also true for early academic failure in grade
school and for first love relationships that were broken off early in puberty. It seems
that several different life events, in interaction with one another, increase the
probability of depression in children. The life histories of children with recurrent
depressive episodes also show an accumulation of parental conflicts, maltreatment,
rejection, separation, several removals, unemployment of both parents, etc. (Garber
& Flynn, 2003)
Blatt, Quinlan, Chevron, McDonald and Zuroff (1982) and Beck (1983)
developed the personality-event congruency hypothesis. When a stressor falls into
an individual's particular area of vulnerability, this would increase the likelihood of

173
depression. For example, individuals whose sense of psychological well-being is
primarily derived from interpersonal relationships (dependent/sociotropic indi-
viduals) have an increased risk of depression when they experience stressors within
the social domain. Those who derive their self-esteem from achievement-related
goals (self-critical/autonomous individuals) are at greater risk of depression when
they encounter failure or an inability to live up to the standards and ideals they had
set up for themselves. However, as in adults (see Blatt, 2004; see also Chapter 4,
this volume), support for the congruency hypothesis in children is mainly limited to
Dependency/Sociotropy (e.g., Fichman, Koestner, & Zuroff, 1997; Hammen &
Goodman-Brown, 1990; Little & Garber, 2000).

Depression in childhood and adolescence is the result of an interaction between

genetic and contextual factors that confront the child with an adversity that cannot
be worked through. There appears to be a "scarring" effect (Garber & Flynn, 2001)
of depression on the explanatory style of the child. Children with a history of
depression tend to develop a so-called "depressive" style of giving meaning to
experiences and of interpreting events. Predominant in this style are feelings of
rejection, of being unloved, not accepted, and inferior to others. These feelings of
insufficiency are associated with a greater sensitivity to signs of rejection and they
influence the way information is processed (i.e., the depressive child only recalls or
reacts to real or fantasized events that confirm his "lack of self-worth").
Furthermore, this negative style can provoke more rejection and social isolation. On
the other hand, even the very minimal amount of negativity that is present in an
otherwise good and positive experience will immediately be picked up and
experienced as a confirmation of all the bad aspects of the self. In that sense, one
negative aspect will have much more of an impact on a depressive child than several
positive experiences, because it is easily linked with and fits in with the negative
expectations of the child. As noted, this explanatory style persists after recovery
from depression. Or, as Nolen-Hoeksema et al. (1992, p. 418) put it: "a period of
depression during childhood can lead to the development of a fixed and more
pessimistic explanatory style, which remains with a child after his or her depression
has begun to subside".
One ofthe strongest predictors of subsequent depression is previous depression.
Compas, Ey and Grant (1993) suggest that depressed mood might be a marker for
subsequent depressive symptoms. Subsyndromal levels of depressive symptoms in
childhood significantly increase the risk of a full major depression in adolescence
(Lewinsohn, Allen, Seeley, & Gotlib, 1999; Pine, Cohen, Cohen, & Brook, 1999;
Rueter et aI., 1999) and in adulthood (Howarth, Johnson, Klerman, & Weissman,
1992; Judd, Akiskal, & Paulus, 1997). This means that depression in childhood
should be treated until full remission is achieved.

174
 Depressotypic organization (Cicchetti & Toth, 1998). Although much research
has been done into the factors that contribute to childhood depression, still little is
known about how these factors contribute to the onset of depression, and about how
these episodes of depression influence further development.
Cicchetti and Toth (1998) developed the concept of depressotypic organization
to describe how distortions in the cognitive, socio-emotional, representational and
biological functions (see Figure I) are present to varying degrees among
individuals with mood disorders and how different etiological factors interact with
one another. In individuals functioning in an adaptive way, there is a coherent
organization among these domains. This means that these individuals usually are
able to consider their thoughts, emotions and representations in the perspective of
the context of experiences and are also able to adapt them to socio-emotional
experiences and relational events. There is a balanced interaction among the
different domains of the internal world and external experience, and when
mismatches or disturbances occur, there is a possibility to change and to adjust. In
depressed individuals one can find either an incoherent organization among these
systems or an organization with one clearly distorted function. The different
structures and domains are not interacting in a balanced and attuned way, i.e., some
intense persistent negative feelings or deeply anchored thoughts about themselves
or the world are leading and influencing all other domains and structures, and there
is little possibility to learn from new experiences. All new experiences fall into the
same track, confirming what already was thought and felt.

Cognitive Socio-emotional
social cognition, affect regulation,
information processing, attachment organization,
attributional style interpersonal relations,
guilt

Depressotypic
organization

Biological Representational
genetics, physiological internal representational
processes, brain models, self schemata,
structural anomalies ... self cognitions

FIGURE 1.
Integration of biological and psychological systems in the emergence of depressotypic organization
(Based on Cicchetti & Toth, 1998).

175
 Childhood Depression and Development

As noted in the introduction to this chapter, whereas childhood depression was

not recognized as such in the literature until recently, there is now a growing body
of knowledge concerning the prevalence, course, comorbidity and general
etiological factors contributing to depression in childhood and adolescence. One
striking characteristic of this literature is that it often followed and sometimes still
follows research lines and theoretical points of view relevant to adult depression.
We still have little understanding of why children become depressive, how the
depression influences further development and how these children grow up into
adulthood. What is needed is a more developmental psychopathological model ::hat
is dimensional. Therefore, in the remainder of this chapter we review some recent
attempts to integrate research findings and developmental theory into
developmentally based and developmentally sensitive models of childhood
depression (e.g., de Wit, 1997; Cicchetti & Toth, 1998; Goodman & Gotlib, 1999).

Depression Interfering With Normal Development

De Wit (1997) emphasizes that depressed children not only are at risk because
of the probability of relapse. Depressive disorders can also interfere seriously with
various developmental tasks during different phases of childhood and adolescence.
In this way, depression can undermine the developmental capacities that are
required for further emotional, cognitive and social growth and health. For instance,
ongoing socialization can stagnate when the child withdraws and feels uncom-
fortable in relationships with peers; apathy and increased inactivity can contribute
to being an "underachiever" and to delay in educational achievement or to strong
feelings of incompetence. Moreover, as we will show further on in this chapter,
depression can be "transmitted" from parent to child (Goodman & Gotlib, 1999).
According to De Wit (1997), a complex model for understanding the interaction
among biological vulnerabilities, temperament, emotional immaturity, specific
educational styles of the parents and cognitive biases is needed. In the context of
this interaction, affect-regulation or mood-regulation problems that are typical of
childhood depression can take place. A typical element of these regulatory
problems is that negative experiences tend to have long-lasting consequences and
their influence is not easily moderated by later experiences. Moreover, even then,
early negative experiences tend to be associated with increased vulnerability when
faced with stress later on in life.

176
The Ecological Transactional Model of Depression (Cicchetti and Toth, /998)'

Cicchetti and Toth (1998) propose an even more complex model to understand
the development of depression in children. In this model, the multiple transactions
among environment, caregiver and child are not only represented, but they are
perceived of as dynamic, reciprocal factors contributing to the likelihood of a
depressotypic organization and the emergence of depressive disorder (see Figure 2).
Moreover, whether depression occurs is determined not only by the presence or
absence of specific vulnerabilities or protective factors. The interplay between these
factors and current and previous levels of adaptation as well as the developmental
period during which risk factors and stressors are experienced are considered to be
important determinants in the pathogenesis of childhood and adolescent depression.

Macrosystem
Macrosystem

Exosystern
Exosystern Ongoingtransactions of
Ongoingtransactions of
potentiating and
potentiating andcompensatory
compensatory
processesat each
processesat eachlevel
levelof
of the
the
social ecology
socialecology

Micro-
system Ontogenic
Development

Distaloutcomes
Depressotypic
organization Dysthymia
Dysthymia
Majordepressivedisorder
Majordepressivedisorder
Doubledepression
Doubledepression
Comorbiddisorders
Comorbiddisorders
Positiveadjustnlent
Positiveadjustment

FIGURE 2.
The interplay of multiple transactions among environmental forces, caregiver and child characteristics
as contributing to a depressotypic organization and to the likelihood of depression
(Based on Cicchetti & Toth, 1998).

'The model proposed by Cicchetti and Toth (1998) is even more complex than our brief summary suggests.
However, even this brief sketch highlights the need for integrative theoretical models that are able to
account for a wide variety of theoretical and empirical findings concerning depression in childhood and
adolescence.

177
 In this transactional approach, ongoing transactions of risk and protective
processes are seen as contributing to the emergence and development of a
depressotypic organization and thus to the onset and recurrence of depressive
disorder. Potentiating factors increase the likelihood that a depressotypic
organization and a depressive illness occur, while compensatory factors decrease
the probability of their occurrence. Potentiating and compensatory factors are
associated with four different levels:
• the macrosystem which contains the beliefs and values of the culture
• the exosystem which includes aspects of the community in which children end
families live
• the microsystem which concerns the immediate environment, mostly the family
• ontogenic development: those factors within the person that affect his or her
adaptation. In ontogenic development, one can differentiate still further between
the developmental biological system on the one hand and the resolution of
developmental tasks on the other hand. Maladaptive resolutions of early
developmental tasks can contribute to the development of pathways to
depression through depressotypic organizations of developmental structures,
According to Cicchetti and Toth (1998), four developmental issues are of
specific importance for the emergence of a depressotypic organization and a
depressive disorder:
- The development of homeostatic and physiological regulation
- Affect differentiation and the modulation of attention and arousal
- The development of a secure attachment relationship
- The development of the self-system

Child Development in the Context of Parental Depressive Psychopathology

Goodman and GotUb's (1999) Integrative Model

As we pointed out before, much research has focused on the adverse effects on
offspring of parents with depression. Recently, Goodman and Gotlib (1999)
proposed an integrative model with the purpose of identifying mediating and
moderating factors in the relationship between parental depression and maladaptive
developmental pathways.
According to Goodman and Gotlib, such an integrative model should incorporate
reciprocal and transactional relationships and should take into account a
developmental point of view. Researchers have increasingly recognized the
importance of such reciprocal and transactional relationships within families. The
child's characteristics, such as gender and temperament, can strongly influence the
developmental course as a function of the complex interplay between parent and

178
child (Gotlib & Wheaton, 1997). A developmental perspective, in turn, is to a
certain extent still lacking in most studies of children of depressed mothers. Yet,
Cicchetti and Schneider-Rosen (1986) already argued that knowledge of develop-
mental processes is essential when studying risk factors for the development of
psychopathology in children and to subsequently develop prevention programs.
Most researchers studied children of depressed mothers in one developmental phase
or across a broad range of developmental phases, which limited the extent to which
their findings could be generalized. For instance, conclusions drawn from studies
of children in one developmental stage cannot be automatically generalized to
children of other developmental stages. Moreover, studies that select children with
a different age can mask important developmental issues. Hence, many studies
failed to address developmental issues. In addition, from a developmental perspec-
tive, it appears essential to take into account the timing of the mother's depression,
especially of the first depressive episode during the child's lifetime. Hence, a
developmental theory must be sensitive to the maturational tasks the child has to
master at the time of the exposure to the depression and its consequences.
Finally, from a developmental perspective, one can see that adult depression
usually is not a single, time-limited event (see also Chapter 1, this volume). As
many depressions are chronic or recurrent, few children are exposed to only one
episode of maternal depression. Research often fails to take into account the
specific moment in his development at which the child is confronted with parental
depression.

General Principles of Goodman and Gotlib's Integrative Risk Model

Goodman and Gotlib (1999) describe four possible transmission mechanisms

involved in the transgenerational transmission of depression from mother to child:
Two of these mechanisms are mainly of a biological nature, such as heritability of
depression and innate dysfunctional neuroregulatory mechanisms. The other
mechanisms have to do with disturbances in the interaction between the child of a
depressed mother and his/her environment, including the exposure to maternal
negative and/or maladaptive cognitions, behaviors and affects and the exposure to
a stressful environment in general (see Figure 3). When the mother is depressed,
one or, in most cases, a combination of these transmission mechanisms is present.
The occurrence of one or more of these transmission mechanisms is also
hypothesized to be associated with the emergence of vulnerabilities in several
domains of the child's functioning: psychobiological, cognitive, affective and
behavioral/interpersonal. These vulnerabilities should not be considered as discrete

, Although Goodman and Gotlib (1999) discuss the effects of depression in the mother on the child. as
noted earlier. there are few differences between paternal and maternal depression with regard to their
effects on child development (Phares et al., 2002)

179
domains, but as interacting with each other. Finally, Goodman and Gotlib propose
that these vulnerabilities interact with three potential moderators: the role of the
father, the timing and the course of maternal depression and different child
characteristics.
This theoretical model allows Goodman and Gotlib to summarize a wide variety
of research on children of depressed mothers. In the next section, we briefly discuss
the main transmission mechanisms as well as moderators of this transmission
described by Goodman and Gotlib.

Risk
Mechanisms

Heritability of _
Vulnerabilities
Psychobiological
.
Outcome
Childhood or
adolescent
Depressed
mother depression
dysfunction

Skilldeficits or
. depression
Otherdisorder..

maladaptive styles
Innate or behavioral
tendencies
dysfunctional ,/

···
neuroregulatory Cognitive
Affective
mechanisms
Behavioral or
Moderaton: interpersonal
Fathers:
·· Availability
Mental health
Timing& course
Exposure to
mother's
negative and/or
of mother's maladaptive
depression cognitions!
Childcharacteristics behaviors and

·· Temperament
Gender
affect

· Intellectual &
social-cognitive
skills Exposure to
stressful
environment

FIGURE 3.
An integrative model for the transmission of risk to children of depressed mothers
(Based on Goodman & Gotlib, 1999).

Mechanisms of Transmission

Mechanism 1: Having a depressed mother confers on the child a generic

predisposition to depression. Genetic vulnerability for depression may be either
indirect, direct, or both. First, children of depressed mothers may inherit depression

180
directly. They may inherit DNA that regulates the biological mechanisms of the
child in a way that is different from that of children of nondepressed mothers.
Secondly, genetic liability might be indirect, by inheriting vulnerability factors for
depression, such as particular personality traits or cognitive or interpersonal styles
that increase the risk of depression.
Summarizing the large amount of empirical support for this mechanism, we can
say that there is less clear support for childhood-onset and adolescence-onset
depression than there is for adult-onset depression. Furthermore, the genetic
contribution seems to be greater for early than for late-onset depression. Herita-
bility has further been found to be stronger for less severe childhood depression and
for girls, in contrast to more severe levels of depression and in boys, where
environmental factors seem to have a stronger influence.

Mechanism 2: Infants of depressed mothers are born with dysfunctional

neuroregulatory mechanisms that interfere with emotional regulation processes and
consequently increase vulnerability to depression. This mechanism assumes that
inheritance of dysfunctional neuroregulatory mechanisms is due to genetic factors
or to negative prenatal experiences. This can be the case when the mother was
depressed during or before pregnancy. The foetus is then exposed to
neuroendocrine alterations associated with the mother's depression, to constricted
blood flow, etc. However, much of this research is new and has yet to be replicated.
Goodman and Gotlib also remark that none of these studies tested the role of early
behavior and interaction as mediating the association between maternal depression
and depression in children.

Mechanism 3: Depressed mothers expose their children to negative or

maladaptive cognitions. behaviors, and affect, which places the children at elevated
risk of developing depression. This mechanism consists of several components.
First, parental depression is associated with negative thoughts, behaviors and affect.
This can have a serious impact on the quality of mothering. Depressed mothers
typically show inadequate, non-supportive parenting behaviors, and are unable to
attune to the child's developmentally dependent social and emotional needs. This in
turn will influence the child's social, cognitive and emotional development.
Through identification, modelling, and social learning, the child often acquires
ways of thinking, behaving and feeling that resemble those of the depressed mother,
putting the child at increased risk of depression.
Goodman and Gotlib (1999) state that there is little question that depressed
mothers are characterized by negative affect, cognition and behavior, leaving them
unable to meet the social and emotional needs of their children. These unmet needs
limit the children's social, emotional and cognitive development. Although there is
little research examining whether this negative way of feeling, thinking and
behaving in children of depressed mothers leads to the development of depression,

181
there is evidence from human and animal studies that this negative functioning
places the children at an elevated risk of depression.

Mechanism 4: Contextual features in families with depressed mothers,

particularly some specific stressors, contribute significantly to the development of
psychopathology in children. As noted in Chapters 4 and 8 in this volume, the
relationship between stress and depression is well documented in adults. Children
of depressed mothers are not only exposed to their mother's depression, but also to
a variety of stressors associated with their mother's depression in the domains of
marital and social relationships, job, finances, etc. These elevated levels of exposure
to stress, in comparison with children of normal parents, further elevate the risk of
depression. Despite the general finding that children of depressed mothers are
living in a more stressful environment, little research has examined the potential
mediating role of stress.

Moderators

Moderator 1: Fathers may increase the risk of psychopathology in children of

depressed mothers if they are absent or if they also show psychopathology.
Similarly, fathers may represent a protective factor if they are healthy, involved and
supportive. According to Goodman and Gotlib (1999), there is growing evidence
that fathers have an important role in families with a maternal depression. They can
either increase or decrease the risk of psychopathology in their offspring. When
fathers are absent and/or suffer from mental disorders themselves, they are not able
to offer supportive, corrective, healthier relational experiences, nor are they able to
support the child's healthy social relationships with others such as peers. From a
clinical point of view, for the child, the absence of a healthy adult often implies the
absence of recognition of the child's need for sensitive, responsive and caring
parenting by adults. In contrast, although research has not always yielded consistent
results, taken together, existing studies suggest that normal, involved fathers may
protect the child from future psychopathology.

Moderator 2: The timing and evolution of maternal depression. In general, the

younger the child is at the time of maternal depression, the bigger the impact.
Because the first year of life represents an important critical period characterized
by rapid development in the physiological, affective and cognitive domains,
maternal depression at this age (e.g., postpartum depression) has particularly
serious consequences for subsequent development. Parental regulation of the

182
infant's emotions, for example, is a very important part of relational exchange, and
postpartum depression can make the mother incapable of noticing the child's needs
and responding to them in a helpful and regulating way.
Another important moderator is the course ofmaternal depression. As with most
depressed patients, depressed mothers recover from their depression sooner or later.
However, little is known about the response of children to this recovery. Yet, the
patterns the child has acquired while the mother was depressed, and which can be
seen as adaptive reactions in dealing with a depressive key figure, may already
become generalized. Hence, these "ways of being with a depressed mother" (Stern,
1995) may have become generalized to other relationships. For instance, while a
withdrawal reaction can be adaptive in the context of inconsistent, unresponsive or
harsh parenting, it can prevent the child from finding supportive interactions with
others: with healthy adults or with friends in one's peer groups. This can further
contribute to feelings of loneliness and social incompetence. Moreover, when
depression in the mother is more chronic or recurrent, which is often the case given
the fact that relapse in depression is the rule rather than the exception (Belsher &
Costello, 1988), the adverse impact on the child's functioning can be expected to be
even bigger. Clearly, more research is needed to find out more about the impact of
the timing and the course of depression in mothers on the subsequent development
of their offspring.

Moderator 3: Characteristics of the child, such as temperament, gender, intel-

lectual and social-cognitive skills, can moderate the association of maternal
depression and developmental child outcomes. From a transactional and clinical
point of view, one could argue that temperament could play an important
moderating role in the relationship between maternal depression and child
dysfunction. Empirical research, however, has found no support for this assumption
(Goodman & Gotlib, 1999).
Research on the impact of gender has yielded conflicting results. Most research
on children of depressed mothers does not report gender differences. However,
some studies have found that maternal depression has more of a negative impact on
the development of girls as compared with boys, while other studies have found the
reverse. Goodman and Gotlib (1999) wonder whether these conflicting findings
could be explained by gender-specific models of the impact of maternal depression.
Based on Murray's (1992) findings that there is a significant increase in insecure
attachment in boys due to a non-sensitive environment, Hay (1997), for instance,
suggested that boys may be more vulnerable to mothers' noncontingent responsi-
veness and negative affect than girls. Other research (e.g., Klimes-Dougan &
Bolger, 1998) found different coping strategies in boys and girls. Leadbeater,
Kuperminc, Blatt and Hertzog (1999) found in a group of early adolescent boys and
girls (age 11-13) that self-critical vulnerability predicted internalizing and
externalizing problems for both boys and girls. Interpersonal vulnerability
predicted increases in internalizing problems. These authors found also that the
quality of adolescent relationships with parents and peers influenced the adjustment

183
of girls. Girls with positive relationships with their parents showed a lower level of
internalizing and externalizing problems, had fewer stressful life events, and less
self-critical vulnerability. For boys, positive relationships were associated with less
self-critical vulnerability and fewer stressful life events but paradoxically with a
higher level of interpersonal vulnerability. However, further research is needed to
investigate the impact of gender.
Intelligence as well as level of social-cognitive functioning seems to protect the
child against adverse outcome. Radke-Yarrow and Sherman (1990), for instance,
found in two different age groups that children characterized by three protective
factors (above-average intelligence, socially engaged and "having a special positive
place in the family") did not meet criteria for any psychiatric diagnosis, even when
the mother clearly suffered from major depression.

To summarize, the theoretical model proposed by Goodman and Gotlib

(1999) not only integrates a wide variety of research on maternal depression, but
it also sets a research agenda for the coming years. In our view, one particular point
that deserves future research attention concerns the mechanisms by
which specific child characteristics exert a moderating effect on the relationship
between maternal depression and child development. These mechanisms might
also include interactions between child and life stress (see also Chapter 4, this
volume).

Summary and Conclusions

Depression in childhood and adolescence has long been a neglected topic in

theory and research. Research in this domain originated only about 25 years ago,
and coincided more or less with the birth of developmental psychopathology. In
these 25 years and mainly under the influence of developmental psychopathology,
the notion of "depression" in childhood and adolescence evolved from a static
condition that was described with clear and delineated symptoms, towards a
developmentally based disorder that can affect people throughout their lives, from
early infancy to later life. Childhood and adolescent depression are not only
associated with a risk of relapse in later life, but also with a more broad, possibly
enduring vulnerability for psychopathology in general. However, at the same time,
the presence of protective factors both in the child and his/her environment might
moderate this vulnerability.
In this chapter, we reviewed some of the main research findings and theoretical
models that integrate these findings from a developmental psychopathological
perspective. As noted, although our insights into childhood and adolesc-ent
depression and the factors contributing to it are already considerable, much more

184
theoretical and empirical research is needed. Ultimately, this knowledge should be
able to inform both preventive and therapeutic programs that are specifically
tailored to the developmental needs of depressed children and adults.
In the next chapter, we will focus on one specific orientation within
developmental psychopathology, namely a psychoanalytic developmental
psychopathological perspective. As psychoanalytic thinking concerning depression
has always focused on childhood factors as contributing to depression, and even
considers depression as an inherent part of human development, a comparison of
psychoanalytic developmental psychopathology with the theory and research
reviewed in this chapter could prove to be a fruitful endeavor to further our
understanding of depression in childhood and adolescence.

References

American Psychiatric Association (1994). Diagnostic and statistical manual of mental disorders (4th ed.).
Washington. DC: Author.
Anderson, J., Williams, S., McGee, R., & Silva, P. (1987). DSM-III disorders in pre-adolescent children:
Prevalence in a large sample from the general population. Archives (if General Psychiatry, 44. 69-76.
Beardslee. w.. Versage, E., & Gladstone, T. (1998). Children of affectively ill parents: A review of the past
ten years. Journal of the American Academy (if Adolescent Psychiatry, 37, 1134-1141.
Beck, A. (1983). Cognitive therapy of depression: New perspectives. In P. Clayton & J. Barrett (Eds.),
Treatment (if depression: old controversies and new approaches (pp, 265-290). New York: Raven Press.
Belga (2003). Twaalfjarige verhangt zich op school. [Twelve year old hangs himself]. De Standaard,
0110312003.
Belsher, Goo & Costello, C. (1988). Relapse after recovery from unipolar depression: A critical review.
Psychological Bulletin, 104. 84-96.
Birmaher, B., Ryan, N., Williamson. D; Brent, D., Kaufman, J., Dahl, Roo et al. (1996). Childhood and
adolescent depression: A review of the past ten years. Part I. Journal (if the American Academy (if Child
and Adolescent Psychiatry. 35. 1427-1439.
Blatt. S. J. (2004). Experiences ofdepression. Theoretical, clinical. and research perspectives. Washington,
DC: American Psychological Association.
Blatt. S. J.. & Homann, E. (1992). Parent-child interaction in the etiology of dependent and self-critical
depression. Clinical Psychology Review, 12,47-91.
Blatt. S.. Quinlan, D., Chevron, E., McDonald, C; & Zuroff, D. (1982). Dependency and self-criticism:
Psychological dimensions of depression. Journal of Consulting and Clinical Psychology. 50. 113-124.
Cicchetti, D; & Schneider-Rosen, K. (1986). An organizational approach to childhood depression. In M.
Rutter, C. Izard & P. Read (Eds.), Depression in young people: Developmental and clinical perspectives
(pp. 71-134). New York: Guilford Press.
Cicchetti, D., & Toth, S. (1998). The development of depression in children and adolescents. American
Psychologist. 53, 221-241.
Compas, 8., Ey, S., & Grant, K. (1993). Taxonomy, assessment and diagnosis of depression during
adolescence. Psychological Bulletin. 114. 323-344.
Cornpas, B., Grant, K., & Ey, S. (1994). Psychological stress and child/adolescent depression: Can we be
more specific? In W. Reynolds & H. Johnston (Eds.), Handbook of depression ill children and
adolescents (pp. 509-523). New York: Plenum.
Coryell, w., Akiskal, H., Leon. A., Winokur, G., Maser, J., Mueller, Too et al. (1994). The time course of
non-chronic major depressive disorder: Uniformity across episodes and samples. Archives (if General
Psychiatry. 51. 405-410.
Costello, E., Pine, D., Hammen, C; March, J., Plotsky, P., Weissman, Moo et al. (2002). Development and

185
 natural history of mood disorders. Biological Psychiatry, 52, 529-542.
de Wit, C. (1997). Depressies bij kinderen en adolescenten. De stand van zaken na vijfentwintig jaar
onderzoek. [Depression in children and adolescents. State of the art after twenty-five years of research].
Kind en Adolescent, 4, 166-184.
Fichman, L., Koestner, R., & Zuroff, D. C. (1997). Dependency and distress at summer camp. Journal of
Youth and Adolescence, 26, 217-232.
Garber, J., Braafladt, N., & Zeman, J. (1991). The regulation of sad affect: An information processing
perspective. In J. Garber & K. Dodge (Eds.), The development of affect regulation and dvsregulation
(pp. 208-240). New York: Cambridge University Press.
Garber, J., & Flynn, C. (200 I ). Vulnerability to depression in childhood and adolescence. In R. Ingram &
J. Price (Eds.), Vulnerability to Psychopathology. Risk across the lifespan (pp. 175-225). New
YorklLondon: The Guilford Press.
Goodman, S., & Gotlib, I. (1999). Risk for psychopathology in the children of depressed mothers: A
developmental model for understanding mechanisms of transmission. Psychological Review, 106, 458-
490.
Goodman, S., & Gotlib, I. (2002). Children of depressed parents. Mechanisms of risk and implications for
treatment. Washington, DC: American Psychological Association.
Gotlib, I., & Wheaton, B. (Eds.), (1997). Stress and adversity over the life course: Trajectories and turning
points. New York: Cambridge University Press.
Hammen, C; & Garber, J. (200 I). Vulnerability to depression across the life span. In R. Ingram & J. Price
(Eds.), Vulnerability to Psychopathology. Risk across the lifespan (pp. 258-267). New York! London:
The Guilford Press.
Hammen, C; & Goodman-Brown, T. (1990). Self-schemas and vulnerability to specific life stress in
children at risk for depression. Cognitive Therapy and Research, 14.215-227.
Hammen, C; Gordon, D.. Burge. D.. Adrian. C.. Jaenicke, C.. & Hiroto, D. (1987). Maternal affective
disorders, illness and stress: Risk for children's psychopathology. American Journal ofPsychiatry, /44,
736-741.
Hankin, B., & Abramson, L. (1999). Development of gender differences: Description and possible
explanations. Annals of Medicine, 31, 372-379.
Hankin, B., Abramson. L., Moffit, T.. Silva. P., McGee. R., & Angell, K. (1998). Development of
depression from preadolescence to young adulthood: Emerging gender differences in a 10··year
longitudinal study. Journal ofAbnormal Psychology: /07, 128-140.
Harrington, R. (1993). Depressive disorder in childhood and adolescence. Chichester: John Wiley and
Sons.
Harrington, R., Fudge, H.. Rutter, M.. Pickles. A., & Hill, J. (1991). Adult outcomes of childhood and
adolescent depression. II. Links with antisocial disorders. Journal of the American Academy of Child
and Adolescent Psychiatry. 30. 434-439.
Harrington, R.. Rutter, M., & Fombonne. E. (1996). Developmental pathways in depression: Multiple
meanings. antecedents. and endpoints. Development and Psychopathology, 8, 601-616.
Hay, D. (1997). Postpartum depression and cognitive development. In P. Murray (Ed.), Postpartum
depression and child development (pp. 85-110). New York: Guilford Press.
Heuves, W. (1990). Depression in young male adolescents. Theoretical and clinical aspects. Unpublished
doctoral dissertation. Rijksuniversiteit Leiden, Leiden, The Netherlands.
Howarth, E., Johnson, J., Klerman, G.. & Weissman, M. (1992). Depressive symptoms as relative and
attributable risk factors for first onset major depression. Archives of General Psychiatry; 49, 817-1123.
Judd, L., Akiskal, H.. & Paulus. M. (1997). The role and clinical significance of subsyndromal depressive
symptoms (SSD) in unipolar major depressive disorder. Journal ofAffective Disorders, 45, 5-18.
Kashani, 1., & Simonds, J. (1979). The incidence of depression in children. American Journal of
Psychotherapy, 136, 1203-1205.
Klimes-Dougan, B., & Bolger. A. (1998). Coping with maternal depressed affect and depression:
Adolescent children of depressed and well mothers. Journal (if' Youth and Adolescence, 27. 1-15.
Kolvin, I., & Sadovski, H. (2001). Childhood depression: Clinical phenomenology and classification. In I.
Goodyear (Ed.), The depressed child and adolescent tZnd ed.). Cambridge Child and Adolescent
Psychiatry (pp. 119-142). New York: Cambridge University Press.
Kovacs. M. (1996). Presentation and course of major depressive disorder during childhood and later years
of the life span. Journal of the American Academy (if' Child and Adolescent Psychiatry. 35. 705-715.
Kovacs, M.. Feinberg, T.. Crouse-Novak. M.. Paulauskas, S., & Finkelstein, R. (I 984a). Depressive

186
 disorders in childhood: I. A longitudinal perspective study of characteristics and recovery. Archives of
General Psychiatry. 41. 229-237.
Kovacs, M., Feinberg, T., Crouse-Novak, M., Paulauskas, 5., Pollock, M., & Finkelstein, R. (1984b).
Depressive disorders in childhood: II. A longitudinal study of the risk for subsequent major depression.
Archives of General Psychiatry. 41.643-649.
Lazarus, A. (2002). Depression in primary care: Where psyche meets soma. Psychiatric Annals, 32.
505-506.
Leadbeater, B. J., Kuperminc, G. P., Blatt, S. J., & Hertzog, C. (1999). A multivariate model of gender
differences in adolescents' internalizing and externalizing problems. Developmental Psychology. 35.
1268-1282.
Lewinsohn. P., Allen, N., Seeley, J., & Gotlib, I. (1999). First onset versus recurrence of depression:
Differential processes of psychological risk. Journal ofAbnormal Psychology. 108. 483-489.
Lewinsohn, P., Hops, H., Roberts, R., Seeley, J., & Andrews, J. (1993). Adolescent psychopathology: I.
Prevalence and incidence of depression and other DSM-III-R disorders in high school students. Journal
ofAbnormal Psychology. 102.133-144.
Little, S. A.. & Garber, J. (2000). Interpersonal and achievement orientations and specific stressors
predicting depressive and aggressive symptoms in children. Cognitive Therapy and Research, 24,
651-670.
Lous, A., de Wit, c.,de Bruyn, E., & Riksen-Walraven, M. (2002). Depression markers in young children's
play: A comparison between depressed and non-depressed 3- to 6-year-olds in various play situations.
Journal of Child Psychology and Psychiatry and Allied Disciplines. 43. 1029-1038.
Lous, A., de Wit, C; de Bruyn, E., Riksen-Walraven, M., & Rost, H. (2000). Depression and play in early
childhood: Play behavior of depressed and non-depressed 3- to 6-year-olds in various play situations.
Journal of Emotional and Behavioral Disorders, 8. 249-260.
Meurs, P., & Cluckers, G. (1999). Psychosomatic symptoms, embodiment and affect, Weaving threads to
the affectively experienced body in therapy with a neurotic and a borderline child. Journal of Child
Psvchotherapy. 25. 71-91.
Monroe, S.• Rohde, P., Seeley, J., & Lewinsohn, P. (1999). Life events and depression in adolescence:
Relationship loss as a prospective risk factor for first-onset of major depressive disorder. Journal of
Abnormal Psychology. 108.606-614.
Murray, L. (1992). The impact of postnatal depression on infant development. Journal ofChild Psychology
and Psychiatry and Allied Disciplines. 33. 543-561.
Nolen-Hoeksema, S., Girgus J., & Seligman, M. (1992). Predictors and consequences of childhood
depressive symptoms: A 5-year longitudinal study. Journal ofAbnormal Psychology, 101,405-422.
Phares, v., Duhig, A., & Watkins, M. (2002). Family context: Fathers and other supports. In S. Goodman
& I. Gotlib (Eds.), Children ofdepressed parents. mechanisms of risk and implications oftreatment (pp,
203-225). Washington, DC: American Psychological Association.
Pine, D., Cohen, E., Cohen. P.• & Brook. J. (1999). Adolescent depressive symptoms as predictors of adult
depression: Moodiness or mood disorder? American Journal of Psychiatry. 156. 133-135.
Pine, D.. Cohen, P., Gurley, D.. Brook, J., & Ma, Y. (1998). The risk for early-adulthood anxiety and
depressive disorders in adolescents with anxiety and depressive disorders. Archives of General
Psychiatry, 55. 56-64.
Plomin, R. (1990). Nature and nurture: An introduction to human behavioral genetics. Pacific Grove, CA:
Brooks/Cole.
Poznanski, E., & Mokros, H. (1994). Phenomenology and epidemiology of mood disorders in children and
adolescents. In W. Reynolds & H. Johnston (Eds.), Handbook ofdepression in children and adolescents
(pp. 19-39). New York: Plenum Press.
Radke- Yarrow, M., & Sherman. T. (1990). Hard growing: Children who survive. In J. Rolf, A. Masten, D.
Cicchetti, K. Nuechterlein & S. Weintraub (Eds.), Risk and protective factors in the development of
psychopathology (pp. 97-119). Cambridge: Cambridge University Press.
Radke- Yarrow, M., & Zahn-Waxler, C. ( 1990). Research on affectively ill parents: Some considerations for
theory and research on normal development. Development and Psychopathology. 2. 349-366.
Rueter, M., Scaramella, L., Wallace, L., & Conger, R. (1999). First onset of depressive or anxiety disorders
predicted by the longitudinal course of internalizing symptoms and parent-adolescent disagreements.
Archives of General Psychiatry. 34. 1618-1628.
Shaffer, D., & Waslick, B. (2002). The many faces of depression in children and adolescents. Review of
Psychiatry, 21. 1-36.

187
Stern, D. (1995). The motherhood constellation: A unified view ofparent-infant psychotherapy. New York:
Basic Books.
Weiss, B., & Garber, J. (2003). Developmental differences in the phenomenology of depression. Develop-
ment and Psychopathology, 15, 403-430.
Weissman, M. M., Gammen, G. D., John, K.. Merikangas, K. R., Warner, v., Prusoff, B. A., & Sholomskas,
D. (1987). Children of depressed mothers: Increased psychopathology and early onset of depression.
Archives of General Psychiatry, 44, 847-853.
Weissman, M., Wolk, S., Goldstein, R., Moreau, D., Adams, P., Greenwald, S., et al. (1999a). Depressed
adolescents grown up. Journal of the American Medical Association, 281, 1707-1713.
Weissman, M., Wolk, S., Wickramaratne, P., Goldstein, R., Adams, P., Greenwald, et al. (1999b). Children
with prepubertal-onset major depressive disorder and anxiety grown up. Archives of General
Psychiatry, 56, 794-80 I.
Zero to three Diagnostic Classification Task Force (1999). Diagnostic Classification ofMental Health and
Developmental Disorders of Infancy and Early Childhood. Washington, DC: National Centre for
infants, toddlers and families.

188
 Chapter 7

'Closed Doors and Landscapes in the Mist' 2.

Depression in Psychoanalytic Developmental

Psychopathology: From Single Track Models

to Complex Developmental Pathways

Patrick Meurs, Nicole Yliegen. & Gaston Cluckers

Developmental psychopathology focuses on charting processes of, or pathways

into, psychopathology. It examines the biological, contextual and psychosocial pro-
cesses involved in the development of psychopathology, as well as the spontaneous
or therapy-induced processes of recovery and its biological, contextual and psycho-
logical correlates.
Cicchetti and Cohen (1995) indicate that the joint study of normal and patho-
logical development forms a crucial aspect of developmental psychopathology:

"Because all psychopathology can be conceived as a distortion, disturbance, or

degeneration of normal functioning, it follows that, if one wishes to understand
pathology more fully, then one must understand normal functioning against
which psychopathology is compared. Not only is knowledge of normal processes
 very helpful for understanding, preventing, and treating psychopathology, (... )
but also the deviations from and distortions of normal development that are seen
in pathological processes indicate in exciting ways how normal ontogenesis may
be better investigated and understood. (... ) Indeed, for many thinkers, the very
essence and uniqueness of a developmental psychopathology approach lies in its
focus on both normal and abnormal, adaptive and maladaptive, ontogenetic
processes." (Cicchetti & Cohen, 1995, p. 3-4).

In this sense, this discipline represents the study of unfolding illness processes
(paths of deviant, maladaptive development) in contrast to normally unfolding
development (paths of normal, adaptive development).
First, we will trace the development of this simultaneous approach to normality
and pathology in psychoanalytic thinking. Next, we will discuss the strengths and
weaknesses of the blueprint of this early developmental psychopathology project in
psychoanalysis. In the third section we will briefly outline the renewed interest in
developmental psychopathology within psychoanalysis. This is followed by an
illustration of the principles of a psychodynamic developmental psychopathology
of depression by means of a clinical vignette. Finally, future challenges for
psychodynamic developmental psychopathology are discussed.

Rise and Stagnation of the Developmental Psychopathology Project

in Psychoanalysis

The simultaneous approach to normal and pathological development is deeply

rooted in psychoanalytic thinking. We can see this in a quotation from A psycho-
analytic perspective on developmental psychopathology (A. Freud, 1974, p. 63-64):

"We have broken with the tradition according to which every mental difficulty is
seen and explained by comparison with severe pathological patterns, and, in-
stead, try to see it against the background of the norm, expectable for the parti-
cular child's age, and to measure its distance from it. (... ) I have attempted to do
this in 1966, in my book Normality and Pathology in Childhood: Assessment of
Development, by introducing the concept of Lines of Development."

The psychoanalytic contribution to developmental psychopathology, however, is

less well known than one might expect from this quotation. In order to understand
this, we will first make a historical digression.

190
 The term "developmental psychopathology" was coined in 1974, both by Anna
Freud in psychoanalytic literature and by Achenbach (1974) - author of the "Child
Behavior Checklist" (Achenbach, 1991) - in experimental clinical child
psychology. As a result, it is not immediately clear who can assume the copyright
to the term. However, a central concept of developmental psychopathology -
developmental lines - is without a doubt attributable to Anna Freud. It had already
appeared in an article of hers dating from 1963. The above quotation from 1974
also makes it clear that the idea of developmental psychopathology - with the
simultaneous study of normal and pathological lines of development - was already
firmly entrenched in the research supported by Anna Freud.
The roots of the developmental psychopathology notion in psychoanalysis do
indeed go back a long way. While Anna Freud (1966) used the concept of
developmental lines as a key concept in one of her principal works - Normality and
pathology in childhood - she had previously used the term developmental distur-
bances quite frequently. This was still a fairly wide concept, referring mainly to a
"disturbance, characterized by deviation from normal development" (A. Freud,
1963, p. 44-45). Anna Freud had already started using this concept during the
Second World War. Since 1942, she had been observing large groups of children
who had been orphaned by war violence, as well as children who presented
problems after having been temporarily separated from their parents and placed in
foster homes in the countryside in order to escape from the German bombing of
major British cities. She learned to look, in these children, for the first signs of
psychological problems, the further evolution of these symptoms, both in their
spontaneous course and under the influence of more specific care and help. What
particularly interested her - and here our historical digression moves closer to the
question of depression - were the mourning responses and the responses to
separation and loss, more specifically early childhood expressions of grief and
depression due to separation from and loss of the beloved care figure. She also
searched for a better understanding of the influence of certain characteristics in the
child or the environment, leading to an elimination or neutralizing of previous
depressive symptoms or, on the contrary, to their emphasis and fixation
(Burlingham & Freud, 1944). Within the context of numerous case descriptions, all
kinds of familial and individual antecedents are described, which could promote or
impede the processing of the complex feelings regarding separation from or loss of
the parents. In that sense, Anna Freud had a good feel for the protective and risk
factors in the socio-emotional development of these "children without families".
Early vulnerabilities and sources of resilience, as well as their influence on the
subsequent evolution of the mourning process or on the child's further
development, are also extensively covered. Research into the normal and deviant
developmental pathways of sad or mourning children, as well as into the factors
influencing this and into the widely varying expressions of depression throughout
childhood, was central to Anna Freud's work in that period.
Her observational research, which began during the I940s, was to a certain
extent a crystallization of an academic project for psychoanalysis written by Heinz
Hartmann (1939) in Ego psychology and the problem of adaptation. To Hartmann,
the simultaneous study of normal and deviant development was an absolute

191
priority.' At the same time as Anna Freud's interpretation of this project, within
psychoanalytically inspired developmental psychology we find this emphasis on
the simultaneous study of normal and deviant phenomena in The first year of life:
Normal and deviant development of object relations. the basic work of Spitz and
Cobliner (1965).
In one of Anna Freud's latest papers, dated 1981, she indicates how much
importance she attaches to the study of early development and to the recognition of
early signs of deviance. The possibilities of prevention or early intervention were
already widely known at that time, but for refinement and adaptation of early
developmentally oriented preventive guidance and child psychotherapy, an ongoing
and far-reaching research project covering normal and deviant development was
needed. Anna Freud was to construct an extensive dataset for this project, among
others, in the form of the Hampstead Clinical Index, which is certainly under-
exploited even today. By using observation methods and setting up datasets in
several developmental domains, Anna Freud became keenly aware of the sticking
points in this academic project; for example, she organized the data of the Clinical
Index in such a way that they could be used for group research or for multiple
measurements within single cases, being aware of the need to undertake statistical
research procedures within psychoanalysis. She also points out how difficult it is to
make a distinction between normality and deviance during earliest child
development and, certainly, during the first phases of a process that could be the
onset of a maladaptive developmental line. She therefore advocates in no uncertain
terms an intensification of research into this aspect:

"What concerns us today are the many characteristics expected from the average
adult which (... ) are described as to their end products, but for which no develop-
mental prestages are itemized. This omission not only leaves a gap in develop-
mental theory; it creates also the false impression that such achievements are
come by easily, in fact that they are simply the result of smooth, undisputed, i.e.
nonconflictual maturation." (A. Freud, 1981, p. 14)

Another important concern of Anna Freud is to demonstrate that development

does not progress in a linear fashion, but with saltatorial developmental changes or
shifts. After such a change, the child needs some time to adapt to a new, more
complex way of functioning. Throughout these changes, development is progres-
sive or it is being transformed and organized in a complex, hierarchical way, but it
also has normal moments of stagnation, contributing to a sense of homeostasis. Nor
is development synchronous in all domains or along all of its lines, as is clear from
the developmental profiles of children. Moreover, development is embedded in
I "His interest in the conflict-free aspects of the ego, relates. he emphasizes time and again, to this desire

that psychoanalysis becomes a general psychology. and not merely a special theory of psychopathology."
(Yankelovich & Barrett, 1970, p. 108).

192
affective-relational dynamics and complexes from childhood, making temporary
stagnation or even regression more understandable. Despite the great effort Anna
Freud had made on the level of developmental psychopathology, her approach had
insufficient following in psychoanalysis. She complained about this at the end of
her life:

"( ... ) I have attempted to do this by introducing the concept of Lines of

Development. (... ) But the list given was by no means considered to be
complete; rather, it was in fact meant as an invitation to the readers of the book
to make their own additions to it. This challenge has so far not been taken up by
any other author." (A. Freud, 1974, p. 64)

Contrary to what Anna Freud suggests, there have been important publications
on developmental psychopathology within psychoanalysis in that period (e.g., see
Nagera, 1970, 1981). But she felt that the domain of psychoanalytic developmental
psychopathology had not evolved into a unified research project. Within certain
influential circles in psychoanalysis, observation studies of children remained
highly controversial until very recently. This empirical methodology on which John
Bowlby (1981), in addition to Anna Freud and Rene Spitz (1945), also based his
work on separation and loss of the attachment figure, and which Robert Emde
(1981) and Daniel Stern (1985) also used as a basis for their work on socio-
emotional development and on the development of the self in contact with the other,
is unfortunately still not accepted in some schools in psychoanalysis today. This
methodological problem in psychoanalysis has been addressed more systematically
in recent years, while empirical research is also strongly advocated by a growing
number of "scientist-practitioners" in psychoanalysis (Barron, Eagle, & Wolitzsky,
1990; Shapiro & Emde, 1995; Emde, 1999; Fonagy & Target, 2003).
Moreover, in the 1960s and 1970s, psychoanalysis lost its dominant position
within mental health care, as a result of which the knowledge acquired within
earlier psychodynamic developmental psychopathology was not adequately
disseminated nor fully exploited or valued. This "political" element, together with
the large methodological differences between psychoanalysis and other theoretical
approaches in psychology meant that psychoanalytic developmental psychopatho-
logy was either left out of the picture or was only quoted as an aside - an
historically interesting footnote - within the broad and well known research domain
which developmental psychopathology - at the intersection between child
psychiatry, developmental psychology and clinical psychology - had now become,
outside psychoanalysis.
Only recently, Peter Fonagy and Mary Target have placed psychoanalytic
developmental psychopathology more accurately and systematically on the map,
initially through specific psychoanalytic contributions to handbooks of
developmental psychopathology (see Fonagy, 1995a, b; Fonagy & Target, 1997),

193
and subsequently with their book, Psychoanalytic theories: Perspectives from
developmental psychopathology (Fonagy & Target, 2003).
Inspired by these trailblazers, we are setting out in search of the foundations for
a psychoanalytic developmental psychopathology of depression with the aim of
taking up the dialogue with mainstream developmental psychopathology. Our
starting point is formed by the two dimensions of depression - anaclitic and
introjective - as described by Blatt (1974), who recently re-emphasized the need to
work out a developmental psychopathological perspective on these dimensions
(Blatt, 2004).
We are looking more specifically for the developmental line of normal anaclitic
attachment and of maladaptive anaclitic depression as well as for the developmental
line of normal introjection of moral imperatives and ideals (normal normative
functioning) and of maladaptive introjective depression (pathological normative
function).

Emergence of the Blueprint for Psychoanalytic Developmental Psychopathology

Historical Background: Some Founding Moments

Freud's introductory lectures to a developmental perspective on psychopatho-

logy. The developmental perspective - referred to in the early years of psycho-
analysis as the "genetic perspective" - did not form one of the three key principles
quoted by Freud (1900) in his first model of the human psyche, in which he
distinguished a dynamic, economic and topographical perspective. The structural
and genetic perspectives were added later. This "late" addition means that the
developmental perspective is not regarded as obvious in psychoanalytic theory. This
is undoubtedly one of the reasons why in some psychoanalytic theories pathways
of depression are not automatically developmental pathways (cfr. infra) and why
developmental psychopathology was not given an enthusiastic welcome in all
psychoanalytic circles.
Freud's innovation was to place the emphasis on early childhood, with the
Oedipus complex as the core complex. His first case studies were aimed at
demonstrating that psychopathology in adulthood is the result - by deferred action
(Nachtriiglichkeit) - of a long and complex pathogenesis. Some systematically
repressed wishes remained active in the deeper layers of the psyche and could
determine, in a complex way, neurotic difficulties later on in life: the Oedipal
dramas from childhood are repeated and worked through in the context of mature
adult psychosexuality.
While female neurotic patients originally played an important and inspiring role
in developing psychoanalytic models, in case discussions from 1909 onwards Freud

194
describes chiefly male patients. Once Freud concentrates on the psychosexual
development of men, he becomes aware not only of the great importance of the
aggressive sub-component of psychosexual development, but also of the previous
history of it in the earliest Oedipal phase.
In the case description of Little Hans (Freud, 1909a), for example, the libidinal
origin of Hans' fears and phobias is set in the broader context of the Oedipus
complex. This core complex appears to imply a genuine development, a path
throughout the Oedipal phase, from the earliest Oedipal strivings to the later
Oedipal phase: Freud makes several remarks on the previous period in which the
symptoms of five-year-old Hans already existed in a milder version. Sexual
curiosity, problems with aggressive feelings and a somewhat anxious or shy attitude
were already present in three-year-old Hans, from the beginnings of the Oedipus
complex on.
This pathway perspective throughout the Oedipus complex cannot yet be
detected in the case study of Dora published in 1905 (Freud, 1905). In the case
study of the Rat Man, published also in 1909, the developmental perspective had
become already very clear: the extreme repression of the aggression toward the
father, activated in the context of the Rat Man's sexual desire for his mother, can
only be understood when in the more archaic early phases of development other
defensive operations against the sexual and aggressive impulses must have been
activated. In this text, Freud (1909b) is rather vague on the exact timing of these
"archaic" mechanisms preceding the extreme repression of the early Oedipus
complex.
When Freud again writes about female psychosexuality in 1931, toward the end
of his life, this developmental perspective has become even more evident, and more
broad: the same word "archaic" now explicitly refers to the pre-Oedipal develop-
ment (see Assoun, 1981). Consequently, the developmental line is considerably
expanded again: from pre-Oedipal roots through early and late Oedipal influences,
to ultimate symptom formation (by "deferred action"). At the same time, Freud has
noticeable difficulty imagining the continuity in this developmental line. He
presents it as if two discontinuous worlds are involved: the pre-Oedipal and the
Oedipal development as "two civilizations from different times in history" (Freud,
1932, p. 225-226). A cut-off point or fracture exists between these two "worlds".
Freud leaves us with this image of discontinuity in psychosexual development, at
least in his later writings, which deal explicitly with female sexuality and, again,
hysteria.
In papers dealing with other clinical syndromes, we implicitly find a different
view of continuity and discontinuity throughout development. In his work on
paranoid psychosis for instance (Freud, 1911, 1922), but definitely in his work on
depression and melancholia (Freud, 1917a), the question is raised of the influence
of pre-Oedipal and Oedipal development on the emergence of psychopathology.
Moreover, in this period important theoretical adjustments occur - the reformula-
tion of the drive theory (Freud, 1920) and the description of the structural model in
his second topographical model of the human psyche (Freud, 1923) - which cause
him to give more weight to the pre-Oedipal preliminary phase. For example, the
formation of moral consciousness, which Freud originally (e.g., Freud, 1905)

195
situated at the end of the Oedipus complex, is situated much earlier. In his writings
between 1925 and 1930, continuity between the pre-Oedipal and Oedipal contri-
butions to superego development is suggested.
We can say that Freud worked hard to construct a developmental perspective,
rethought it several times, developed it, frequently hesitated and was even some-
times contradictory about it, depending on the context within which he was
discussing it. We will now look in more detail at his developmental perspective of
depression itself.
We start with his Mourning and Melancholia (Freud, 19l7a). Here, he disrin-
guishes two types of depression: neurotic and narcissistic depression. The role of
aggression in the onset of depression receives much of Freud's attention. More
specifically, the fundamental characteristic of affective ambivalence - the fact that
human love is never perfect, never fully positive - is referred to by Freud as the key
aspect in the psychodynamics of depression. People suffer from depression because
of their ambivalence, because of the aggressive component of their love.
Usually, people can tolerate this imperfection in their love lives (ambivalence
tolerance). They can keep the aggressive component of their love in the background
(and yet remain sufficiently in touch with it to mobilize aggression against the love
object, where necessary) or use the aggression constructively, for example to make
their loving feelings powerful and assertive. Furthermore, in 1917 Freud (Freud,
1917a) describes another way of making constructive use of ambivalence: people
use their aggression to go through the process of mourning. Just like separation,
mourning demands some aggression: in protesting against the loss, in assigning the
blame for having been left alone, in daring to feel the gift of still being alive, in
accepting that the link with the lost object does not have to be idealized. However,
depressive people cannot tolerate the imperfection of their love. When they are
confronted with loss or even the threat of it, they massively suppress aggressive
feelings toward the beloved object. Moreover, these feelings are turned against the
self and only in this way, in the form of self-reproach, can they resurface.
Aggression, therefore, can only be experienced at a conscious level if it is aimed at
the self.
Self-reproach and self-criticism for not having taken good enough care of the
loved one, the feeling that I am responsible for the loss of the other, that I did not
do enough to prevent the loss - these are clinical manifestations of turning
aggression toward the self although it was originally aimed at the other.
Freud's text from 1917 may well be criticized because of its rather limited
view of the etiology of depression, but from a phenomenological point of view
the text is still highly relevant. Particularly the comparison between normal,
neurotic and melancholic (narcissistic) mourning processes is of the utmost
importance to the developmental psychopathology perspective we have in mind.
In melancholia - labelled a narcissistic neurosis by Freud - aggression against
the loved one and the self is primitive (oral) in nature. As a result, the aggression
is even more destructive than in psychoneuroses, as an example of which Freud
uses obsessional neurosis in this text. In obsessional neurosis, the aggression
is of an Oedipal nature, with possible regression to the anal-sadistic realm. The
primitive nature of oral aggression can lead to more destructive self-punishment

196
and self-criticism in melancholic patients, which increases the likelihood of suicide
compared to obsessional neurotics without melancholic psychopathology,
according to Freud (l917a).
In another text from 1917, Freud (1917b) clearly indicates that unfortunately he
is not yet in a position to give us more information about the difference between
these two levels of pathological ambivalence, the neurotic and the narcissistic. They
produce different mourning processes, while the aggression involved has another
developmental origin (oral versus Oedipal). In a certain sense, in this text Freud set
out a program for further theoretical and empirical research on depression due to
primitive (oral) sadistic love or to ambivalent (Oedipal) neurotic love. Other authors
have attempted to fill in the theoretical gaps which Freud (1917b) left at the time,
citing the influence of "earlier" oral themes and "later" Oedipal aspects in
depression, the difference between neurotic and pre-neurotic depression, and the
assumption that another developmental layer is associated with every form of
depression. Freud was the first to suggest that two levels of depressive
psychopathology are linked to two different developmental stages: depression in
narcissistic disorders is about oral sadism while depression in neurotic disorders is
about Oedipal ambivalence. We will now see how Abraham and Klein continued
this line of thinking.

Abraham and Klein: Expansion of Freud's developmental perspective into the

earliest developmental phases. For Freud, moral conflicts about aggression towards
the beloved object should be placed within the context of the Oedipal phase of
psychosexual development. At the end of this phase, the child is usually able to
understand in a certain way the emotional contradiction in the fact that the beloved
object is also treated aggressively. The child evaluates his love and hate self-
critically and sees that it has not met the ego ideal or the normative standards of the
superego. The Oedipal phase is the period during which the child can gradually
grasp the contradiction between love and hate, as well as the conflict between
sexual desire and prohibition. Normally, at the end of the Oedipal phase, the child
makes a judgement of condemnation. Freud (1909a) described this mild
mechanism of defense only once. It is a mechanism that is almost conscious in
nature: henceforth, the normally developing child is better able to make a
judgement and to rein in aggression towards the parents. Pathological development
is characterized by repression, occurring early in Oedipal development, as is the
case with the phobic problems of little Hans (repression as a more extreme
mechanism of defense, unconscious in nature), possibly even preceded by a
primitive isolation or separation of both tendencies of love and aggression, as is the
case with the obsessional neurosis of the Rat Man. In this latter case, the anal-
sadistic component of the love was of a particular kind already in the pre-Oedipal
development phases. But Freud hasn't given us a further answer about these early
pre-Oedipal precursors of problems with aggression and their influence on self-
reproach and self-evaluation. He only knew that this raised important questions,
since he gave a lot of thought to the pre-Oedipal love-hate balance in texts on
paranoia (Freud, 1911, 1922).

197
 Since 1917, other authors have looked at these precursors in particular,
more specifically at aggression and its role in early moral functioning. For instance,
Karl Abraham (1921/1953a, 1924/1953b) and Melanie Klein (1928) des-
cribed, respectively, the sphincter morality of the anal phase and the depressive
crises in the context of archaic Oedipal anxieties in the second and third years of
life.
Once the child realizes for the first time that its sadism is targeted at the same
vitally necessary and beloved object (e.g., the mother) and that the child itself
consequently has not only loving, positive tendencies within him, but also
aggressive, negative tendencies, the child reaches the depressive position. At a
certain moment in development - Abraham and Klein originally place this in the
second part of the second year of life - it is possible to see in most children that
aggression towards the beloved objects no longer remains without moral conflicts
or critical self-evaluation. As a consequence of this early moral reaction, the child
can protect the mother from its aggression.
These first moral reactions are an expression of an awareness of the gap between
the (extreme) ideal self-image and the "harsh" reality of the "ambivalent" self: the
child not only loves the mother but also causes her harm. Moreover, the child
gradually realizes that it previously did this in an unconcerned and sometimes
ruthless way (to use one of Winnicott's concepts; Winnicott, 1945/1958). The
capacity for empathy or concern begins to develop, while aggression is better
neutralized, channelled and merged with love.

Klein (1935, 1940) describes these first moral reactions in child development
within the context of her theory of the depressive position. The first time a child
realizes that the object towards which love and hate are targeted is one and the
same, it leads to a painful affective crisis which prompts the child towards
reparational tendencies. From then on, the mother becomes more acknowledged
and accepted as a differentiated person with her own rights (a whole object); she is
no longer an object in the service of the self (a partial object). The central fear of
the child at this age - Klein has placed this moment gradually earlier in
development - is depressive in nature: the child fears having injured the mother
object, banished it from himself or even destroyed it, and now more keenly realizes
that he is not a perfectly loving child. The child first realizes his affective
ambivalence when he has a certain sense of the consequences of his aggression. He
now sees that the mother is not perfect and does not entirely coincide or merge with
the child; therefore she cannot be controlled in an omnipotent way. Moreover, the
child realizes that a gap exists between the real self and the ideal self. This leads to
feelings of guilt, moral reflexes and a genuine desire to repair, initially in the form
of perfectionism or of a misunderstanding of the gravity of the consequences of the
ambivalence (obsessive versus manic defenses against depressive pain), later in
more realistic attempts at reparation.
The desire for perfection - the obsessive defense - is an understandable first
attempt to deal with the ambivalences and to close the gap between ideal and real

198
self. In normal development, this quickly assumes the form of "wanting to be as
perfect as possible but also accepting that this can never succeed: a certain tolerance
of imperfection and ambivalence, a negative capability" (Bion, 1962, p. 118). Or,
as Ricoeur (1971, p. 34) has put it: "accepting the best possible human world"
instead of "reaching after a perfect world, beyond our human capabilities". In the
acceptance process, the ideals of the child, step by step, become more realistic and
have a motivating effect, while the reality of imperfection becomes less dramatic
and less de-motivating.

The Emerging Model: Single Developmental Lines into Pathology

and One-To-One Associations Between Development and Pathology

In New Introductory Lectures on Psychoanalysis, Freud (1932) praised Abraham

for the distinctions he made within the pre-Oedipal developmental phases. He
particularly emphasized the developmental changes in the aggressive aspect of
psychosexual ontogenesis and the implications these changes have on the develop-
ment of guilt, morality and prosocial tendencies. Important steps in moral develop-
ment are clearly observable in the middle of the anal phase and toward the latter
part of the Oedipal phase.
In the middle of the anal phase, the child can no longer approach the loved object
in an unconcerned ruthless (and sometimes aggressive') way. Abraham's theory of
the first development of guilt as a depressive anxiety, and as a motivating element
for reparational tendencies in the latter anal phase, has been taken up by Klein in
her theory of the onset of the depressive position, situated in the earliest moments
of the Oedipus complex (in her theory coinciding with what traditionally had been
described as later oral phase and anal phase).
These first moral reactions are joined in a subsequent developmental phase by
guilt and concerns over the Oedipal ambivalence. In this sub-phase, when libidinal
tendencies have become dominant in a genital sexual organization, moral
consciousness is strengthened to such an extent that the object is better safeguarded
from aggressive impulses. Ambivalence is diminished.
In Abraham's theory, the concepts of pre-ambivalent, ambivalent and post-
ambivalent phases became associated with this threefold division: the pre-
ambivalent development from birth to the middle of the oral phase; the ambivalent
development from the middle of the oral phase to the later Oedipal phase; the post-
ambivalent development from the later Oedipal phase onward. This blueprint was
initially adopted as such by Klein in 1935 (the paranoid phase that precedes the

, Aggression in this sense means a non-intentional epiphenomenon of the narcissistic phases of

development. In the middle of the anal phase, an enormous shift is made towards object relatedness. This
brings it about that aggression can be kept at bay and that it can be used more intentionally. This
developmental change in the line of aggression is associated with a stronger awareness of one's own
responsibility for aggressive acts of the self against the object: a first clear sign of moral development.

199
depressive position that is passed through toward a post-depressive position). The
fact that Klein and other psychoanalysts subsequently changed the timetable of this
and adjusted their conceptual framework does not eliminate the threefold structure
(functioning according to a paranoid-schizoid mode, a depressive mode and a post-
depressive mode). We will now explain this threefold division in early childhood
development which is still widespread in psychoanalysis and which has been linked
to levels of psychopathology.

General idea: Three developmental phases connected to three levels ofpsycho-

pathology. The idea persists in both Abraham and Klein that, at a certain point in
development, the perception of the aggressive sub-component of the libido brings
about a morally inspired self-evaluation and a depressive reaction. They place this
in the ambivalent phase of libidinal development. A series of well-known
psychoanalysts such as Jacobson (1953), Zetzel (1965), Mahler (1972), Parens
(1979) and Kernberg (1984) each described this threefold division in the devel-
opmental perspective in their own way. As a result, this three-phase model -- of
which Abraham is the "godfather" - recurs in variants in many authors who link a
developmental perspective to their psychopathological theory. For example,
Winnicott (196011984) links various forms of psychopathology to either the phase
of absolute dependence, to the phase of relative dependence or to the Oedipal
phase. Fairbairn (1941/1952a) distinguishes between the phase of infantile
dependence, the transitory phase and the phase of complexity in relationships, three
moments on the developmental line from infantile to mature dependence. Mahler
(1972) and Mahler, Pine and Bergman (1975) refer to the pre-rapprochement phase,
the rapprochement phase and the Oedipal phase, while Blatt (1974) distinguishes
between psychopathology associated with problems with object permanency,
object constancy and affect constancy. Parens (1991) talks of the first precursors of
ambivalence, the first (pre-Oedipal) and the second (Oedipal) generation of
ambivalence conflicts.
These three compartments in development are not always comparable at the level
of content and time, but they are in terms of structure. Psychoanalytic psycho-
pathology theory depends on this structure: for the psychoses, borderline conditions
and neuroses, crucial factors are sought in, respectively, the earliest, central and
later phases of early childhood development. The underlying assumption is: the
more serious the psychopathological syndrome, the earlier in development the
deviation from a normal path must have occurred. This three-level framework
certainly is a strong element of psychoanalytic theory; the way they are associated
to levels of psychopathology raises important questions, as we will argue further
on. The strengths and limitations of this framework will be discussed later in this
chapter. However, we first discuss the work of two authors (Glazer, 1979;
Antonovsky, 1991) who have tried to link depression to the different developmental
steps at which deviance from normal development can occur. Their work can be
situated in the low-profile period of psychoanalytic developmental psychopatho-
logy, between the mid-seventies and the early nineties. Both authors situate the

200
etiology of depression within the perspective of the development of personal
standards, more specifically the ego ideal and the superego.

Glazer's Theory of the Developmental Trajectory Towards Object-Related

or Narcissistic Depression

Glazer (1979) distinguishes two levels of depression, "object-related depression"

and "narcissistic depression". His attempt to understand the pathogenesis which
leads to two levels of depression in a developmental context immediately brings us
a step closer to the developmental psychopathology project we are seeking.
Glazer's starting point is that ideal and reality are not the same. One possible
reaction to this is depressive in nature, heralded by the helplessness of the self
which is confronted with the gap between ideal and reality, as well as by the
possible aggression which is provoked as a result, or which is involved in some
other way. Moreover, this aggression can be turned against the self because in the
ideal self image, for example, there is no place for anger towards the object.
Glazer also raises the question of how we can understand, from the helplessness
of the self, that depression can be narcissistic or object-related in nature. He argues
that the two deviations in the direction of depression start from the same
developmental line (the normative function line), but from different locations or
developmental moments. Particularly as a result of the latter, the pathway into
different levels of depression takes on the nature of a true developmental line.
According to Glazer, the central aspect of object-related depression is that the
ambivalence with respect to the object is too conflicting. In order to avoid
helplessness with respect to this ambivalence, the self will coincide with the object
in an extremely positive symbiotic relation within which aggression is no longer
present. However, the helplessness does not disappear because the self feels
restrained (inhibited) and the possibility is lost of making creative use of aggression
within an individualized relation of an autonomous self with the differentiated
other. The object relationship itself is now no longer at risk, but its differentiated
nature is. In other words, the object relationship is distorted because certain aspects
of it are unbearable.
By contrast, in narcissistic depression, helplessness in the self arises as a result
of the threat of loss or destruction of a relationship with a self-object. That self-
object is vitally necessary to the homeostasis of the self; moreover, it is an object
with which the person was previously merged. In this form of depression, the object
relationship itself is at issue: it is not a question of a distortion of the object
relationship between a differentiated self and object, but rather of the loss of an
undifferentiated link with an object, as a result of which the self is also damaged or
emptied.
Glazer emphasizes that, in the developmentally more advanced object-related
depression, helplessness and inhibition are a consequence of affective ambivalence,

201
which is a source of conflict for the self. More realistic individuation should be
worked on during therapy, within which positive and negative components can be
present simultaneously or can be integrated. In the developmentally more primitive
narcissistic depression, the helplessness is much more of a reaction to the threat that
the link with an as yet unseparated but vitally necessary self-object will be lost.
Treatment in this case will primarily focus on safe separation, with the presence of
the object in the background, rather than on more complex individuation with a
differentiated relationship to whole objects and with an integration of negative and
positive elements.
As can be seen from the terms used, Mahler's developmental framework is
introduced by Glazer in order to put the cause of the helplessness into its proper
context: unbearable aspects of individuation (aggression and ambivalence) lead to
object-related depression characterized by the fear of loss of approval by the
superego or the moral authority; unbearable aspects of separation lead to
narcissistic depression characterized by the fear of loss of the relation with the self-
object. In other words, while a normative conflict comparable to the conflicts from
the rapprochement subphase forms the basis for object-related depression, narcis-
sistic depression is linked to an unbearable loss of a self-object, set within the
context of the development of the pre-rapprochement subphase.
The developmental perspective is not taken further by Glazer. He stops with the
role of the normative function in the emergence of the two levels of depression. In
object-related depression, the ideal is disrupted by ambivalence, which is inevitable
in a complex, integrated picture of self and other; in narcissistic depression the ideal
is disrupted by aggression, which could destroy the vitally necessary extremely
idealized merger, but which is also needed to come to any differentiation between
self and other. In object-related depression, the discordant aggression is repressed;
in narcissistic depression unbearable aggression is isolated and split off. In Glazer's
words, it is a matter of two forms and levels of deviance of the same developmental
line: that of moral development. On this developmental line, aggression is averted
in a phase-specific way - in one case because the moral conscience rejects the
affective contradiction of love and hate (due to unbearable guilt anxiety), in the
other case because, within the most primitive form of moral consciousness, ag-
gression evokes fear of revenge and destruction (annihilation anxiety).

Antonovsky's Theory of the Depressive Experiential Process

Anna Antonovsky (1991) discussed the emergence of depression in a summary

article about "the holding of ideals and idealization". In this article, she emphasi led
how important it is to be able to make the transition from idealization to the holding
of ideals, from extreme to realistic ideals or moral standards.
The first love of the child for the parental figures and the self is highly idealizing,
but this idealization of self and others is gradually tempered in normal develop-

202
ment, giving way to the holding of ideals. Holding ideals is, in other words, more
realistic than idealization. According to Antonovsky, a primordial etiological factor
in depression is found in problems related to making the transition from (extreme)
idealization to tempered, more realistic ideals. Personal standards and moral
judgement remain strict as a result, rather than presenting a mildness; people suffer
from these extreme, unrealistic ideals, while self-reproach and self-punishing
trends are prominently present. In depressive illness, people become stuck on the
line from extreme idealization to the holding of realistic ideals.
Ego psychology chiefly stressed the adaptive aspects of covering the path from
idealization to the holding of ideals. By the holding of ideals, development is
facilitated and a better sense of reality is established as well as a more integrated
superego in which the complexity of a mature personality structure can be
expressed.
From a clinical point of view, Jacobson (1964, 1971) is of the opinion that the
development of the normative function is driven by a back-and-forth movement on
this line between "idealization" and "the holding of ideals", from idealizing love
towards holding realistic ideals concerning the other or the self. On the other hand,
there is the reverse trend in which a continued longing for more idealistic circum-
stances and representations of self and other is expressed. The first movement (from
extreme to realistic ideals) is never completed; the second (from realism to extreme
idealism) is doomed to disillusionment. What is crucial is that, in normal
development, de-idealization becomes acceptable, thereby tolerating the tension
brought by the realization of imperfection. After having longed for more perfect
conditions, gradually a re-libidinalization of the imperfect picture of the self and the
other occurs, described by Jacobson as a more mature kind of idealization, and
regarded by Antonovsky as "holding ideals in an adaptive way". This new libidinal
cathexis of "the fallible other" or "the fallible self' is an adequate counterbalance
to the yearning for the perfect situation. Against this background of sufficient love
for the imperfect, the gap with the perfect ideal is tolerated, people feel happy
enough with the reality of imperfections, without misrepresenting them too much,
but with the possibility of striving for certain unfulfilled objectives which lie within
the bounds of possibility and of evaluating endeavours and possibilities sufficiently
realistically.
A crucial change, which launches the subject on the path to depression, is that
the tension which brings the necessary disillusionment of idealization is not
tolerated. It is replaced by a feeling of helplessness. Moreover, aggression can also
be present within this helplessness in the form of reproach against the self and the
other. The helplessness of the self, confronted with the inability to reach or maintain
a highly ideal situation, opens up the way to depression.
Among other psychoanalysts, who chiefly stress the drive-related background of
depression, even more attention will be devoted to the aggressive component in the
etiology of depression. According to Melanie Klein (1945), for example, the
idealized object arises as a result of splitting, the mechanism which helps to defend
against aggressive impulses. These are then attributed to a persecutory ("bad")
object, while the beloved object appears perfect or highly ideal. In a normal
experience process, integration of both parts of self and other will then follow in the

203
depressive positron. In this integration, the extreme idealization of the good
object can be renounced. Goodness is no longer attributed to a bad other; bad
aspects in the good object are no longer defended against. The representations
of the self and the other become less extreme and more subtle or complex;
they become more realistic too. The bad sides of the self and the object become
regarded as an inevitable part of the whole and are also less unbearable. How-
ever, this is only true in the case of normal integration of the aggressive, negative
self and object parts into the loving, positive parts. In the depressive crisis,
by contrast, people experience helplessness, a complex affective crisis of guilt,
shame, pining and self-reproach about their own aggression and imperfec-
tions. More than anything, they want to cure the crisis but believe that this can
only be done by returning to the perfect situation and by undoing the imperfection
in an omnipotent way, rather than by accepting imperfection and negativity in love
or by acknowledging that this imperfection is simply an inevitable component of
human relationships. Imperfection is still harshly judged. It is impossible to reach
the desired perfect situation and this realization leads to the feeling of helplessness;
in that case, the depressive position cannot be tolerated, leading to depressive
illness.
While ego psychology primarily regards depression as a reaction to a situation
of extreme helplessness of the self (Bibring, 1953), drive psychology places more
emphasis on the role of aggressive impulses in the induction of this helplessness.
However, ego psychology and drive psychology concur on what they refer to as
crucial etiological factors. The emergence of depression is situated in the
intervening moment of disillusion with extreme idealization, the inevitably
associated helplessness and the (lack of) tolerance for ambivalence associated with
this process of disillusionment.
The experiential line of depression therefore runs from extreme idealization to
the holding of realistic ideals, with several steps in the development of tolerance of
disillusionment and of ambivalence along the way. The line leading to depression
crosses clear problems of disillusionment with extreme idealization. These are
expressed in ambivalence-intolerance. The imperfection and ambivalence are not
tolerated and, at the same time, one realizes that the extremely ideal situation is no
longer attainable. This conflict becomes a trap; the self feels helpless because it has
no way out since the extreme idealistic images are unrealistic and the realistic
images are intolerable.
Antonovsky offers us a "developmental line perspective" on depression in the
sense that she describes in great detail the course or development of the experiential
processes leading to depression. In principle, the experiential pathway she mentions
is applicable to several moments in the development from more extreme to more
realistic moral or normative structures. However, the object of study is not about the
critical points in this developmental line from idealization to the holding of ideals:
Antonovsky's brilliant study is about lines of depressive experience.
Comparing Glazer and Antonovsky, we could say that in a certain way a change
in emphasis has occurred, from a developmental psychopathology at the end of the
seventies to a theory of dynamic and experiential pathogenesis. Both contributions
are situated within another context: Glazer was writing at a period when searching

204
for the crucial developmental line of a psychopathological syndrome was still in
vogue, Antonovsky at a period where this focus was less of a concern.

Guiding Principles for an Up-To-Date Psychoanalytic

Developmental Psychopathology:
From Single Trajectory to Complex Interacting Developmental Lines

Several events inside and outside psychoanalysis brought the developmental

psychopathological project back into prominence. The broader acceptance of
empirical research methods in psychoanalysis is one of these events: complexity in
longitudinal processes can be investigated, combining data from several variables
(for example, different developmental domains or pathways). This broadens the
existing model of developmental psychopathology: pathological outcomes can be
understood as the results of changes and deviations on several developmental
pathways and on several crucial moments on these pathways. This carries
psychoanalytic developmental psychopathology away from the idea of one single
track into psychopathology, possibly from one crucial moment of deviance on. This
evolution in psychoanalysis is nurtured to a large extent by the vigour of empirical
(experimental) developmental psychopathology within clinical psychology,
developmental psychology and (child) psychiatry from the second half of the
eighties on (Cicchetti, 1984, 1989; Cicchetti & Toth, 1992; Sroufe & Rutter, 1984).
In the field of the psychoanalytic depression literature Blatt (1974) already
advocated this new line of thinking in the mid-seventies; it is no accident that this
developmental psychopathological element of Blatt's work received an enormous
new impetus from the nineties on.

Blatt's Perspective on Depression as a Pathological Outcome of Two

Developmental Lines: Introjection and Anaclitic Relatedness

Blatt expands upon the developmental perspective formulated by Glazer in two

ways. On the one hand, he specifies that two developmental lines leading to two
forms of depression are involved, rather than talking about two levels of deviance
from one and the same developmental line (i.e., the development of personal
standards). Introjective (self-critical) depression is conceived as a deviation from
the line of development of self-definition, including the development of personal
standards and morality; anaclitic (dependent) depression as a curve on the line of
object relational development (Blatt & Shichman, 1983). Subsequently, he even
talks about two dimensions which are present in every depressive problem, albeit to

205
a variable extent and often with one of them dominating: the introjective (structural)
and the anaclitic (object relational) dimensions of depressive illness. The idea exists
of a dominant depressive style (structural/introjective or relational/anaclitic), but in
no way are these styles exclusive (Blatt, 1998).
With his theory of two kinds of depressive experiences, Blatt gives full weight to
an aspect that in Glazer's theory remained in the background. Specifically, Glazer
points to the possibility of depressive reactions as a result of loss of the approval by
the moral conscience (falling short with respect to norm and ideal in "object-related
depression") as well as resulting from loss of the self-object (the ideal link with it
would be lost by the aggression of separation, which would give rise to "narcissistic
depression"). Blatt will give more weight to the (object) relational components in
what Glazer calls narcissistic depression, as we shall see, without neglecting the
possible influence of the superego aspects in these relational depressions, and also,
by hinting at relational depressive illness in a broader area than that of the
narcissistic level.
As an intermediate consideration, we should also mention here that the use of
both sets of terminology can confuse the reader: Glazer's object-related depression
corresponds mostly to the introjective depression of Blatt' while the narcissistic
depression of Glazer chiefly relates to what Blatt calls (object) relational depression
(although Blatt clearly sees this relational component more broadly than Glazer:
not only dyadic, but just as much triadic)!
Blatt also does justice to an object-relational origin of a pathological superego
structure in depression, while most of the theories of depression in psychoanalysis
stress the ego-psychological origin of depression (the helplessness of the ego),
whether or not in interaction with aggressive impulses (Bibring, 1953).
As summarized in Chapter 3, in contrast to Glazer, Blatt distinguishes two
developmental lines and he indicates that development into normality or pathology
is the result of a dialectic interaction between these two lines. The end result of
normal development along the anaclitic developmental line becomes clear in the
capacity for satisfactory, intimate, interpersonal relationships; the end result along
the introjective line takes the form of a stable, positive feeling of autonomy and
identity. In this latter line, the genesis of the ego is central and the formation of ego
ideal and superego also plays an important role, more specifically through
processes of identification and introjection. Essentially, this second line is
concerned with self-definition, self-evaluation, a sense of self-competence and
identity.

, In Glazer, object-related depression refers to depression within the context of whole object relations and
complex personality structures with a distinct id, ego, superego, while at this level of depression, Blatt
chiefly emphasizes the structural characteristics (the introjection leading to the emergence of the ego ideal
and superego). Glazer's narcissistic depression is then located at the level of partial relationships or self-
object relationships, while Blatt's relational depression additionally concerns the loss of whole object
relationships,
, Blatt's approach covers the whole range of personality functioning, from psychotic to normal-neurotic.
At each level, depression can occur within which both dimensions (introjection and anaclitic relatedness)
can be intermingled. Glazer's approach seems to refer to a more restricted level of pathology: object-related
depression is Oedipal in nature and associated to neurotic illness, narcissistic depression is pre-Oedipal and
related to pre-neurotic character pathology.

206
 Based on this view of development - development as interaction between several
developmental lines - Blatt has also described two clusters of psychopathology
(Blatt & Shichman, 1983; Blatt & Zuroff, 1992): the anaclitic and the introjective.
Influenced by many types of biological, contextual and/or psychological factors,
the subject "chooses" one of the two. This "symptom choice" is, as many before us
have indicated, still largely uncharted territory. It is perhaps more important to
know that Blatt indicates that whichever "developmental line" and "pathology
cluster" is "chosen", the other line can also be entwined with it alongside this
dominant line and cluster.
In anaclitic psychopathology, the focus is on problems with interpersonal
relationships, merger and individuality, dependence and autonomy, whether within
relations of a dyadic or triadic nature. In introjective psychopathology, the focus is
on difficulties with self-experience, self-appraisal, self-definition, self-criticism and
identity. Introjective aspects will also resonate within dominant anaclitic problems,
just as relational aspects will playa role within dominant introjective problems.
Just as we can speak of anaclitic and introjective psychopathology, we can also
refer to anaclitic and introjective depression. Anaclitic depression is related to
conflicts in the interpersonal realm, particularly relating to dependence and
(threatened or real) object loss. Feelings of loneliness, helplessness and weakness
dominate. There is a highly intense longing for love and approval, but just as great
a fear of rejection and loss of contact. Aggression is largely repressed, in order to
protect the longed-for relationship or the necessary dependence. Repressed or
denied aggression not infrequently finds an outlet in psychosomatic symptoms.
Introjective depression is linked rather to problems concerning the sense of self,
self-evaluation and self-appreciation. Feelings of guilt and self-deprecation are
prominent. The criticism targeted at the self is strong: the drive of mastery and
extreme perfectionism do not permit failure in attaining the strict ideals. Aggression
towards the self or the other occasionally comes clearly to the surface, certainly
when ideals are not attained by the self or ideal expectations are not met by the
other.
Blatt and Zuroff (1992) also link gender considerations to both forms of
psychopathology and depression: women present more anaclitic problems, men
tend towards introjective problems. Also, it is possible that in ego-oriented Western
societies, depressions tend to be introjective, while in societies where identity is
determined much more socially, depressions are more likely to be anaclitic or
relational. The appearance or phenotype of a psychopathological syndrome may
also be determined by the culture of the patient. This aspect is referred to in cross-
cultural psychology as the pathoplasticity ofa syndrome (Kleinman & Good, 1985;
Kortmann, 1995). Culture is one element in this plasticity. Markus and Kitayama
(1991) were among the first cross-cultural psychologists to argue that instruments
that are more sensitive to culture-specific idioms of distress and psychopathology
are urgently needed. In this respect, Blatt's Depressive Experiences Questionnaire
(Blatt, 0' Afflitti, & Quinlan, 1976) - with an emphasis on self-critical introjective
and on socially oriented relational depression - could have enormous benefits for
depression research in multicultural societies. For example, not all depression will
have a connotation of a downward movement of the self. This self-evaluative

207
emphasis has been identified by Littlewood (2003) as a typically Western or
European idiom for distress or a typical Western format for depressive illness from
the l S" century on. Kareem (1992) argues that depression in non-Western
immigrants is as often associated with loss or breakdown of social networks as it is
with not living up to the ideals and standards of a "successful migration". The
relational pathway into depression was perhaps not emphasized sufficiently in
depression literature; depression in children, woman and immigrants from non-
Western cultures brings this dimension more into prominence.

Both forms or clusters of psychopathology - relational and introjective - emerge

from an accumulation of factors which may be located at various moments of
development. In this sense, Blatt presents a complex model of developmental
psychopathology. In Glazer we still find the idea that narcissistic depression occurs
on the basis of problems during the period before the rapprochement crisis (pre-
rapprochement phase pathology), while object-related depression occurs during the
rapprochement crisis and/or later. Blatt does not make this distinction in
developmental origin between the two types of depression, or certainly not in the
same way. Although Blatt considers anaclitic depression as more primitive than
introjective depression, he also refers to the idea of progression/regression and
deferred action. In his theory, the subject is presented as vulnerable to anaclitic,
sociotropic elements or to introjective, autonomous aspects. Various psycho-
pathological syndromes are possible, based on this vulnerability on one of the two
developmental lines. Within the framework of two (interacting) developmental
lines, disturbances can be present at various levels: psychotic, borderline, neurotic.
For example, non-paranoid schizophrenia, hysterical borderline personality
disorders, infantile ("narcissistic") personalities and anaclitic depressions are
situated as deviations from the anaclitic developmental line. The introjective
developmental line is concerned with paranoid schizophrenia, borderline persona-
lity disorders with the emphasis on ideational problems, obsessional neurosis
(obsessive-compulsive personalities), phallic-narcissistic problems (with an
exaggerated involvement with the self) and introjective depression (see Blatt &
Shichman, 1983; see also Chapter 3, this volume).
These various syndromes are not only closely linked within each dimension of
psychopathology, they can also alternate through various kinds of regression
and progression through time or occur in a mixed form at a well-defined moment.
At the same time, restrictions are imposed (Blatt, 1995): both forms of
schizophrenia are typical of difficulties which first occur on both developmental
lines, specifically when in development object permanence emerges (libidinal
recognition constancy). Both borderline conditions are then situated within the
perspective of problems with object constancy (evocative object constancy), while
the other syndromes are related to affect constancy (post ambivalence constancy).
Using this division, schizophrenia and borderline conditions are preneurotic
personality disorders - related to processes in pre-Oedipal development - while the
other conditions (including both kinds of depression) are related to neurotic,

208
intrapsychic Oedipal conflicts. Blatt therefore speaks not only of two devel-
opmental lines, but also of several developmental moments at which both devel-
opmental lines can be turned in the direction of various psychopathological
syndromes.
This developmental psychopathological perspective fits in with the complexity
of developmental lines and processes that contribute to a psychopathological
outcome (Blatt & Homann, 1992). Yet, it still leaves questions unanswered. Firstly,
although in theory Blatt's model links both kinds of depression to Oedipal
development, it is still conceivable for introjective depression to be regarded as the
most differentiated form of depression, while anaclitic depression could again be
regarded as less differentiated (not triadic, but dyadic; not Oedipal but pre-Oedipal).
The tendency to involve relational models in psychoanalysis, particularly for a
better understanding of the first three - pre-Oedipal - years of life, while
introjection in the classical sense is approached from an Oedipal perspective,
certainly plays a role here. Object-relational models in psychoanalysis are closely
linked to the discovery of the socio-emotional development in the first years of life,
while superego formation in classical Freudian theory was associated with Oedipal
development. It is hardly surprising then that the relational depressions can be
linked to earlier developmental tracks more easily than the introjective depressions.
Secondly, although the various pathological syndromes linked to each of the two
developmental lines can interact, each syndrome is linked to one developmental
origin: pathology as a result of problems with object permanency, pathology as a
result of problems with object constancy, or pathology as a result of problems with
affect constancy. Without taking the complexity of Blatt's model well enough into
consideration, the reader could inadvertently regard this again as a linear
developmental model in psychoanalysis, a uniformity myth or single track model
of which criticism is possible: "Psychoanalytic developmental models aim at a level
of abstraction where there is a one-to-one relationship between a particular pattern
of abnormality and a particular developmental cause" (Fonagy & Target, 2003,
p. 10).
Thirdly, the various syndromes in both dimensions of psychopathology are
linked to a developmental origin. But what exactly is meant by "origin"? Is the
developmental line already deviating in the direction of depression at that critical
moment in development, and how is this deviation expressed at these moments?
Does vulnerability on a developmental line occur for the first time here? Is
development under pressure, while resilience ensures that the developmental line
still remains within the limits of normal variance? Does this pressure or risk factor
then form a vulnerability which subsequently, in interaction with later develop-
mental interferences, leads to deviance from the normal developmental line? In
other words, how should we conceptualize continuity and discontinuity between
early childhood risk or vulnerability, subsequent developmental interference and
deviant developmental outcome? What interactions are conceivable between
various influencing factors, from several developmental phases, in the emergence
of psychopathology and, more specifically, depression? The phase specificity of the
object tie of the child is a very important element to mention here. Winnicott (1964)
stresses the importance of the father as a shield around the mother-child unit in the

209
oral phase, enabling the mother and child to attune to one another and to make the
first steps toward differentiation and individuation (see also Chapter 6, this
volume). In the anal and Oedipal phases the father has another function for the
child: helping the child with further ego-integration, personality structure building,
gender identity, and psychosexual development. In the process of internalizing
moral standards, the father also plays an important role. We do not want to suggest
here that the mother-child tie is the only relational matrix, and the father-child
relation the only introjective, structure-building matrix. Father and mother play an
important part in both dimensions, but they do this in many different ways
depending on the developmental phase the child is in.
The loss of a parent can also have differential effects on the relational and
introjective dimension, depending on the developmental phase in which the loss is
experienced. Whether a child is more vulnerable on the relational or the introjective
dimension has to do with subjective aspects but perhaps also with the phase-
specificity of the early experience of loss. Further research on the relational and
introjective impact of childhood loss will be of great importance (e.g., Bravo, 2001;
Dietrich & Shabab, 1989; Nagera, 1970).
Compared to the linear developmental models in psychoanalysis, Blatt's
model has greater explanatory value but can still be reinforced in the
aforementioned manner. It is a question of introducing further developmental
differentiation within the concepts of "anaclitic" or "introjective" depression.
Introjective problems within the context of dyadic relationships are not the same as
introjective problems within triadic relationships; both may also possibly make a
different contribution to the onset of depression and have a different influence on
the progress of depression and the therapeutic process. The same applies to
anaclitic problems which also emerge from different developmental phases:
problems of relatedness in the pre-object phase, in the phase of dyadic relationships
or in the phase of triadic relationships. It is no coincidence that Blatt and Spitz lise
the concept of "anaclitic depression" to refer to phenomena from widely varying
developmental phases, Oedipal and pre-Oedipal. Moreover, greater developmental
complexity will allow a more precise developmentally oriented therapeutic
intervention.
It is therefore high time for psychoanalytic developmental psychopathology to
discern how depression can assume different forms depending on whether aspects
from various developmental phases are prominent (and, for example, are expressed
subsequently during the various phases of a psychoanalytic process). Associated
with this is the question of how anaclitic and introjective dimensions of depression
differ in various clinical pictures which are thought to diverge at specific
developmental moments - phase-specific - from normal development (in other
words, what is anaclitic/introjective depression like in schizophrenia, in border-
line conditions and in neurotic conflicts?). It is at this level that a develop-
mental psychopathology perspective within psychoanalysis should be able to
make progress. In the next section, we briefly discuss how a developmental
model that would distinguish between three phases or moments in both the anac-
litic and introjective developmental line might organize current and future
research.

210
Identifying Three Moments on the Anaclitic and Introjective Developmental Line

The term "anaclitic" is - under the influence of Spitz (1945, 1957) -

spontaneously associated with very early pre-Oedipal, relationally determined
depression. Introjective depression as a consequence of a certain perception of the
gap between ideal and reality - is then spontaneously associated with later pre-
Oedipal or Oedipal development.
Helplessness in the first months of life, therefore, is not so much the result of
critical self-evaluation, but of contact problems or deficiencies in the earliest
mother-child relation. The young child feels helpless, for example, because he
cannot attain the necessary symbiosis with the mother: the child lost his mother or
she is emotionally unavailable. This impossibility of making contact with the object
or inability to remain in contact with a significant object forms an early
developmental path to depression. Hermann (1936) and Szondi (1965) described
this early contact dimension based on a double movement of "clinging to and going
on search". Insufficient possibility to do this leads to relational depression, certain
extreme forms of which are described in Spitz (1945, 1946): hospitalism and
anaclitic depression.
Depression here is particularly related to difficulties in the earliest contact
between the child and the outside world. Helplessness is therefore related to the
impossibility of reaching the mother, of forming a symbiotic link with her, finding
a self-object, encountering a safe relational matrix within which initial self-
development can take place and from which the child can subsequently separate:
mothers have to be there to be left (Furman, 1988). Fairbairn (I 944/1952b) would
argue that this is not depression because of aggression in love, but because of the
feeling that one's own early love or binding impulses are not good enough.
Winnicott (1963), in turn, would relate this kind of depression to the fact that the
child's "contributing in (the first relationship) is not met". Tustin (1987) would
refer to depression because of an inability to communicate early-childhood needs
to the other either because that other is not there to be reached or because the child
himself does not possess basal capacities for reaching out to the other. Her view is
also interesting because she indicates that the relational dimension of depression
has various levels: helplessness and depression in contact at the psychotic level, at
the borderline level, at the neurotic level or in normal variation. The normal
counterbalance to this early-childhood helplessness is a feeling of competence and
agency of the child in relation to the other. Interestingly enough, Alvarez (1992)
describes the need for live company and the echoes a deficiency in this company
leads to in several levels of psychopathology.
Blatt indicates, not coincidentally, that we still do not know enough about this
dimension of depression, certainly concerning the early, pre-Oedipal factors.
Recent research in psychoanalysis addresses these early developmental phases and
their importance for understanding depression and other psychopathological
syndromes (e.g., Bravo, 2001; Cramer, 1987; Green, 1983; Kreisler, 1992; Meurs
& Jullian, 2003; Stern, 1985; Vliegen, 2003).

211
 To conclude our discussion of the anaclitic dimension, we should say that this
object-relational dimension is also present in the later developmental phases, for
example, in the fact that moral fear and guilt are not infrequently described in terms
of fear of loss of the object, fear of loss of approval or love of the object, or fear of
loss of approval by the internalized voice of the parental figures. In this sense .- in
the midst of moral reactions in the later pre-Oedipal and Oedipal phase - relational
concerns and fears can occupy center stage.

The term "introjective" is used much more for the later moments in early
childhood development because moral development or even the perception of the
gap between ideal and reality requires some complexity in the structure of
personality. Previously in this paper we have already extensively discussed authors
who concentrated on this developmental line (see also Chapter 3 and 5, this
volume).
A preferred angle of approach in describing this developmental line is the role of
aggression in attachment and in love: the role of ambivalence in the emergence of
introjective depression. As we have seen, two generations of ambivalence conflicts
are described by Parens (1979): on the one hand, the ambivalence which is a normal
part of a complex self-image (with positive and negative aspects), to which one can
react with depressive moods; on the other hand, the ambivalence which is a normal
part of love and hate toward both parents in the Oedipal constellation at a time when
the child is more able to be aware of affective contradiction and the moral conflict
in it, and also able to feel guilt more fully. By distinguishing these moments in
moral development, specific developmental anchor points of introjective problems
are suggested. The earliest revert to developmental phases that are often described
as being relational in nature. In this way, the connection between relational-
anaclitic and introjective-normative developmental lines is made clear,'

"Closed Doors and Landscapes in the Mist": Clinical Illustration of a Relational

Depression with Introjective Elements in a Grade-School Boy

In a short clinical vignette we will illustrate the way introjective and relational
depressive themes are intertwined. Three fragments are extracted from a long-term
, Blatt clearly suggests that introjective depression (situated at the higher neurotic level) is associated with
manifest ambivalence. this in contrast to the obsessional neurotic. in which aggression is clearly
repressed/displaced. This perhaps suggests still another level of dealing with aggression. which is perhaps
also more closely associated with narcissistic personality. the obsessional neurosis occupying a position
between neurotic illness and preneurotic disturbances. thereby requiring a more extreme form of repression
and isolation of aggression and ambivalence. At the same time. for constitutional or other reasons. in
obsessional neurosis the aggression can be more pronounced; therefore a more extreme defense against
aggression and some clear manifestations of strong ambivalence can be present simultaneously.

212
psychoanalytic child psychotherapy with a latency boy. These and other elements
of this therapy have been published previously (see Meurs & Cluckers, 1999). In
this chapter, we have chosen fragments from the therapy report that enable us to
illustrate that a child seems to have a "preference" for one kind of depression but
that the other depressive aspect comes into play after a while, leading to the
situation that both depressive themes - relational and introjective - get connected
in the clinical material.

Jan, a 9-year-old boy, was admitted to the paediatric wing of a local children's
hospital with persistent stomach pain and nausea. An extensive somatic
examination could find no adequate explanation for the symptoms, so Jan and his
parents were referred to the child psychotherapist at the hospital. His father and
stepmother admitted from the outset that things had been difficult at home with Jan
for some time. Jan alternated between a high level of cooperation and sudden,
violent aggression. At other times, the aggression assumed more covert forms. Jan
agreed to everything, but just did not do what his stepmother asked him to do.
Jan's mother had a long history of depressive illness. After Jan's birth, however,
she seemed to be improving; she was very excited about her boy. When Jan was one
year old, however, she seemed to lose interest, a change that became very clear
when Jan was a very active child in his rapprochement crisis. When Jan was 2 years
old, his mother was admitted for long-term psychiatric care, due to depression that
had become more severe. She had also had several psychotic episodes in the
meantime. After a while, the mother broke off all contact with the family, and Jan
had not seen her since he was 4 years old. At that point, Jan's mother was well
aware of her mental problems; she even insisted that "when she goes too far", her
husband should live his own life. Eventually, his father decided to go it alone. They
were divorced when Jan was nearly 5 years old, and Jan continued to live with his
father. The mother accepted the divorce but asked for a legal agreement about the
possibility to have contact with Jan, especially when the boy would request it. She
herself tended to withdraw from her family; on one occasion, when we spoke to her
at the end of treatment, she said that in that period she was convinced that she
would cause harm to her child by being with him in a such painful, depressive state.
When Jan was six years old, his father met his new partner and, when Jan was
nearly 8, his father and his new partner moved in together. About a year later, Jan's
mother unexpectedly renewed contact and asked to be able to visit Jan fairly
regularly. This possibility had been left open at the time of divorce, but his mother
had never taken it up. The occurrence of Jan's stomach complaints roughly
coincided with his mother's request to re-establish contact.
Jan was diagnosed as suffering from childhood depression. He had serious
doubts about himself and was never sure that he was doing well enough in contacts
and tasks. Whenever he did something he thought of as being wrong, he wanted to
make amends, yet at the same time feared it was never good enough. His father said
that Jan frequently lost himself in contact with another: he would give everything
away, just to be sure that the other would not lose interest in him.

213
 The diagnostic examination took place while Jan was hospitalized for two
weeks. After release, father and stepmother agreed to weekly child psychotherapy
and parental guidance once every month. When Jan started visiting his mother, we
also invited her to a parental guidance session. We spoke to her altogether three
times over a period of eighteen months. Jan's psychotherapy lasted for two years.
At first, Jan was quite withdrawn in the play room. He played by himself,
but whenever the therapist asked something about his play, Jan answered with
an elaborate story. The therapist soon felt that the therapeutic alliance had
been established; Jan seemed to be a very sensitive boy. In Session 8 Jan pro-
duced a drawing - "The landscape in the mist" - that was important for several
reasons:

"There is a misty, gray landscape with all sorts of shapes and silhouettes. These
figures (with no face) are lost in a wood and can no longer find one another.
When the therapist asked what struck Jan most about his drawing, he said, 'They
have no faces, they do not know if they can see or hear anything, a signal or
something'".

The drawing evoked a certain vagueness that can occur in affective commu-
nication with a depressed person: Jan saw people as vague silhouettes without the
sensory means to address one another affectively. People are no longer able to find
one another; they even don't have the means to locate one another, to reach out to
one another. Not being able to see, to reach out or to communicate: this is an
expression of Jan's relational depression in the period when his mother had
vanished after having had at first a meaningful and intense relationship; up till now
he could not re-establish contact with her. The same anxieties seemed to playa role
in the therapy room: after the first sessions, Jan became fearful about losing contact
with his therapist. He was desperate about how to continue contact once it had been
established. In the next sessions, he wanders around in the play room, not knowing
what to do, repeating several times his question if he would be allowed to return to
his therapist the next week. For one reason or another, he fears losing contact
without having any control in this process. Grey colors in everything he draws
accentuate this depressive impression. It is the contact dimension - the relational
theme - that is at stake: joining the other, touching the other, approaching the other,
keeping in touch.

In Session 20, after a period of despair for Jan - his first contact with his mother
had been planned but also delayed; in the therapy room he was very difficult to
reach - the therapist decided to intervene in the following way:
Th.: 'Jan, you have already led me to understand that people who want to be in
contact with one another sometimes find it difficult. For some time now, I have

214
 felt that you were happy to come here. But you also have your doubts about how
things will be between us from now on.'
Jan then produced another drawing: 'A house in the distance'. A child wants to
visit the house. The inhabitants have left. The door is closed. The child wanted
to go inside. Suddenly, Jan corrected this and said that the boy would certainly
not try to go inside but would sit and wait at the door until somebody comes
home. Nobody knew if and when they would return.
Th.: 'The boy knows that he wants to find someone there, but he feels that he is
never certain whether and when somebody will be there for him.'
Jan nods in agreement. 'In any case, he will wait, then they cannot forget him
when they come home.'
Th.: 'The boy thinks the only thing he can do is wait. He has the impression that
everything else depends on the others. In any case, he makes sure that the others
cannot get past him because he is waiting at the door.'
Jan: 'It might take a long time. They will probably not like him lying against the
house. They would probably not let him in. The boy does not know how to do
any good. Then he falls asleep.'

After having described a world where people can fade away (the relational
depressive theme), Jan moves closer now to his own desire to make contact ("going
to a house" and "wanting to go inside"). But this first longing for contact is con-
nected with a sense of doing wrong. Jan's sudden correction could be understood
as a defensive act (a kind of reaction formation or undoing), afterwards verbalized
by saying "the boy does not know how to do any good". Anyhow, the door is closed:
the residents have gone or they do not let the boy in. In reaction to this, the boy has
a sense of acting in the wrong way and of being left behind. A first introjective (self-
critical) theme comes into play: Is it my fault? Is the loss caused by my actions?

This combination of longing for relationships and fearing to drive away the
longed-for other, now became much more important. In Session 41 Jan plays on
"closed doors".
He pretends to be a boy waiting for a message from a distant world where the
"earth mothers" live, deep underground. The boy sends various sound signals to this
unknown world. This distant world and the world of the ordinary people in which
the boy lives are separated by a wall. The world of the "earth mothers" is populated
by motherly old women.
The sounds coming back from the world of the "earth mothers" come too quickly
or become merged and confused in the very small opening in the wall through
which the messages and sounds have to pass to reach the boy. This makes the
message incomprehensible. Somewhere, there is a magic key that opens the closed
door in the wall between the two worlds just enough to make the sounds and signals
from the "earth mothers" comprehensible. The boy has the key. However, once he

215
receives a signal from the "earth mothers", he cannot wait. He sets to work too
quickly, so that the messages again become incomprehensible. The boy briefly
notices an old man who knows how to use the key to understand the message better.
But the boy wants to use the key to leave the wall behind and to go into or even
merge with the world of the "earth mothers" as quickly as possible. The boy
hesitates a little regarding the old man, but the man is eventually rejected with the
words, "Do not take my key now".
Jan's desire for contact with a distant and sometimes incomprehensible mother
becomes very clear now. After a period of being quite uncertain as to whether he
was allowed to undertake any action to establish contact, he is now very much in
contact with his relational desires. He is even very active at it, still in search of a
good balance between activity and passivity. The symbolic play material of this
session has an enormous complexity: the relational maternal theme - with a desire
for merger - is connected to a paternal theme, with some fears of exclusion and/or
castration. Fears of being overwhelmed by a "third" figure (the old man), of being
unable to attune to messages from the earth world (the other world of the mothers)
and of having been denied by the father figure the means (the key) to re-establish
contact with the mother world, are different layers of this material. Specially with
this last element, Jan moves closer to an introjective depressive fear: the boy can be
punished for doing the wrong thing by acting upon his wish for contact (the
punishment or exclusion is expressed in the following play themes: not being
allowed to enter the house in Session 20, or, by being "castrated" for his desires and
activities in Session 41).
The background still is Jan's fear of not getting connected (the relational
depressive theme), but once he is in touch with his wish for contact, these relational
themes and the associated activities become associated with introjective depressive
themes ("due to what I am undertaking, I will lose contact or not be allowed to have
contact or be punished"; "due to my activity, I will be judged as being wrong").
From now on, interpretations can focus on this complex weave of both core themes
in depression: fears of relational loss, associated with one's activity (in the light of
motivations, drives, desires, wishes) that is criticized by the superego.

Future Challenge: Formulating Complex Developmental Pathways and

Reintegrating Psychoanalysis Into Mainstream Developmental Psychopathology

The widespread single-track model in psychoanalytic developmental

psychopathology suggests that neurotic psychopathology can be regarded
exclusively as a remnant of Oedipal concerns and conflicts (between the ages of 3
and 6). Personality disorders are then a remnant of deficient personality structuring
in the second and third years of life, while psychoses are thought to go back to very
earliest childhood:

216
 "There is a tacit assumption of an isomorphism between pathology and
development, which permits bi-directional causal inference between childhood
and pathology. The assumption covers all psychopathology and all stages of
development. (... ) There is a single model for borderline personality disorder,
narcissistic pathology, neurotic disorders, etc. Empirical studies, on the whole,
are at odds with these accounts. (... ) Developmental psychopathology research
has demonstrated that developmental continuity is an empirically elusive and
conceptualIy complex problem, and cannot be simply assumed, as psycho-
analysts are wont to do" (Fonagy & Target, 2003, p. 5-10).

While the map of the first three years of life gradualIy has been drawn in much
detail, the possibilities for clarifying the developmental origin of psychopatho-
logical disorders increased. This provided enormous opportunities (better under-
standing of pathological functioning based on some comparison with early develop-
ment, detection of early risk factors in development) as welI as disadvantages
(reducing psychopathology to this early phase or deterministically thinking that
difficult developmental phases would always lead to a certain type of pathology).
Among the advantages of psychoanalytic developmental psychopathology one
could point to the thoughtful way of connecting psychopathological syndromes
with crucial developmental phases. This has led to an enormous body of knowledge
on the socio-emotional psychodynamics of the different developmental steps and
complexes of childhood. These are understood and reconstructed from within the
conflicts that are considered as reminiscences, later on in life, of these early
childhood dynamics and complexes; at other moments, psychoanalysis hinted at
domains that could be the object of childhood observation studies. Without psycho-
analysis, Cicchetti & Cohen (1995) admit, these important normal developments
(for example, the Oedipus complex, the separation-individuation process, the
narcissistic and object-oriented phases, etc.) would not have been documented the
way they are today. In these crucial processes the origins are found for vulnerabilit-
ies that later in life lead to psychopathology: in these crucial childhood psycho-
dynamics the crystal (personality) (Freud, 1932) receives its structure, with its
strong aspects and with certain hidden ruptures. When later in life a stressor directly
acts upon the hidden vulnerability, the crystal can break.
Another advantage is that in psychoanalysis a lot of knowledge is acquired about
the pre-school phase; in mainstream developmental psychopathology this period of
precursors of disturbances is the least welI documented phase. Not surprisingly, in
autism research, for example, psychoanalysis has a lot to say about the risks and
vulnerabilities of this syndrome; in the same way psychoanalysis emphasizes the
importance of problems with object constancy in children or adults with borderline
characteristics and anti-social behavior or of problems with affect constancy in
children with emotional disturbances or depression. Even in what we have been
indicating as a low profile period for psychoanalytic developmental psychopath-
ology (1975-1990), the developmental perspective within psychoanalysis became
more important. Recently, under the influence of a wider acceptance of quantitative

217
methodology in psychoanalysis and motivated by the success of mainstream
developmental psychopathology, the developmental and the clinical model in
psychoanalysis were brought in much stronger interaction again (Shapiro & Emde,
1995). This, in turn, has led to a revival of the developmental psychopathology
project within psychoanalysis; in this revival a more complex model of
developmental continuity is presented.'
We are in need of that complex model, since there are disadvantages to the
psychoanalytic developmental model as it has been described above. The broader
flow of developmental psychopathology represents a huge challenge to the
dominant developmental model in psychoanalysis. The interaction of influences
from various developmental phases is a complex matter. The same vulnerability in
a certain developmental phase can produce widely varying forms of developmental
trajectories (multifinality), while the same outcome can be achieved from widely
varying lines of development (equifinality) (Von Bertalanffy, 1968; Cicchetti &
Cohen, 1995; Cicchetti & Rogosch, 1996). These aspects are definitely exerting an
influence on psychoanalysis, as is evident from the amount of research into
pathways along which the internal representations of early experiences with care
figures have an impact on the formation of subsequent relationships later in life
(Emde, 1988; Fivaz-Depeursinge, 2004).
However, although authors like Shapiro and Emde (1991) have been working out
this complexity in their developmental theories, we still regard this as a further
challenge for future psychoanalytic developmental psychopathology. For the time
being, insufficient clarity exists in psychoanalysis about the respective influences in
the emergence of psychopathology of the three developmental moments in early
childhood (distal antecedents) or of early childhood and subsequent school-age and
adolescent development or adulthood (proximal antecedents) (Blatt, 2004). This
complexity is recognized on a theoretical level, but has been seriously under-
researched; we are still starting to work out a "complex-enough" developmental
perspective on psychopathology that will exert a decisive influence on our clinical
practice.
Mainstream developmental psychopathology teaches us that normal and
pathological developmental lines emerge through a series of associated factors and
mechanisms throughout all the phases of child development and life-span.
Moreover, a more complex picture also allows us to predict that various clinical
pictures will also have many shared aspects (for example, anaclitic or introjective
features in depression and other psychopathological syndromes), while similar
clinical pictures also have striking differences (for example, not all anaclitic or
introjective psychopathology is equal).
If we regard these questions as future challenges for psychoanalysis, the
specificity of its theories will increase. Crucial questions include, "What are
examples of earlier and later developmental factors which contribute to

o American psychoanalysis, intluenced by Hartmann's original empirical project, has probably been leading
in this effort: it has always been closer to developmental psychology, the search for evidential basis for its
theories and the reflection on complex developmental mechanisms such as differentiation, integration.
continuity and transformation (see Shapiro & Emde. \991) than most of European psychoanalytic thought.

218
depression?", "What is earlier and later anaclitic/introjective vulnerability and
resilience?".
The challenge consists of sufficiently complicating the psychodynamic develop-
mental perspective so that, within the broader flow of developmental psychopatho-
logy, it has its place. Psychoanalysis has always had a developmental perspective.
Since a developmental perspective is not obvious in (child) psychiatry or clinical
(child) psychology', or may even be lacking there, psychoanalysis has a huge
advantage in this respect. After all, even in the broader flow of developmental
psychopathology, a tendency remains to include "age" as an "empty variable" in
research or to look at "development" as the evolution of incidence of psycho-
pathology at different ages ("developmental epidemiology") or to limit it to
cognitive micro processes (the development or the process of a pathological,
depressive thought or of cognitive information processing). In response to these
limitations, psychoanalysis possesses certain advantages: no other school of
thought emphasizes to such an extent that developmental phases are well-rounded
elements from a content point of view, with their own distinct processes, specific
psychodynamics and conflicts. However, in order not to be drowned out in the
broad field of psychiatry and psychology, the psychoanalytic researcher must also
allow himself to draw inspiration from developmental psychopathology.
For depression research, this means that earliest development does not determine
subsequent development; pathological functioning is not the same as early
childhood functioning. The final clinical picture is the result of a pathway through
many developmental phases and influenced by complex interactions of
psychological, biological and contextual factors. Early developmental achieve-
ments are integrated in further developmental steps and phases; they support later
developmental achievement, but do not guarantee them. Early developmental
stressors reduce the probability of further adaptive development, but they do not
rule it out altogether. This probabilistic, rather then deterministic, developmental
reasoning shows that the ultimate outcome is the result of many developmental
lines and several crucial influences at different moments on these developmental
lines. Research should make this complexity easier to grasp; investigating the
crucial determinants of adaptive and maladaptive development as well as their
relative predictive power over the outcome (adaptive or maladaptive development)
requires a complex model and well chosen statistical analyses within a quantitative
methodological approach.
Clinically oriented psychoanalysts and their "developmental research-minded"
colleagues will also be able to interact better: what Freud described as "deferred
action" is after all a clinical way of thinking about how a certain outcome, whether
pathological or not, comes about through a complex developmental pathway in
which earlier "dynamic" impressions and experiences are continuing to work in

7 "The impact of development on clinical dysfunction and how this information can be integrated into
treatment may have important implications for conclusions about the effectiveness of treatment. Outcome
research rarely considers features of development that may affect treatment. (... ) Chronological age
represents an empirical beginning. However, age, as a construct for research, is only seldom identified in
specific processes ( ... ) (e.g. attachment; moral, cognitive and socio-emotional development, etc.)"
(Kazdin, 2004, p. 12)

219
later ones, attracted there by new similar experiences, reworked in the present
and/or influencing the significance of present experience. The clinical concept of
"deferred action" is much closer to a model of complex developmental pathways
than to single track developmental models, as is illustrated by the following quote
from three leading authors in the field of developmental psychopathology:

"It is also an error to assume that pathways between early and later development
are relatively simple and straightforward (e.g., early depression predicts later
depression). In fact, the wayan individual responds to events is a function of past
history or experience, as well as current experience. (...) Moreover, during
development, an individual may move back and forth between pathways of
adjustment and maladjustment, or show different constellations of problems at
different ages. The assumptions of a pathway model provide a launching point
for considering fundamental issues of continuity and discontinuity, (... ) and the
need to consider the stage or period of development" (Cummings, Davies &
Campbell, 2000, p. 18).

The expression of depression changes in childhood (see also Chapter 6, this

volume). Early depression can lead to many different outcomes (from normal
development to serious psychopathology), while late depression can arise at various
times, both from previous normal development and psychopathology.

Conclusions

Certainly, there is work at hand for psychoanalytic developmental psychopatho-

logy. This observation, however, should not be a cause for pessimism. The blueprint
we described contains a rich clinical realm and a sound developmental perspective.
Usually, the two are not found together. However, in order to grow, psychoanalysis
must be able to see over the wall of the psychoanalytic model and to answer
questions posed from interaction with the broader flow of developmental
psychopathology. Its integration, together with the use of various research methods,
offers psychoanalysis new opportunities as well as the hope of a stronger position
in interaction with other models.
In the meantime, we have reached the point where we have filled in important
gaps in the relational-anaclitic and structural-introjective developmental lines of
depression. Emde and colleagues (e.g., Emde, Polak, & Spitz, 1965; Emde,
Johnson, & Easterbrooks, 1988) and Blatt (2004) - among others - are important
authors presenting major models in psychoanalysis concerning depression,

220
integrating drive, ego, object relational and self psychologies in a developmental
perspective.
With the formulation of the separate anaclitic dimension of depression and the
observation that pre-Oedipal and Oedipal elements as well as childhood and later
developmental stages are important in both dimensions - the anaclitic and the
introjective - the question of continuity and discontinuity throughout the entire
course of development is clearly posed. Biological, psychological and contextual
factors are sought which playa role in these different phases. At this level, progress
is expected in the near future, both from carefully chosen case studies and from
empirical research.

References

Abraham, K. (1953a). Contribution to the theory of the anal character. In K. Abraham (Ed.), Selected
papers on psychoanalysis (pp. 370-392). New York: Basic Books. (Original work published 1921)
Abraham, K. (I 953b). A short study of the development of the libido. In K. Abraham (Ed.), Selected papers
on psychoanalysis (pp. 418-501). New York: Basic Books. (Original work published 1924)
Achenbach, T. (1974). Developmental psychopathology. New York: John Wiley & Sons.
Achenbach, T. (1991). Manual for the Child Behavior Checklist/4-/8 and /99/ Profile. Burlington:
University of Vermont, Department of Psychiatry Publications.
Alvarez, A. (\992). Live company. London: Kamac Books.
Antonovsky, A. M. (1991). Idealization and the holding of ideals. Contemporary Psychoanalysis, 27.
389-404.
Assoun, P.-L. (1981). L'archaique chez Freud [The concept of 'the archaic' in Freud's work]. Nouvelle
Revue de Psychanalyse francaise, 26. 11-44.
Barron, J. M., Eagle, M. N., & Wolitsky, D. L. (1992). The interface of psychoanalysis and psychology.
Washington, DC: American Psychological Association.
Bibring, E. (1953). The mechanism of depression. In P. Greenacre (Ed.), Affective disorders.
Psychoanalytic contribution to their study (pp. 13-48). New York: International Universities Press.
Bion, A. (1962). Learning from experience. London: Heinemann.
Blatt, S. J. (1974). Levels of object representation in anaclitic and introjective depression. The
Psychoanalytic Study of the Child. 29. 107-157.
Blatt, S. J. (\995). Representational structures in psychopathology. In D. Cicchetti & S. L. Toth (Eds.),
Emotion. cognition. and representation (Rochester symposium on developmental psychopathology
(Vol. 6, pp. 1-33). New York: University of Rochester Press.
Blatt, S. J. (\ 998). Contributions of psychoanalysis to the understanding and treatment of depression.
Journal of the American Psychoanalytical Association. 46, 722-752.
Blatt, S. J. (2004). Experiences ofdepression. Theoretical. clinical. and research perspectives. Washington,
DC: American Psychological Association.
Blatt, S. 1., D' Afflitti, 1. P., & Quinlan, D. M. (1976). Experiences of depression in normal young adults.
Journal ofAbnormal Psychology. 85. 383-389.
Blatt, S. J., & Homann, E. (1992). Parent-child interaction in the etiology of dependent and self-critical
depression. Clinical Psychology Review. /2.47-91.
Blatt, S. J., & Shichman, S. (1983). Two primary configurations of psychopathology. Psychoanalysis and
Comtemporary Thought. 6. 187-254.
Blatt, S. J., & Zuroff, D. C. (1992). Interpersonal relatedness and self-definition: Two prototypes for
depression. Clinical Psychology Review. /2. 527-562.
Bowlby, J. (1981). Attachment and Loss. London: Hogarth Press.
Bravo, I. (200 I). The impact of early loss on depression: Dynamic origins and empirical findings. In 1. R.

221
 Brandell (Ed.), Psychoanalytic approaches to the treatment of children and adolescents (pp. 47-69).
New York: The Haworth Press.
Burlingham, D., & Freud, A. (1944). Infants without families. The case for and against residential nurses.
London: George Allen & Unwin Ltd.
Cicchetti, D. (1984). The emergence of developmental psychopathology. Child Development, 55, 1-7.
Cicchetti, D. (1989). Developmental psychopathology: Some thoughts on its evolution. Development and
Psychopathology, I, 1-4.
Cicchetti, D., & Cohen, DJ. (1995). Developmental psychopathology. Vol. I: Theory and methods. New
York: John Wiley & Sons.
Cicchetti, D., & Rogosch, E A. (\996). Equifinality and multifinality in developmental psychopathoogy.
Development and Psychopathology, 8, 597-600.
Cicchetti, D., & Toth, S. L. (1992). Developmental perspectives on depression. Rochester: University of
Rochester Press.
Cramer, B. (1987). Objective and subjective aspects of parent-infant relations: An attempt at correlation
between infant studies and clinical work. In J. Osofsky (Ed.), Handbook of infant development (pp.
1037-1057). New York: John Wiley & Sons.
Cummings, E. M., Davies, P. T, & Campbell, S. B. (2000). Developmental psychopathology andfamily
process. Theory, research. and clinical implications. New York: Guilford Press.
Dietrich, D. R., & Shabad, P. C. (1989). The problem of loss and mourning. Psychoanalytic perspectives.
Madison, CT: International Universities Press.
Emde, R. N. (1981). Changing models of infancy and the nature of early development: Remodelling the
foundation. Journal of the American Psychoanalytic Association, 29, 179-219.
Emde, R N. (1988). Development terminable and interminable. II: Recent psychoanalytic theory and
therapeutic considerations. International Journal of Psychoanalysis, 69, 283-286.
Emde, R. N. (1999). Moving ahead: Integrating influences of affective processes for development and
psychopathology. International Journal of Psychoanalysis, 80, 317-339.
Emde, R. N., Johnson, W. E, & Easterbrooks, M. A. (\988). The do's and don'ts of early rroral
development: Psychoanalytic tradition and current research. In J. Kagan & S. Lamb (Eds.), The
emergence of morality (pp. 245-277). Chicago: University of Chicago Press.
Emde, R N., Polak, P. R, & Spitz, R. A. (\965). Anaclitic depression in an infant raised in an institution.
Journal of the American Academy of Child Psychiatry, 4, 545-553.
Fairbairn, R (1952a). A revised psychopathology of the psychoses and psychoneurosis. In R. Fairbairn,
Psychoanalytic studies of personality (pp, 28-38). London: Tavistock. (Original work published 1(41)
Fairbairn, R (1952b). Endopsychic structure considered in terms of object relationships. In R Fairbairn,
Psychoanalytic studies ofpersonality (pp. 82-136). London: Tavistock. (Original work published 1(44)
Fivaz-Depeursinge, E. (in press). Attachment et intersubjectivite [Attachment and intersubjectivity].
Lausanne: Unil.
Fonagy, P. (\ 995a). Psychoanalytic perspectives on developmental psychopathology. In D. Cicchetti & D.
J. Cohen (Eds.), Developmental psychopathology (pp. 290-306). New York: John Wiley & Sons.
Fonagy, P. (1995b). Psychoanalytic and empirical approaches to developmental psychopathology: An
object-relations perspective. In T Shapiro & R. N. Emde (Eds.), Research in psychoanalysis:
Processes, development, outcome (pp. 245-260). Madison, CT: International Universities Press.
Fonagy, P., & Target, M. (1997). Psychodynamic developmental theory for children: A contemporary
application of child psychoanalysis. In E. D. Hibbs & P. S. Jensen (Eds.), Psychosocial treatment
research with children and adolescents (pp. 53-77). Washington, DC: National Institute of Health and
the American Psychological Association.
Fonagy, P., & Target, M. (2003). Psychoanalytic theories: Perspectives from developmental
psychopathology. London/Philadelphia: Whurr Publications.
Freud, A. (1963). The concept of developmental lines. The Psychoanalytic Study ofthe Child, 18, 245-265.
Freud, A. (\966). Normality and pathology in childhood: Assessment of development. The writings of
Anna Freud (Vol. 6, pp. 1-186). Harmondsworth: Penguin Books.
Freud, A. (1974). A psychoanalytic view of developmental psychopathology. The writings ofAnna Freud
(Vol. 8, pp. 57-74). Harmondsworth: Penguin Books.
Freud, A. (\981). The concept of developmental lines: Their diagnostic significance. The Psychoanalytic
Study of the Child, 36, 129-136.
Freud, S. (1900). The interpretation of dreams. In J. Strachey (Ed. and Trans.), The standard edition ofthe
complete psychological works of Sigmund Freud (Vol. 4 and 5, pp. 1-715). London: Hogarth Press

222
Freud, S. (1905). Fragment of an analysis of a case of hysteria. In 1. Strachey (Ed. and Trans.), The
standard edition of the complete psychological works of Sigmund Freud (Vol. 7, pp. 7-122). London:
Hogarth Press.
Freud, S. (1909a). Analysis of a phobia of a five-year-old boy. In 1. Strachey (Ed. and Trans.), The standard
edition of the complete psychological works of Sigmund Freud (Vol. 10, pp. 1-147). London: Hogarth
Press.
Freud, S. (l909b). Notes upon a case of obsessional neurosis. In J. Strachey (Ed. and Trans.), The standard
edition ofthe complete psychological works ofSigmund Freud (Vol. 10, pp. 153-320). London: Hogarth
Press.
Freud. S. (1911). Psycho-analytic notes upon an autobiographical account of a case of paranoia (Dementia
Paranoides). In J. Strachey (Ed. and Trans.), The standard edition of the complete psychological works
of Sigmund Freud (Vol. 12, pp. 3-82). London: Hogarth Press.
Freud, S. (l917a). Mourning and melancholia. In J. Strachey (Ed. and Trans.), The standard edition of the
complete psychological works of Sigmund Freud (Vol. 14, pp. 237-258). London: Hogarth Press.
Freud. S. (l917b). Introductory lectures on psycho-analysis. In 1. Strachey (Ed. and Trans.), The standard
edition of the complete psychological works of Sigmund Freud (Vol. 15, 16. 17). London: Hogarth
Press.
Freud, S. (I 920a). Beyond the pleasure principle. In J. Strachey (Ed. and Trans.), The standard edition of
the complete psychological works (if Sigmund Freud (Vol. 18, pp. 1-64). London: Hogarth Press.
Freud, S. (1922). Some neurotic mechanisms in jealousy, paranoia and homosexuality. In J. Strachey (Ed.
and Trans.), The standard edition of the complete psychological works of Sigmund Freud (Vol. 18, pp.
225-230). London: Hogarth Press.
Freud, S. (1923). The ego and the id. In 1. Strachey (Ed. and Trans.), The standard edition of the complete
psychological works of Sigmund Freud (Vol. 19, pp. I-59). London: Hogarth Press.
Freud, S. (1931). Female sexuality. In J. Strachey (Ed. and Trans.), The standard edition (if the complete
psychological works of Sigmund Freud (Vol. 21, pp. 221-246). London: Hogarth Press.
Freud, S. (\ 932). New introductory lectures on psychoanalysis. In 1. Strachey (Ed. and Trans.), The
standard edition of the complete psychological works of Sigmund Freud (Vol. 22, pp. 1-182). London:
Hogarth Press.
Furman, E. (1988). Mothers have to be there to be left. The Psychoanalytic Study of the Child, 43,
345-367.
Glazer, M. W. (\979). Object-related vs. narcissistic depression. A theoretical and clinical study. The
Psychoanalytic Review, 66, 323-337.
Green, A. (1983). The dead mother. In A. Green (Ed.), On private madness (pp. 48-58). London: Karnac
Books.
Hartmann, H. (1939). Ego-psychology and the problem ofadaptation. New York: International Universities
Press.
Hermann. I. (1936). Sich anklammern - Auf Suche gehen [Clinging and searching]. lntemationale
Zeitschrift fur Psychoanalyse, 22, 45-58.
Jacobson, E. (1953). The affects and their pleasure/unpleasure qualities in relation to the psychic discharge
processes. In R. M. Loewenstein (Ed.), Drives, affects and behavior (pp. 38-66). New York:
International Universities Press.
Jacobson, E. (1964). The self and the object world. New York: International Universities Press.
Jacobson, E. (1971). Depression. Comparative studies of neurotic and psychotic conditions. New York:
International Universities Press.
Kareem, J. (\992). Ideas and experience in intercultural therapy: The Nafsiyat Intercultural Therapy
Centre. In J. Kareem & R. Littlewood (Eds.), Intercultural therapy: Themes, interpretations and
practice (pp. 14-37). Oxford: Blackwell.
Kazdin, A. E. (2004). Psychotherapy for children and adolescents. In M. J. Lambert (Ed.), Bergin and
Garfield's handbook of psychotherapy and behavior change (5th ed.) (pp, 540-597). New York: John
Wiley & Sons.
Kernberg, O. (1984). Object relations theory and clinical psychoanalysis. New York: Jason Aronson.
Klein, M. (1928). Early stages of the Oedipus complex. In M. Klein (Ed.), The writings of Melanie Klein
(Volume I): Love, guilt and reparation and other works 1921-1945 (pp. 186-198). London: Hogarth
Press.
Klein, M. (1935). A contribution to the psychogenesis of manic-depressive states. In M. Klein (Ed.), The
writings of Melanie Klein (Volume I): Love, guilt and reparation and other works 1921-1945 (pp. 262-

223
 289). London: Hogarth Press.
Klein, M. (1940). Mourning and its relation to manic-depressive states. In M. Klein (Ed.), The writings of
Melanie Klein (Volume I): Love. guilt and reparation and other works 1921-1945 (pp. 344-369).
London: Hogarth Press.
Klein, M. (1945). The Oedipus complex in the light of early anxieties. In M. Klein (Ed.), The writings of
Melanie Klein (Volume I): Love. guilt and reparation and other works 1921-1945 (pp. 370-419).
London: Hogarth Press.
Kleinman, A., & Good, B. (1985). Culture and depression. Berkeley: University of California Press.
Kortmann, E A. M. (1995). Psychopathologie, cultuur en omgeving [Psychopathology, culture, and
environment]. Tijdschrift voor Psychiatrie, 37. 3-14.
Kreisler, R. (1992). Le nouvel enfant du desordre psychosomatique [The new infant with psychosomatic
symptoms]. Toulouse: Privat.
Littlewood, R. (2003). Pathologies of the West. An anthropology of mental illness in Northern America and
Europe. London: Continuum.
Mahler, M. (1972). On the first three sub-phases of the separation-individuation process. International
Journal of Psycho-Analysis. 55. 333-338.
Mahler, M., Pine, E, & Bergman, A. (1975). The psychological birth of the human infant. Symbiosis and
individuation. New York: Basic Books.
Markus, H. R., & Kitayama, S. (1991). Culture and the self: Implications for cognition, emotion, and
motivation. Psychological Review, 98. 224-253.
Meurs, P., & Cluckers, G. (1999). Psychosomatic symptoms, embodiment and affect. Weaving threads to
the affectively experienced body in therapy with a neurotic and a borderline child. Journal of Child
Psychotherapy. 25. 71-91.
Meurs, P., & Jullian, G. (2003). Cultuur sensitieve ontwikkelingsbegeleiding. Preventief werken aan
veerkracht in de vroege ouder-kind interactie bij allochtone gezinnen in een kansarme wijk van Brussel
[Culture sensitive developmental guidance. Preventive intervention and promoting resilience in early
parent-child interaction in immigrant families in a disadvantaged urban quarter of Brussels], Medische
Antropologie. 15. 221-236.
Nagera, H. (1970). Children's reactions to the death of important objects: A developmental approach. The
Psychoanalytic Study of the Child. 25. 360-400.
Nagera, H. (1981). The developmental approach to childhood psychopathology. New York: Jason Aronson.
Parens, A. (1979). The development of aggression in early childhood. New York: Jason Aronson.
Parens, A. (1991). A view of the development of hostility in early life. In T. Shapiro & R. N. Emde (Eds.),
Affect: Psychoanalytic perspectives (pp. 75-108). Madison. CT: International Universities Press.
Ricoeur, P. (1971). Discours, communicatibilite et ecriture [Discours, communication, and writing].
Nijmegen: s.n,
Shapiro, T., & Emde, R. N. (1991). Developmental perspectives: Introduction. In T. Shapiro & R. N. Emde
(Eds.), Affect: Psychoanalytic perspectives (pp. 3-5). Madison. CT: International Universities Press.
Shapiro, T., & Emde, R. N. (1995). Research in psychoanalysis: Process. development. and outcome.
Madison, CT: International Universities Press.
Spitz, R. (1945). Hospitalism: An inquiry into the genesis of psychiatric conditions in early childhood. The
Psychoanalytic Study of the Child. 1.53-73.
Spitz, R. (1946). Anaclitic depression: An inquiry into the genesis of psychiatric conditions in early
childhood. The Psychoanalytic Study of the Child. 2. 313-342.
Spitz, R. (1957). No and yes: On the genesis of human communication. New York: International
Universities Press.
Spitz, R., & Cobliner G. (1965). The first year of life. New York: International Universities Press.
Sroufe, L. A., & Rutter, M. (1984). The domain of developmental psychopathology. Child development.
55. 17-29.
Stern, D. (1985). The interpersonal world of the infant. A view from psychoanalysis and developmental
psychology. London: Karnac Books.
Szondi, L. (1965). Schicksahlsanalyse: Wahl in Liebe. Freundschaft, Beruf, Krankheit und Tod [Analysis
of destiny. Choice in love relations, friendship, profession, illness and death]. Basel: Schwabe.
Tustin, E (1987). The rhythm of safety. Winnicott Series. 2.
Vliegen, N. (2003). Affectieve communicatie en depressie in de moeder-kind dyade aan het einde van het
eerste halfjaar [Affective communication and depression in the mother-child dyad at the end of the first
semester]. Unpublished manuscript, Department of Psychology, University of Leuven, Belgium.

224
von Bertalanffy, L. (1968). General systems theory. New York: Braziller.
Winnicott, D. W. (1958). Primitive emotional development. In D. W. Winnicott (Ed.), Collected papers:
Through paediatrics to psychoanalysis (pp. 262-277). New York: Basic Books. (Original work
published 1945)
Winnicott, D. W. (1963). The development of the capacity for concern. In D. W. Winnicotl (Ed.), The
maturational processes and the facilitating environment: Studies ill the theory of emotional
development (pp, 73-82). New York: International Universities Press.
Winnicotl, D. W. (1964). What about father'? In D. W. Winnicott (Ed.), The child, the family. and the outside
world (pp, 113-118). London: Penguin.
Winnicott, D. W. (1984). Aggression, guilt, and reparation. In C. Winnicott, R. Shepherd & M. Davis
(Eds.), Deprivation and delinquency (pp. 136-144). London: Tavistock. (Original work published 1960)
Yankelovitch, D., & Barrett, W. (1970). Ego and instinct: The psychoanalytic view of human nature -
revised. New York: Random House.
Zetzel, E. (1965). Depression and the incapacity to bear it. In M. Schur (Ed.), Drives. affects and behavior
(Vol. 2, pp. 243-274). New York: International Universities Press.

225
 Chapter 8

Corticotropin-Releasing Factor (CRF)

and Major Depression: Towards an Integration

of Psychology and Neurobiology in Depression Research

Stephan J. Claes and Charles B. Nemeroff

Major depressive disorder (MDD), as defined by the Diagnostic and Statistical

Manual of Mental Disorders, 41h edition (DSM-IV; APA, 1994), is a very common
disorder as evidenced by the results of two large epidemiological studies, in which
more than 10% and 16.2% of the population respectively were found to experience
at least I MDD episode during their lifetime (Blazer et al., 1994; Kessler et al.,
2003). The prevalence is higher in women compared to men, with a ratio
approaching 2 to I. It is a chronic disorder, with a high tendency for relapse in those
who have recovered from a depressive episode (Frank et al., 1990). MDD is
characterized by a variety of symptoms from different categories. Emotional
dysregulation (low mood, instable mood, irritability, anxiety, emotional apathy) and
cognitive dysfunction (concentration, memory) are accompanied by severe
disturbances in fundamental biological systems such as sleep, appetite regulation,
sexual drive and autonomic regulation. The disorder is also associated with an
increased risk for a number of somatic disorders, such as osteoporosis and several
cardiovascular diseases. This results in markedly increased morbidity and mortality
in individuals with major depression, even if factors like smoking, diet and alcohol
intake are corrected for (Vaillant, 1998). The Baltimore portion of the national
Epidemiological Catchment Area (ECA) study revealed that major depression
subsequently leads to a four- to fivefold increase in the risk for myocardial
infarction (Pratt et al., 1996).
Psychosocial risk factors for developing major depression have been studied
extensively. Life stress has been found to be an important risk factor, as well as the
personality structure of the person undergoing the stressful experiences (see also
Chapter 4, this volume). For this purpose, stress can be subdivided into three
categories: acute stress, chronic stress and early life stress.
Acute stress, more specifically life events with a connotation of loss or
humiliation, has been posited to be depressogenic. Kendler et al. (1999) reported
that individuals reporting a major stressful life event in the last month had a large
increase in the risk for the onset of major depression (Odds Ratio 5.64). However,
the impact of life events on depression vulnerability is highly dependent on other
factors such as gender and neuroticism (Kendler et aI., 2004). At least equally
important in the etiology of major depression is chronic stress (Tennant, 2002). ln
mice, chronic mild stress produces a number of signs and neurobiological
alterations similar to the pathophysiology of major depression in humans (Grippo
et al., 2003). Finally, early life stress (ELS) has recently been demonstrated to play
an important role. Early life stress is a broad concept, of which physical and
sexual child abuse are undoubtedly the most extreme forms. Several studies have
shown that individuals exposed to severe ELS are particularly sensitive to develop
major depressive disorder as adults. A study on almost 2,000 women linked a
history of childhood abuse to an increase in depressive symptoms and suicide
attempts (McCauley et al., 1997). Childhood abuse has also been associated
with earlier onset and a more chronic course of depression (Heim & Nemeroff,
2001).
The personality structure of the individual confronted with stress is important as
well. For instance, many studies have shown that anxiety-related personality traits
increase the risk to develop major depression (Kendler et aI., 2004).

Biological Factors in Depression

Genetic factors play an important role in the pathogenesis of major depression.

Family studies show a high familial aggregation of the disorder: first degree family
members of a patient with major depression have a risk that is 1.5-2 times higher
than the general population (Jones et aI., 2002). Twin and adoption studies have
shown that a substantial part of this increased familial risk is genetic. The
heritability of the disorder (defined as the extent to which individual genetic
differences contribute to individual differences in observed behavior) is estimated
at 40-50% (Jones et al., 2002).

228
 Since the inception of the modern era of biological research in psychiatry, many
hypotheses have been generated regarding the patho-etiology of depression. There
is compelling evidence for a significant role of at least three different physiological
systems: the monoamine neural circuits (serotonin, norepinephrine and dopamine),
the immune system and the hypothalamic-pituitary-adrenal (HPA) axis. In those
three areas, specific abnormal patterns have been shown to accompany the clinical
phenotype of major depression. Because monoamine neurotransmission, immunity
and the HPA axis are functionally interrelated at multiple levels, all these findings
might reasonably be considered as different sides of the same coin.
In the I950s, two classes of antidepressants, monoamine oxidase inhibitors
(MAOIs) and tricyclic antidepressants (TCAs), were discovered serendipitously.
These compounds were subsequently found to enhance neurotransmission by
serotonin, norepinephrine and/or dopamine. For this reason, much biological
research has focused on a potential role of these three monoamines in the patho-
genesis of major depression. It is not the aim of this chapter to comprehensively
review the massive data emanating from these studies. A large number of
biochemical, challenge, imaging and post-mortem studies have demonstrated an
abnormally reduced function of central serotonergic systems in major depression
(Charney et aI., 1981; Lopez et al., 1998, Orevets et aI., 1999). Findings implicating
dysfunction of norepinephrine and dopamine in the pathogenesis of major
depression are also available (Mendels & Franzer, 1974; Meyers et aI., 1999;
Ressler & Nemeroff, 1999).
A second important field of biological research in major depression concerns the
immune system. There is evidence that major depression is characterized by
specific alterations in a number of important immune parameters. More speci-
fically, the disorder is associated with activation of several aspects of cellular
immunity, specifically the hypersecretion of proinflammatory cytokines (Kronfol,
2002; Musselman et al., 2003).
Finally, the most scrutiny has been directed towards the role of the hypothalamic
- pituitary - adrenal axis (HPA axis) in major depression, which is the main focus
of this chapter. The HPA axis plays a pre-eminent role in the endocrine, autonomic,
immunological and behavioral response to stress. Evidence from multiple sources
indicates that HPA axis hyperactivity is a common neurobiological feature of
MOD. There is considerable evidence that this HPA axis hyperactivity is caused by
a hypersecretion of the pre-eminent hormone which controls the activity of the axis,
which is corticotropin releasing factor (CRF). Therefore, we focus on the
physiology and pathophysiology of CRF in relation to affective disorders.
First, we describe the physiology of the HPA axis and of CRF, in both its
endocrine and extra-endocrine role. Subsequently, the evidence summarizing
alterations in CRF secretion and HPA axis function in major depression and in other
stress-related psychiatric disorders is reviewed. How does such dysregulation arise?
We describe how genetic factors, chronic stress and early life stress can all lead to
chronic CRF hypersecretion. Finally, directions for further research integrating
psychology and neurobiology are discussed in the last part of this chapter. This
research integrates biological and psychological theories of the pathogenesis of
major depression.

229
 CRF, Stress and Psychiatry

Physiological Functions of CRF

Vale and colleagues (1981) first elucidated the structure of a hypothalamic

hormone, corticotropin releasing factor (CRF), which controls the activity of the
HPA axis. In mammals, however, CRF has multiple extra-endocrine functions,
including orchestrating the behavioral, immune and autonomic responses to stress.
The primary role of hypothalamic CRF, as noted above, is to control the activity
of the HPA axis. The HPA axis plays a pivotal role in the response of the organism
to stress, broadly defined as any change in the environment threatening homeo-
stasis. Within seconds after exposure to stress, there is enhanced release of CRF
from the median eminence in the hypothalamus. Arginine vasopressin (AVP) is a
peptide hormone that augments the effect of CRF on the HPA axis, and that is
cosecreted with CRF from parvocellular hypothalamic cells. CRF and AVP are
transported by the portal vascular system to the anterior pituitary, where they
stimulate the secretion of adrenocorticotropin (ACTH) from pituitary corticotrophs
(Vale et aI., 1981), leading in humans to the secretion of cortisol from the adrenal
cortex. The HPA axis is schematically represented in Figure I.
As is the case of many biological systems, the HPA axis contains its own
negative feedback system, which regulates its activity, particularly after acute
activation. Cortisol acts as a negative feedback regulator at both the hypothalamus
and pituitary, decreasing the secretion of CRF and ACTH respectively (Swanson &
Simmons, 1989), and also exerts a negative feedback function on the HPA axis via
the hippocampus (Herman et aI., 1989). Cortisol acts at both mineralocorticoid
receptors (MR) and glucocorticoid receptors (OR). These two receptors differ in
their affinity for cortisol and in their distribution in the CNS. Under physiological
cortisol concentrations, negative feedback occurs mainly through the sensitive
MR's in the hippocampus. Under stress and high cortisol concentrations, negative
feedback is mainly achieved through OR's in the hippocampus, hypothalamus and
pituitary.
In response to stress, the secretion of CRF is increased not only in the
hypothalamus, but in the central nucleus of the amygdala as well. The amygdala
plays a seminal role in the emotional response to stress, including anxiety
(Davidson, 2002). The results of several studies have suggested that hypothalamic
CRF secretion is quite sensitive to physiological stress, whereas CRF secretion
from the amygdala is more reactive to psychological stress (Makino et aI., 1999).
Indeed, considerable evidence indicates that the amygdala plays a paramount
role in the autonomic, endocrine and behavioural responses to stress (Davis,
1992).
CRF is found not only in the hypothalamus and the amygdala, but in many other
brain regions, including the cerebral cortex, several limbic areas, and brainstem

230
 Hippocampus
OR
Dentate gyrus
MR

Hypothalamus
Paraventricular nucleus GR

CRF AVP_

Anterior Pituitary GR
ACTH

Adrenal cortex
Icortisol
FIGURE I.
Schematic representation of hypothalamus pituitary adrenal (HPA) axis function. Parvocellular neuron s in
the paraventricular nucleu s of the hypothal amus cosecrete corticoptropin-re leasing factor (CRF) and
vasopressin (AVP), which stimulate the secretion of adrenocorticotroph in (ACTH) from the anterior
pituitary into the blood. ACTH initiates the synthes is and secretion of cortisol from the adrenal cortex.
Con isol exerts a negative feedback control over the HPA axis. Under physiolog ical cortiso l co ncentrations,
negative feedback occurs mainly through the sensitive mineralocorticoid receptors (MR) in the
hippocampu s. Under stress and high cortisol concentrations, negative feedb ack is mainly achieved through
glucocorticoid receptors (GR) in the hippocampus, hypothalamus and pituitary.

nuclei, the latter responsible for autonomic function. A distinct population of CRF
containing PVN hypothalamic neurons stimulates the locus coeruleus in the
brainstem, leading to increased activity of noradrenergic neurons (Swanson et aI.,
1983) and of the sympathic autonomous nervous system . CRF also acts as a
neurotransmitter in the median and caudal dorsal raphe , suggesting a role in the
regulation of serotonin system s (Austin et aI., 2003) . Finally, CRF is expressed in
many peripheral tissues including the adrenal medulla, lymphocytes, pancreas ,
lung, liver, the gastrointestinal tract and skin (Orth, 1992).
The acute stress response, associated with a number of metabolic, neuro-
endocrine and autonomic adaptations, is largely mediated by CRE The se
adaptations include inhibition of sexual (Heinrichs et aI., 1997) and ingestive

231
behaviour (Heinrichs & Richard, 1999). It also inhibits growth hormone secretion
(Raza et aI., 1998) and the hypothalamic-pituitary-gonadal (HPG) axis (Nikolarakis
et aI., 1986). CRF-mediated secretion of cortisol from the adrenal cortex has
numerous acute adaptive effects, while chronic HPA axis activation is associated
with several adverse effects. Together with CRF, cortisol inhibits the growth
hormone and reproductive axes. It also reduces the cellular immune system.
Cortisol increases available energy by promoting gluconeogenesis, glycolysis and
proteolysis, and augmenting insulin resistance. The activation of noradrenergic
systems by CRF mediated activation of the locus coeruleus increases blood
pressure, heart rate and blood sugar, and decreases gastro-intestinal bloodflow. All
these adaptions allow the organism to react adequately to acute stressors and
threats. This so-called "fight or flight reaction" has obvious and crucial importance
in allowing the individual organism to respond adequately to acute threats,
increasing chances of survival.

CRF in Depressive Disorders

Major depressive disorder. Beginning in the late 1960s, numerous studies

reported cortisol hypersecretion in major depressive disorder (MOD), documented
by elevated plasma corticosteroid concentration (Carpenter & Bunney, 1971),
increased levels of cortisol metabolites (Sachar et aI., 1970) and elevated 24-hour
urinary free cortisol concentrations (McClure 1966; Carroll et al., 1976). In contrast
to Cushing's disease, hypercortisolism may not occur every day in depressed
patients. For example, urinary free cortisol excretion may be elevated for 21 days
out of the month in depressed patients compared with 4-5 days out of the month
for control subjects (Gold et aI., 2002). Hypercortisolemia in MOD is considered to
be a state marker, and not a trait marker, because it tends to normalize in most
patients after clinical improvement. Postmortem studies revealed enlarged pituitary
and adrenal glands in depressed patients, consistent with chronic HPA axis
hyperactivity (Nemeroff et aI., 1992).
Chronically elevated glucocorticoid levels have been shown to be associated
with loss of neurons in the hippocampus of mammals including rats and monkeys
(Sapolsky, 1990). Some investigations have found a decrease of hippocampal
volume in MOD patients (Bremner et aI., 2000), while others have not (Vakili et aI.,
2000). Posener et al. (2003), using high-resolution magnetic resonance imaging
(MRI), obtained brain scans from 27 patients with MOD and 42 healthy
comparison subjects. They found that MOD patients might have structural abnor-
malities of the hippocampus that can be detected by analysis of hippocampal shape
but not volume, which might partly explain the conflicting earlier findings.
Moreover, although hippocampal function seems to be disturbed after the first
depression, hippocampal volume probably only decreases after multiple depressive
episodes (McQueen et aI., 2003). Finally, hippocampal volume reduction has been

232
reported to be restricted to the specific subgroup of MDD patients with a history of
early life trauma (Vythilingam et aI., 2002).
The increased cortisol secretion in MDD has been associated with central CRF
hypersecretion. Several studies reported elevated CRF concentrations in
cerebrospinal fluid (CSF) of untreated MDD patients (Nemeroff et al., 1984; Banki
et aI., 1987; Risch et aI., 1992). Indeed, Raadsheer et al. (1995) observed that CRF
mRNA expression and CRF concentration in the PVN of depressed patients is
markedly higher than those of comparison subjects. When the concentration of
CRF is increased in the brain, downregulation of CRF receptors is to
be expected. Nemeroff et aI. (1988) found a marked (23%) reduction in the den-
sity of CRF receptor binding sites in the frontal cortex of suicide victims compared
with controls, and these findings were recently confirmed (Merali
et aI., 2004). Another recent study found a shift in the ratio of the two known types
of CRF receptors (CRF-RIIR2) in the pituitaries of suicide victims (Hiroi et aI.,
2001).
The hypersecretion of CRF clearly is not limited to the HPA axis, but is also
present in other brain regions where CRF acts as a neurotransmitter. In a recent
study, Austin et aI. (2003) studied post-mortem brain material from II male suicide
patients and II matched controls. They found that the level of CRF was increased
by 30% in the locus coeruleus, 39% in the median raphe and 45% in the caudal
dorsal raphe in depressed suicide subjects compared to controls. No difference in
CRF was found in the dorsal tegmentum or medial parabrachial nucleus between
the subject groups. These findings indicate that CRF availability is specifically
increased in norepinephrine- and serotonin-containing pontine nuclei of depressed
suicide men, and this is consistent with the hypothesis that CRF neurotransmission
is elevated in extra-hypothalamic brain regions of depressed subjects. The specific
involvement of the locus coeruleus and the raphe nuclei in this study provides a
direct link between CRF hypersecretion and disturbances in the monoamine
(noradrenaline and serotonin) systems which have long been posited to be integral
in the pathophysiology of MDD.
In addition to the finding of hypercortisolemia and CRF hypersecretion in MDD,
more sensitive provocative endocrine challenge tests such as the CRF stimulation
test have been evaluated. Intravenous CRF administration provokes a marked
increase of ACTH secretion from the pituitary gland in healthy control subjects. In
MDD patients, ACTH response is blunted (Heim et aI., 2001). This is at least in part
secondary to downregulation of anterior pituitary CRF receptors, in response to
CRF hypersecretion from the hypothalamus that is likely the primary cause of HPA
axis dysfunction in MDD (Nemeroff, 1992; Heim & Nemeroff, 1999). Recently,
Newport et aI. (2003) reported that the magnitude of the blunted ACTH response to
CRF is correlated with the patient's basal CRF concentrations in cerebrospinal fluid
(CSF).
Other HPA axis functional studies included long term monitoring with the
dexamethasone-suppression test (DST) in patients suffering from a depressive
episode (Holsboer et aI., 1988). The standard DST is composed of administration
of a low dose (1-2 mg) of dexamethasone (DEX) at 23:00 h and the subsequent
measurement of cortisol levels at one or more time points the following day. In

233
healthy controls, the secretion of ACTH and cortisol is suppressed. In contrast, in
depressed patients, the suppression of cortisol secretion by DEX is markedly
reduced. Numerous studies contributed to the DST literature (for a review, see APA
Task Force on Laboratory Tests in Psychiatry, 1987). The sensitivity of the DST
(rate of a positive outcome, or non-suppression of cortisol) in MDD is modest
(about 40-50%) but is higher (about 60-70%) in very severe, especially
psychotic, affective disorders, including MDD with psychotic as well as
melancholic features, mania, and schizoaffective disorder. The specificity (true
negative outcome) of the DST in normal control subjects is above 90%, but it varies
from less than 70% to more than 90% in psychiatric conditions that often need to
be separated from major affective disorders. In addition, a number of medical
conditions, including severe weight loss and use of alcohol and certain other drugs
(barbiturates, anticonvulsants, and others) can produce false positive results. Early
life trauma likely contributes to increased HPA axis activity (see also below).
Therefore, the usefulness of the DST as a diagnostic tool is limited. It can be used
as a predictive instrument, because failure to convert to normal suppression of
cortisol with apparent symptomatic recovery from depression is associated with an
increased risk for relapse into depression or suicidal behaviour (Ribeiro et aI.,
1993).
A considerably more sensitive test for HPA axis dysfunction is the combined
Dex-CRF-challenge test (Dex-CRF test) (Holsboer-Trachsler et al., 1991). In
healthy controls, the secretion of ACTH and cortisol is almost completely
suppressed by DEX administration, even after IV administration of CRE Depressed
patients release significantly more cortisol and ACTH after DEX + CRF in
comparison with age-matched controls. The sensitivity of the Dex-CRF test for
MDD (about 80%) greatly exceeds that of the standard DST. Increased HPA axis
activity as identified with the Dex-CRF test after clinical remission in depressed
patients is a strong predictor of relapse (Zobel et al., 1999). Recently, Krieg et aI.
(2001) reported abnormal Dex-CRF test results in asymptomatic first-degree
relatives of MDD patients.
Taken together, these studies provide convincing evidence of HPA axis
abnormalities in MDD. The concatenation of data suggests increased cortisol
concentrations in depressed patients, caused at least partly by CRF hypersecretion,
which is not adequately suppressed by any negative feedback system.
Chronic HPA axis overactivity leads to a number of serious adverse effects on
the organism (reviewed in Gold & Chrousos, 2002), including excessive fear and
anxiety, increased atherosclerosis (through insulin resistance and visceral fat
deposition), osteoporosis and decreased immune function. Remarkably, a number
of these findings have been reported in major depression (Thakore et aI., 1997). The
increase in intra-abdominal fat likely contributes to the increased incidence of
cardiovascular morbidity and mortality in depressed patients. Michelson et aI.
(1996) and Schweiger et aI. (2000) showed that past or current depression in
women is associated with decreased bone mineral density.
Recently, a preclinical study revealed that the mesolimbic dopaminergic (DA)
system is suppressed by hypercortisolemia (Pacak et aI., 2002). If confirmed this
preliminary finding would provide a link between HPA axis overactivity and an

234
impaired OA mediated reward system, possibly responsible for anhedonia (lack of
the ability to experience pleasure), a core feature of major depression. This would
add to the growing database supporting a role for OA circuit hypofunction in
depression.
An important question is whether the neurobiological changes described here
should be considered as concomitants, consequences, or causes of MOD. Most
studies suggest that the HPA axis overactivity tends to normalize after clinical
improvement (for a review, see Hatzinger, 2000). This seems compatible with
the hypothesis that the HPA axis dysfunction is a consequence or concomitant
of depression. However, in a subgroup of MOD patients, the function of the
HPA axis does not return to normal after clinical remission (Zobel et al.,
1999). Furthermore, in patients who fully recovered from bipolar depression,
Watson et al. found that the Oex-CRF test was abnormal, even though the
disturbance was smaller than in patients with current bipolar depression (Watson
et al., 2004). Finally, Holsboer et al. (1995) reported that the abnormal Oex-
CRF test results observed among patients with depression were also present in first-
degree relatives of these patients, who had no history of psychiatric illness
themselves. The most likely hypothesis is that a number of individuals show a
chronic liability for HPA axis hyperactivity, which can be attributed to genetic
factors or experiences of early abuse (see also below). This dysfunction aggravates
during MOD, and falls back to basal, but not necessarily normal, levels after
clinical remission.

Dysthymia. Dysthymia is a syndrome characterized by a prolonged period (at

least 2 years) with unremitting depressive symptoms, not sufficient to continuously
fulfil the criteria for MOD, but still very invalidating for the patient. Although
dysthymia was once viewed as a prominent feature of personality disorders, and
best treated with psychotherapy, it is now clear that the disorder also has biological
underpinnings, and can respond to antidepressant pharmacotherapy. Studies
regarding CRF and HPA axis function in dysthymia are scarce. Catalan et al. (1998)
studied 10 dysthymia patients, together with 36 MOD patients, and found a
comparable increase in serum CRF and cortisol concentrations in both groups.
However, the many issues associated with the measurement of CRF in serum and
its interpretation renders these findings preliminary. Oshima et al. (2000) tested 20
MOD patients, 8 dysthymia patients, and 12 healthy controls with the Oex-CRF
test. In the MOD group, the ACTH increase to the Oex-CRF test was significantly
increased compared to the controls. The dysthymia group exhibited ACTH levels
intermediate between MOD patients and controls, but not significantly different
from either groups; insufficient statistical power was available to test the
hypothesis. These findings suggest that dysthymia may share similar HPA axis
abnormalities compared to MOD, but further research is needed.

235
 Atypical depression and seasonal affective disorder. Atypical depression is
characterized by an inversion of the somatic symptoms seen in melancholic MOD.
Patients show hypersomnia, hyperphagia, lethargy, fatigue, and relative apathy.
Seasonal affective disorder (depression associated with a specific period of the
year) is clinically similar to atypical depression. As the somatic symptoms in
melancholic depression can be attributed to CRF hyperdrive, the symptoms in
atypical depression have been posited to be due to a hypofunction of the CRF s'ress
response system.
Casper et al. (1988) utilized the OST to measure cortisol levels in blood, CSF,
and urine in 45 hospitalized male and female patients with primary major
depressive disorder who reported hypersomnia or increase in appetite. They
provided preliminary evidence that HPA activity in depression is diminished in the
presence of hypersomnia and/or increased appetite. These findings were confirmed
and extended by a study of Geracioti et al. (1997), who reported that in 10 patients
with symptoms of atypical depression, CRF concentration in CSF and ACTH
concentrations in plasma tended to be low, with normal plasma cortisol. Gold and
Chrousos (2002) measured plasma ACTH and cortisol concentrations every three
minutes for a 24 h period in 4 patients with atypical depression and 4 controls.
Patients with atypical depression showed significant reductions in plasma ACTH
secretion in the presence of normal cortisol concentrations, compatible with
hypothalamic CRF hypoactivity. Although interesting, these studies are very snaIl
and should be considered preliminary.

CRF and other Stress-Related Psychiatric Disorders

Posttraumatic stress disorder. Experiences of overwhelming traumatic stress can

lead to posttraumatic stress disorder (PTSO), a syndrome characterized by re-
experiencing aspects of the traumatic experience and generalized arousal. Because
of the seminal role of stress in the disorder, several studies have scrutinized BPA
axis function in PTSO.
Similar to the findings in MOD, PTSO patients exhibit elevated CSF CRF
concentrations (Bremner et al., 1997; Baker et al., 1999). The ACTH response to a
standardized CRF stimulation test is diminished (Smith et al., 1989). One half of
the PTSO patients also fulfilled criteria for MOD in this study.
However, unlike MOD, basal and stress-induced cortisol output is reportedly
diminished in PTSO patients (For review, see Yehuda, 200 I). Stein et al. (1997)
studied 19 female survivors of childhood and/or adolescent sexual abuse. These
women had significantly enhanced suppression of plasma cortisol in response to a
low dose (0.5 mg) of OEX compared to women without an abuse history. These
observations are consistent with findings in male veterans with combat-related
PTSO (Yehuda et al., 1995; Kellner et al., 1997). This suggests that, in PTSO, the
negative feedback of the HPA axis by cortisol is enhanced. Perhaps the low cortisol

236
secretion leads to an upregulation of glucocorticoid receptors (GR's), resulting in
enhanced negative feedback of the system in response to a low dose of DEX.
Upregulation of GR's was indeed observed in lymphocytes of PTSD patients
(Yehuda et al., 1993).
In a recent study, Rinne et al. (2002) studied 39 patients with borderline
personality disorder (BPD) with (n= 24) and without (n= 15) a history of childhood
abuse and comorbid PTSD (n= 12) or MDD (n= II) and II healthy control
subjects. They observed super-suppression of cortisol and ACTH using the Dex-
CRF test in the PTSD group.
Taken together, these data suggest that PTSD, like MDD, is characterized by
central CRF hyperdrive, but that in contrast to MDD, the negative feedback of the
HPA axis is enhanced in PTSD, resulting in a low cortisol output.

Chronic fatigue syndrome and fibromyalgia. Chronic fatigue syndrome (CFS) is

characterized by persistent or relapsing debilitating fatigue for at least 6 months in
the absence of any medical diagnosis that would explain the clinical presentation.
CFS and fibromyalgia, two clinically related syndromes, are etiologically complex
disorders. Infectious, immunological, neuroendocrine, sleep, and psychiatric
mechanisms have been investigated; however, a unifying etiology has yet to
emerge. Exposure to stress increases the vulnerability for CFS (Kang et al., 2003),
and a recent comprehensive review of neuroendocrine studies reported an altered
physiological response to stress (Parker, Wessely, & Cleare, 200 I). Clinically, CFS
shares some similar features with atypical depression. Neuroendocrinological
studies suggest a disturbance of the HPA axis similar to that observed in atypical
depression, characterized by CRF hyposecretion. Low basal cortisol values in
plasma and low 24-hour urinary free cortisol excretion have been reported in CFS
compared to healthy controls in some, but not all, studies (Demitrack et al., 1991).
In a recent review, Parker et al. (200 I) found that approximately one-third of the
studies reported baseline cortisol secretion in CFS and fibromyalgia patients to be
low, in approximately one-third of patients. Scott et al. (1999) found a volume
reduction of over 50% of the adrenal glands in 8 CFS subjects with abnormal
endocrine parameters (compared to 55 controls), indicative of significant adrenal
atrophy. Similarly, 24-hour urinary free cortisol levels of patients with fibromyalgia
tend to be in the lower part of the normal range. With respect to dynamic
responsiveness of the HPA axis, the DST results are normal in CFS, but marked
ACTH hypersecretion in response to severe acute stressors has been reported,
perhaps indicative of chronic CRF hyposecretion and CRF receptor upregulation
(Geenen et al., 2002).

237
 Factors Underlying Persistent CRF Dysregulation

There is clearly abundant data indicative of dysregulation of the CRF system in

major depression, and in a number of other stress-related neuropsychiatric
disorders. The next logical question is how such dysregulation arises. In the case of
major depression, the question can be formulated more specifically: Are the
biological and psychological factors that are known to increase the risk for major
depression individually and collectively leading to a lasting disruption of the CRF
and HPA axis system? In the next section, we scrutinize three well established risk
factors for major depression, and evaluate their possible influence on the HPA axis:
genetic factors, chronic stress and early life stress.

Genetic Factors

As stated earlier, family, twin and adoption studies have all shown that genetic
factors play an important role in the vulnerability for major depression. The
question here is whether genetic variations could contribute to persistent
dysregulation of CRF secretion as well as HPA axis function, which would render
such individuals more prone to depression.
Holsboer et al. (1995) reported that the HPA feedback disturbance observed
among patients with depression was also present in otherwise healthy individuals
who are at high risk for psychiatric disorders because they have a first-degree
relative with an affective illness. Moreover, this disturbance was shown to be stable
over a 5-year period (Modell et al., 1998). These data suggest that some individuals
have a genetically determined vulnerability to develop chronic CRF hypersecretion
and MOD. Whether they actually develop MOD probably depends on a number of
factors such as early trauma experiences, chronic and acute stress.
It is not yet clear which genes underlie the individual susceptibility for such
dysregulation. Several studies are currently underway examining the impact of a
number of genes implicated in CRF and HPA axis regulation. In some cases, animal
models have been developed that under- or overexpress these genes, and the stress
coping and depression/anxiety-related behaviors of these transgenic animals have
been evaluated. Other studies have searched for possible associations between
polymorphisms of these genes and MOD (for a review, see Claes, 2004).
As stated earlier, CRF and AVP are the main neurohormones activating the HPA
axis, and therefore genes coding for these hormones and their receptors are obviolls
functional candidate genes. There are 2 CRF receptor subtypes, CRF-RI and CRF-
R2, and the availability of CRF is further regulated by its binding to the CRF
binding protein (CRF-BP). Preclinical studies have examined CRF-RI gene
"knockout" mice, transgenic mice in which both copies of the CRF-RI gene have
been inactivated by molecular genetic techniques. These animals exhibit HPA axis
disruption and reduced anxiety, indicative of a key role for this gene in the

238
development of a functional HPA axis (Timple et a\., 1998). CRF-R2 knockout
mice have been reported to be hypersensitive to stress, though discordant findings
exist (Coste et a\., 2000; Bale et a\., 2000; Kishimoto et a\., 2000). CRF-BP
knockout mice exhibit increased anxiety on several behavioral measures (Seasholtz
et aI., 200 I). In humans, there are no systematic studies looking at the association
of the CRF-R I gene with major depression. A preliminary genetic study revealed
no association between CRF-R2 and major depression (Villafuerte et aI., 2002). In
contrast, the gene encoding for CRF-BP was associated with increased
vulnerability for major depression in a Swedish population (Claes et aI., 2003).
Because most patients with major depression show a deficient negative feedback
of the HPA axis, and because this feedback is regulated through the GR, this
represents another functional candidate gene. This is also true for the genes coding
for the so-called chaperone molecules, proteins that interact functionally with the
GR protein.
From animal models, a central role of the GR gene in stress coping seems
plausible. Complete absence of functional GR is incompatible with life (Reichard,
1998). However, mice with functionally impaired GR have been studied, as well as
mice with tissue-specific knockout of the GR gene. These animals show profound
behavioral changes and elevated plasma corticotropin concentrations in response to
stress. Treatment with moclobemide, a monoamine oxidase inhibitor
antidepressant, reversed the behavioral deficits in this mouse model. Tronche et al.
(1999) generated viable adult mice with loss of GR function in selected tissues.
Loss of GR in the central nervous system (CNS) impaired regulation of the HPA
axis, resulting in increased glucocorticoid plasma levels. Conditional mutagenesis
of the GR in the CNS resulted in an impaired behavioral response to stress and
reduced anxiety.
In humans, several genetic polymorphisms have been identified in the GR gene
(De Rijk et aI., 2002 for a review). Some of these influence the regulation of the
HPA axis, and are associated with metabolic and cardiovascular alterations. Data on
GR polymorphisms and psychiatric disorders are scarce. Five missense variants in
the amino-terminal domain of the GR showed no association with puerperal
psychosis or schizophrenia (Feng et aI., 2000). An association analysis between GR
polymorphisms and the vulnerability for MOD in two independent populations has
been performed by one of us (SC) and no evidence for an association was observed
(Van West et aI., submitted).

Chronic Stress

In chronic stress, multiple mechanisms further activate the CRF system. These
mechanisms have been described by Makino et al. (2002). In situations of chronic
arousal, the glucocorticoid negative feedback system is less effective, possibly by
downregulation of hippocampal GR. Furthermore, glucocorticoids upregulate CRF

239
secretion in the amygdala and increase the expression of CRF receptors in the PVN
(Rivest et aI., 1995). These data perhaps explain why in some cases, chronic stress
does not lead to HPA axis downregulation through the negative feedback effects of
cortisol, but to sustained CRF hypersecretion, which likely contributes to MOD.

Early Life Stress

At birth, the stress response system is not fully mature. In rodents, the postnatal
HPA axis is remarkably different from that of the adult, both in structure and
function. The first 2 weeks postnatally are characterized by a "silent period" during
which the developing animal is relatively hyporesponsive to stress (the so-called
stress hyporesponsive period), followed by a new and unique phase of stress
responsiveness when the animal fails to swiftly terminate glucocorticoid secretion
(Vazquez, 1998). Increasing evidence from animal and human studies has revealed
a critical period postnatally when sustained stress (trauma) disrupts the normal
development of the stress response system. We will review the available data in
three different species: rats, non-human primates and humans. A comprehensive
review on this topic has recently been published (Heirn et aI., 2004).
Rats. Plotsky et aI. (1993) studied rat pups 2-14 days of age, that were exposed
daily to handling (H) (15 min of separation from mother and home cage), maternal
separation (MS; 180 min of comparable separation), or were left entirely
undisturbed (non-handled; NH). As adults, MS rats showed increased hypothalamic
CRF mRNA expression compared with NH rats. CRF mRNA expression in H rats
were significantly lower than either MS or NH animals. Hypothalamic CRF
concentration under basal conditions exhibited the same pattern as CRF mR\lA
expression. Restraint stress produced significantly greater increases in plasma
corticosterone in MS and NH animals than in H animals. In a subsequent
experiment, Ladd et aI. (1996) found that adult male rats previously isolated for 6
hours daily during postnatal days 2-20 exhibited an increase in both basal and
stress-induced serum ACTH concentrations. Moreover, these rats exhibited a 12:5%
increase in immunoreactive CRF concentrations in the median eminence and a
reduction in the density of CRF receptor binding in the anterior pituitary.
Alterations in extrahypothalamic CRF systems were also apparent: these animals
showed a 59% increase in the number of CRF receptor I - binding sites in the raphe
nucleus. Marked increases in CRF mRNA expression and concentration in the
central nucleus of the amygdala, the bed nucleus of the stria terminal is and the
locus coeruleus of the adult exposed to ELS were also observed. These results
indicate that maternal deprivation before weaning in male rats produces effects on
CRF neural systems throughout the central nervous system and pituitary that are
apparent several months later and are associated with persistent alterations in
behavior. In a recent study, Huot et aI. (2002) confirmed the persistent HPA axis
changes in maternally deprived rats, and found that maternal separation and

240
subsequent reunion with the dam resulted in elevated plasma corticosterone levels
at a time when rat pups are normally hyporesponsive to stressors and show limited
pituitary-adrenal responses. Structural studies in these traumatized rats showed
decreased mossy fiber density in the stratum oriens region of the hippocampus.
These changes may be the result of neonatal exposure to elevated glucocorticoids
and/or changes in other signaling systems in response to maternal separation.
Overall the results suggest that repeated, daily, 3 h maternal separations during
critical periods of hippocampal development can disrupt hippocampal
cytoarchitecture in a stable manner. The resulting change in morphology may
contribute to the subtle but consistent learning deficit and overall stress hyper-
responsive phenotype observed in these animals.
Primates. Similar to the studies in rodents, several investigators have evaluated
the effects of prolonged deprivation or repeated separations of infant non-human
primates from their mothers or peers. Evidently, primates are more valid models for
human behavior compared to rats. Monkey and human infants indeed demonstrate
similar acute behavioral and physiological reactions to separations (Sanchez et aI.,
2001). When tested as adults, non-human primates exposed to prolonged periods of
maternal or social deprivation exhibit marked behavioral, physiological, and
neurobiological changes. These include increased fearfulness and anxiety, social
dysfunction, aggression, aberrant stereotypic and self-directed behaviors, altered
ingestion and anhedonia, as well as altered neurochemical, immune, and autonomic
function (reviewed in Sanchez et aI., 200 I; Pryce et al., 2002; Gilmer & McKinney,
2003).
Two studies on squirrel and marmoset monkeys respectively, showed lowered
basal and stress induced cortisol excretion and stronger negative feedback of the
HPA axis in monkeys exposed to separation stress early in life (Lyons et aI., 1999,
2000; Dettling et al., 2002). Infant bonnet macaques raised by mothers foraging
under unpredictable conditions exhibited persistently elevated cerebrospinal fluid
(CSF) concentrations of CRF as adults, compared to monkeys reared by mothers
foraging under predictable conditions (Coplan et aI., 1996). Summarizing, these
studies suggest that monkeys with early adverse experiences show hyperactivity of
CNS CRF systems, a strong negative feedback and a reduced cortisol output,
similar to what is seen in human individuals with PTSD.
Humans. More recently, studies in humans have become available. The most
common forms of ELS in humans are sexual, physical, and emotional
maltreatment, as well as parental loss. Other forms of ELS include accidents,
surgeries and chronic illness, natural disasters, war and terrorism-related events.
Heim et aI. (2000) studied women with a history of childhood abuse, some of which
fulfilled criteria for major depression. Traumatized women exhibited increased
pituitary-adrenal and autonomic responses to stress compared with controls. This
effect was particularly robust in women with current major depression. Indeed,
women with a history of childhood abuse and a current major depression diagnosis
exhibited a more than 6-fold increase in ACTH response to stress.
In a subsequent study by the same group (Vythilingham et al., 2002),
hippocampal volume was measured using MRI. A smaller hippocampal volume in
adult women with major depressive disorder was found only in the group with a

241
history of severe and prolonged physical and/or sexual abuse in childhood. Taken
together, the data in humans, although preliminary, seem comparable to the
observations in rodents and primates: specific stressors early in life can cause
structural changes in the limbic system (hippocampus and probably other
structures) and permanently dysregulate the stress response system.
It is probable that the lifelong disturbance of HPA axis function induced by early
life trauma is dependent on several factors: the critical time-window, the nature: of
the stressor, the presence or absence of a supportive environment and the genetic
liability of the individual. Elucidation of these factors is an important target for
subsequent research.
It is important to be aware of the limitations in comparing animal and human
studies. In animal studies, the paradigm used is a model of maternal neglect, which
differs from the abuse experiences studied in humans. Whether maternal neglect
and abuse lead to similar neurobiological consequences, also needs to be
established in further research.

Outlook: Integrative Research

In CRF and HPA axis research, integration of psychological and neurobiological

parameters of major depression has already begun. The most obvious example is
our growing understanding of the neurobiology of early life stress. The concept of
early physical or sexual abuse leading to lifelong disturbances in stress resilience,
and to an increased vulnerability for stress-related disorders such as major
depression will sound familiar to psychodynamic psychotherapists. Many questions
remain to be answered in this area. Which specific early life stressors are most
disruptive of the neurobiology of the stress response system? Looking more closely
at the specific model of early life stress in rats, the neurobiological consequences
are not solely due to the effect of being taken away from the dam for three hours a
day, but also a response to the way these pups are treated by the dam once they He
returned to the nest.
It is generally believed that there is a critical time window, stretching from
infancy to puberty, in which abuse experiences are particularly deleterious, but we
have only scarce scientific data to support this assumption. We don't know whether
this is one homogenous period, or whether it should be subdivided into different
periods, each having a unique sensitivity for HPA axis and CRF circuit disruption
by specific stressors.
A number of factors are thought to protect children against the negative effects
of early life stress: attachment to trustworthy important others and certain
personality traits. Whether these factors can mitigate the negative effect of early life
trauma on the function of the HPA axis, is unknown.
Another clinical observation is that there are clearly individual differences in the
response to the deleterious effects of early abuse. In some families, several children

242
will undergo important early life stress, but not all of them will develop psychiatric
disorders. Genetic factors most probably play some role in such individual
differences, but more research is needed to identify which genes are involved.
Future studies should further evaluate whether ELS associated alterations in the
HPA axis and the other demonstrated biological alterations, normalize in response
to different treatment regimens in depression. Indeed, little is known in clinical
populations as to whether certain treatments can alter or reverse the neurobiological
consequences of ELS. In rat models, SSRI's such as paroxetine reversed the effects
of maternal separation (Huot et aI., 2001).
If childhood trauma leads to disruptions in HPA axis function in genetically
sensitive individuals, would pharmacotherapy or psychotherapy prevent the
development of psychiatric disorders such as major depression in these individuals?
If so, which form of psychotherapy or pharmacotherapy should be utilized and for
how long? One recent study showed that treatment with fluvoxamine, an SSRI,
reversed HPA axis hyper-responsiveness in the dexamethasone/CRF test,
particularly in those borderline personality disordered patients who reported
sustained histories of child abuse, independent of symptomatic status (Rinne et al.,
2003). In order to improve the treatment response in depression, it is likely that new
treatment strategies directly target the underlying pathophysiology. These
treatments that directly target the neuronal circuits and mechanisms that are
modified by ELS, such as CRF receptor antagonists, might be particularly effective
in the treatment or prevention of depression related to ELS.
However promising, this type of integrative research is fraught with difficulties.
First, it is labor intensive, requiring collaboration between experienced clinicians
and neurobiologists. One of the key problems is gathering retrospective information
about early childhood experiences, and linking this to sensitive functional tests of
the HPA axis. As Sigmund Freud already discovered, gathering retrospective
information about abuse experiences in childhood is notoriously difficult.
In order to address complex questions such as the interaction between ge-
netic liability and specific early life stressors, very large samples of trauma-
tized patients and healthy control persons will be needed. For all these reasons,
it would be inappropriate to assume that these questions can be solved rapidly.
However, the importance of this type of integrative research cannot be oversta-
ted.
Many of these factors are also relevant when assessing the impact of chronic life
stress on the HPA axis and subsequent vulnerability to stress related psychiatric
disorders. We have solid data linking chronic stress to chronic CRF hypersecretion,
but many questions remain. Which forms of chronic stress are most relevant? How
do they interact with genetic factors to lead to chronic CRF hypersecretion? What
is the role of personality in this matter? Which factors protect an individual against
HPA axis disruption by chronic stress?
In terms of the question of specificity of stressors, the available data are simply
too discordant to come to any firm conclusions and we must await additional data.
It is known that physical stressors such as blood volume loss and psychological
stressors such as fear elicit clearly distinct neurobiological responses.
Neurobiological and neuroendocrinological responses to different psychosocial

243
stressors also likely exist but it is premature to draw any firm conclusions in this
area.
These and related questions represent a fascinating challenge for psycho-
therapists, psychiatrists and neurobiologists alike, and these groups will need to
collaborate in order to achieve progress. Alterations in HPA axis activity have been
reported in several different psychiatric disorders. As expected, these endocrine
alterations are not limited to specific DSM-IV diagnostic entities. The central CRF
hyperdrive seen in MDD is, for example, similar to what is seen in anorexia nervosa
and end stage Alzheimer's disease. The data in atypical depression, chronic fatigue
syndrome and fibromyalgia, although preliminary, seem more compatible with a
CRF hypofunction, characterized by low cortisol secretion. PTSD is characterized
by central CRF hyperdrive and reduced cortisol secretion possibly due to strong
negative feedback associated with an upregulation of glucocorticoid receptors.
These data suggest the existence of specific HPA axis dysfunctional phenotypes
that transcend DSM-IV diagnostic boundaries. For that reason, future research
efforts should emphasize the link between early adverse experiences, genetic
factors, specific types of HPA axis function and clinical symptoms, regardless of
the DSM category in which these symptoms occur. Such an approach should lead
to a disease classification system with increased etiological and neurobiological
validity, compared to the current DSM-IV classification.

Conclusion

CRF circuits play a major role in the physiological adaptations to acute and
chronic stress. CRF is the main regulator of the HPA axis, but also acts as a
neurotransmitter in numerous brain regions. Whether or not this stress response
system functions adequately is probably determined by the interplay of genetic
factors and environmental events. A growing body of evidence from preclinical and
human studies indicates that adverse experiences during a critical period in
childhood can lead to persisting dysfunctions of CRF and HPA axis function
throughout adult life. These findings are changing the face of psychiatry, because
they make the link between neurobiological and psychological factors underlying
psychiatric disorders.
Numerous studies have revealed HPA axis abnormalities in MDD, and data are
also accumulating for other psychiatric disorders. Based on these findings, an
integrative view of the pathogenesis of MDD can be formulated. This is graphically
demonstrated in figure 2. Genetic factors and adverse early experiences can lead to
a stable vulnerability pattern of increased sensitivity to stress, CRF hyperdrive in
response to stress, and hippocampal shrinkage. When individuals with this
phenotype are confronted with acute or chronic stress, further CRF hyperdrive and
HPA axis dysregulation occur, leading to the clinical phenotype of MDD, possibly
involving changes in monoamine neurotransmitter and immune function.

244
 Genetic
Early adverse experience
vulnerability

• Increased sensitivity to .....

stress
Acute and
chronic Vulnerable phenotype •• CRF
CRF hyperdrive
hyperdrive in
in response
response toto stress
stress
•• abnormal
abnormal development
development of of hippocampus
hippocampus
stressors

Serotonin
Noradrenaline Major depressive
Chronic CRH hyperdrive Immunity
disorder

FIGURE 2.
An integrative model. Genetic vulnerability and early life stress (or the interaction between both) lead to
persistent dysregulations of the HPA axis, characterized by increased stress sensitivity, CRF hyperdrive and
neuronal damage in the hippocampus (and possible other brain areas). When individuals with this
vulnerable phenotype are confronted with chronic stress or negative life events, the system is prone to
further dysregulation and CRF hyperdrive. This will in turn cause changes in other neurobiological systems
such as the immune system and neurotransmission by serotonin and norepinephrine. The combination of
these neurobiological changes leads to the development of clinical psychiatric disorders such as major
depression or other stress related disorders.

The central CRF hyperdrive seen in MOD is probably very similar to what is
seen in dysthymia. The data in atypical depression, chronic fatigue syndrome and
fibromyalgia, though preliminary, seem more compatible with CRF hypoactivity,
characterized by low cortisol secretion. In PTSD, central CRF hyperdrive goes
together with a strong negative feedback and low cortisol output.
These data suggest the existence of specific HPA axis dysfunctional phenotypes
irrespective of DSM diagnostic borders. The available evidence indicates that
several stress-related psychiatric disorders are characterized by similar patterns of
CRF circuit abnormalities. This challenges the borders DSM has defined between
disorders on the basis of clinical criteria. In future OSM editions, clinical entities
that have been separated might be reunited based on a common neurobiological
background.
It is possible that the finding of central CRF hyperdrive in MOD is in fact
restricted to a (large) proportion of MOD patients: those with a history of childhood
abuse (Nemeroff, 2003). This is also compatible with the study of Rinne et al.
(2002), who found CRF hyperdrive in borderline personality patients with
childhood abuse, regardless of whether they were depressed or not.
These data suggest the existence of a specific vulnerability phenotype, caused by
sustained childhood abuse, and characterized by hyperreactivity to stress, central
CRF hyperdrive, hypersecretion of cortisol and impaired function and decreased
volume of the hippocampus (Nemeroff, 2003). Whether this phenotype leads to
MOD or other stress related psychiatric diseases in a particular individual, would
then depend on other, yet unknown, genetic and environmental factors.

245
 A large and fascinating field for research integrating psychological and
neurobiological factors in psychiatry lies open before us. It seems appropriate to
warn against undue optimism however: This research will require large numbers of
patients being investigated by experienced clinicians, neurobiologists and
geneticists, and will therefore be laborious, time consuming and expensive.
Nevertheless, it is only this kind of scientific approach that can do justice to the
complex nature of psychiatric disorders, and will undoubtedly lead to better
treatments for our patients.

Acknowledgements

The authors are supported by the following grants: NIH MH-92088 (CBN), MH-
39415 (CBN), MH-58922 (CBN).

References

APA Task Force on Laboratory Tests in Psychiatry (1987). The dexamethasone suppression test: An
overview of its current status in psychiatry. American Journal of Psychiatry, 144, 1253-1262.
American Psychiatric Association (1994). Diagnostic and statistical manual of mental disorders (4th ed.).
Washington, DC: Author.
Austin, M. C., Janosky, J. E., & Murphy, H. A. (2003). Increased corticotropin-releasing hormone immuno-
reactivity in monoamine-containing pontine nuclei of depressed suicide men. Molecular Psychiatry, 8,
324-332.
Baker, D. G., West, S. A., Nicholson, W. E., Ekhator, N. N., Kasckow, J. W., Hill, K. K., Bruce, A. 8., Orth,
D. N., & Geracioti, T. D., Jr. (1999). Serial CSF corticotropin-releasing hormone levels and
adrenocortical activity in combat veterans with posttraumatic stress disorder. American Journal of
Psychiatry, 156, 585-588.
Bale, T. L., Contarino. A., Smith, G. w., Chan, R., Gold, L. H., Sawchenko, P. E., Koob, G. E, Yak, W.
w., & Lee, K. E (2000). Mice deficient for corticotropin-releasing hormone receptor-2 display anxiety-
like behaviour and are hypersensitive to stress. Nature Genetics, 24, 410-414.
Banki, C. M., Bissette, G., Arato, M., O'Connor, L., & Nemeroff, C. B. (1987). CSF corticotropin-
releasing factor-like immunoreactivity in depression and schizophrenia. American Journal of
Psychiatry, 144,873-877.
Blazer, D. G., Kessler, R. C, McGonagle, K. A., & Swartz, M. S. (1994). The prevalence and distribution
of major depression in a national community sample: The National Comorbidity Survey. American
Journal of Psychiatry, 151,979-986.
Bremner, 1. D., Licinio, J., Darnell, A., Krystal, 1. H., Owens, M.J., Southwick, S. M., Nemeroff, C. 8., &
Charney, D. S. (1997). Elevated CSF corticotropin-releasing factor concentrations in posttraumatic
stress disorder. American Journal of Psychiatry, 154, 624-629.
Bremner, J. D., Narayan, M., Anderson, E. R., Staib, L. H., Miller, H. L., & Charney, D. S. (2COO).
Hippocampal volume reduction in major depression. American Journal of Psychiatry, 157, 115-1 J 8.
Carpenter, W. T. Jr., & Bunney, W. E. Jr. (1971). Adrenal cortical activity in depressive illness. American
Journal of Psychiatry, 128, 31-40.
Carroll, B. J., Curtis, G. c., Davies, B. M., Mendels, J., & Sugerman, A. A. (1976). Urinary free cor:isol

246
 excretion in depression. Psychological Medicine. 6. 43-50.
Casper. R. C .• Kocsis. 1.. Dysken. M.• Stokes. P.. Croughan, J.• & Maas J. (1988). Cortisol measures in
primary major depressive disorder with hypersomnia or appetite increase. Journal I~I' Affective
Disorders. 15. 131-140.
Catalan. R.• Gallarta, J. M.. Castellanosa, J. M.• & Galarda, R. (1998). Plasma corticotropin-releasing
factor in depressive disorders. Biological Psychiatry. 44. 15-20.
Charney. D. S.. Menkes. D. B.• & Heninger. G. R. (1981). Receptor sensitivity and the mechanism of action
of antidepressant treatment. Implications for the etiology and therapy of depression. Archives I~I'
General Psychiatrv, 38. 1160-1180.
Claes, S. J. (2004). CRH. stress. and major depression: A psychobiological interplay. Vitamins and
Hormones. 69. 117-150.
Claes. S.• Villafuerte. S.. Forsgren. T.. Sluijs, S.• Del-Favero. 1.. Adolfsson. R.. & Van Broeckhoven. C.
(2003). The CRF binding protein is associated with major depression in a population from Northern
Sweden. Biological Psychiatry. 54. 867-872.
Coplan. J. D.• Andrews. M. w.. Rosenblum. L. A.. Owens. M. 1.. Friedman. S.. Gorman. J. M.. &
Nemeroff. C. B. (1996). Persistent elevations of cerebrospinal fluid concentrations of corticotropin-
releasing factor in adult nonhuman primates exposed to early-life stressors: Implications for the
pathophysiology of mood and anxiety disorders. Proceedings of the National Academy "I' Sciences I~I'
the United States ofAmerica. 93. 1619-1623.
Coste. S. C .. Kesterson. R. A.. Heldwein, K. A.• Stevens. S. L.. Heard. A. D.. Hollis. J. H.. Murray. S. E.•
Hill. J. K.• Pantely. G. A.. Hohimer, A. R.• Hatton. D. C.. Phillips. T. J.. Finn. D. A.. Low. M. 1..
Rittenberg. M. B.. Stenzel. P.. & Stenzel-Poore. M. P. (2000). Abnormal adaptations to stress and
impaired cardiovascular function in mice lacking corticotropin-releasing hormone receptor-Z. Nature
Genetics. 24. 403-409.
Davidson. R. J. (2002). Anxiety and affective style: Role of prefrontal cortex and amygdala. Biological
Psychiatry. 51. 68-80.
Davis. M. (1992). The role of the amygdala in conditioned fear. In J. P. Aggleton (Ed.), The amygdala (pp.
255-3(6). New York: Wiley.
Dernitrack. M. A.. Dale. 1. K.. Straus. S. E.• Laue. L.. Listwak, S. J.• Kruesi, M. J.• Chrousos, G. P.• & Gold.
P. W. (1991). Evidence for impaired activation of the hypothalamic-pituitary-adrenal axis in patients
with chronic fatigue syndrome. The Journal 01' Clinical Endocrinology and Metabolism. 73. 1224-
1234.
DeRijk. R. H.. Schaaf. M.. & de Kloet, E.R. (2002). Glucocorticoid receptor variants: Clinical
implications. Journal ofSteroid Biochemistry and Molecular Biology. 81. 103-122.
Dettling. A. c.. Feldon, J.• & Pryce. C. R. (2002). Repeated parental deprivation in the infant common
marmoset (Callithrix jacchus, primates) and analysis of its effects on early development. Biological
Psvchiatry.Bl, 1037-1046.
Drevets. W. c.. Frank. E.. Price. J. C .. Kupfer. D. J.. Holt. D.. Greer. P. J.. Huang. Y. Gautier. C; & Mathis.
C. (1999). PET imaging of serotonin IA receptor binding in depression. Biological Psychiatry. 46.
1375-1387.
Feng, 1.. Zheng. 1.. Bennett. W. P.• Heston. L. L. Jones. I. R.. Craddock. N.. & Sommer. S. S. (2000). Five
missense variants in the amino-terminal domain of the glucocorticoid receptor: No association with
puerperal psychosis or schizophrenia. American Journal ofMedical Genetics. 96. 412-417.
Frank. E.. Kupfer. D. J.. Perel. J. M.. Comes, C .. Jarrett. D. B.. Mallinger. A. G.. Thase, M. E.. Mcfiachran,
A. B.. & Grochocinski, V. J. (1990). Three-year outcomes for maintenance therapies in recurrent
depression. Archives 01' General Psvchiatry, 47. 1093-1099.
Geenen, R.. Jacobs. J. W.. & Bijlsma, J. W. (2002). Evaluation and management of endocrine dysfunction
in libromyalgia. Rheumatic Disease Clinics "I' North America. 28. 389-404.
Geracioti, T. D. Jr.. Loosen. P. T.. & Orth, D. N. (1997). Low cerebrospinal tluid corticotropin-releasing
hormone concentrations in eucortisolemic depression. Biological Psvchiatry, 42. 165-174.
Gilmer. W. S.. & McKinney. W. T. (2003). Early experience and depressive disorders: Human and non-
human primate studies. Journal ofAfj'eetil'e Disorders. 75.97-113.
Gold. P. w.. & Chrousos, G. P. (2002). Organization of the stress system and its dysregulation in
melancholic and atypical depression: High vs low CRF/NE states. Molecular Psychiatrv, 7. 254-275.
Gold. P. W.. Drevets, W. c.. & Charney. D. S. (2002). New insights into the role of cortisol and the
glucocorticoid receptor in severe depression. Biological Psvchiatrv, 52. 381-385.
Grippo. A. 1.. Beltz. T. G.. & Johnson. A. K (2003). Behavioral and cardiovascular changes in the chronic

247
 mild stress model of depression. Phvsiology & Behavior, 78, 703-710.
Hatzinger, M. (2000). Neuropeptides and the hypothalamic-pituitary-adrenocortical (HPA) system: Review
of recent research strategies in depression. World Journal of Biological Psychiatry, I, 105-111.
Heim, C; & Nemeroff, C. B. (1999). The impact of early adverse experiences on brain systems involved
in the pathophysiology of anxiety and affective disorders. Biological Psychiatry, 46, 1509-1522.
Heim, C., & Nemeroff, C. B. (2001). The role of childhood trauma in the neurobiology of mooc. and
anxiety disorders: Preclinical and clinical studies. Biological Psychiatry, 49, 1023-1039.
Heim, C., Newport, D. 1., Bonsall. R.. Miller, A. H., & Nemeroff, C. B. (2001). Altered pituitary-adrenal
axis responses to provocative challenge tests in adult survivors of childhood abuse. American Journal
of Psychiatry, 158. 575-581.
Heim, C.. Newport, D. J., Heit, S.. Graham, Y. P., Wilcox. M.. Bonsall, R., Miller. A. H., & Nemeroff. C.
B. (2000). Pituitary-adrenal and autonomic responses to stress in women after sexual and physical
abuse in childhood. Journal of the American Medical Association, 284. 592-597.
Heim, C.. Plotsky, P. M., & Nemeroff, C. B. (2004). Importance of studying the contributions of early
adverse experience to neurobiological findings in depression. Neuropsvchopharmacology. 29, 641· 648.
Heinrichs, S. C.• Min, H., Tamraz. S.. Carmouche. M.. Boehme. S. A.• & Vale, W. W. (1997). Anti-sexual
and anxiogenic behavioral consequences of corticotropin-releasing factor overexpression are centrally
mediated. Psychoneuroendocrinology, 22. 215-224.
Heinrichs, S. C, & Richard, D. (1999). The role of corticotropin-releasing factor and urocortin in the
modulation of ingestive behavior. Neuropeptides, 33, 350-359.
Herman, J. P.• Schafer, M. K.. Young. E. A.. Thompson, R.. Douglass. J.• Akil, H.. & Watson, S. J. (1989).
Evidence for hippocampal regulation of neuroendocrine neurons of the hypothalamo-pituirary-
adrenocortical axis. Journal of Neuroscience. 9. 3072-3082.
Hiroi, N., Wong, M. L., Licinio, J., Park. C.. Young. M.. Gold, P. w..Chrousos, G. P.. & Bornstein, S. R.
(200 I). Expression of corticotropin releasing hormone receptors type I and type II mRNA in suicide
victims and controls. Molecular Psychiatry. 6. 540-546.
Holsboer, E, Lauer, C. 1.. Schreiber. w.. & Krieg. J. C. (1995). Altered hypothalamic-pituitary-
adrenocortical regulation in healthy subjects at high familial risk for affective disorders.
Neuroendocrinology. 62. 340-347.
Holsboer, E. Liebl, R.. & Hofschuster, E. (1988). Repeated dexamethasone suppression test during
depressive illness. Journal ofAffective Disorders. 4. 93-10 I.
Holsboer-Trachsler, E.. Stohler. R.. & Hatzinger. M. (1991). Repeated administration of the combined
dexamethasone-human corticotropin releasing hormone stimulation test during treatment of depression.
Psychiatry Research. 38. 163-171.
Huot, R. L., Plotsky, P. M.. Lenox. R. H.. & McNamara. R. K. (2002). Neonatal maternal separation
reduces hippocampal mossy fibre density in adult Long Evans rats. Brain Research. 950, 52-63.
Huot, R. L.. Thrivikraman, K. V.. Meaney. M. 1.. & Plotsky, P. M. (2001). Development of adult ethanol
preference and anxiety as a consequence of neonatal maternal separation in Long Evans rats and
reversal with antidepressant treatment. Psychopharmacology. 158. 366-373.
Jones. I., Kent, L., & Craddock. N. (2002). Genetics of affective disorders. In P. McGutlin, M. J. Owen &
I. I. Gottesman (Eds.), Psychiatric genetics and genomics (pp, 211-246). Oxford: Oxford University
Press.
Kang. H. K., Natelson, B. H.. Mahan. C. M.. Lee. K. Y.. & Murphy, E M. (2003). Post-traumatic stress
disorder and chronic fatigue syndrome-like illness among Gulf War veterans: A population-br.sed
survey of 30.000 veterans. American Journal of Epidemiology, 157. 141-148.
Kellner. M.. Baker, D. G.. & Yehuda, R. (1997). Salivary cortisol in Operation Desert Storm returnees.
Biological Psychiatry. 42. 849-850.
Kendler, K. S.• Karkowski, L. M.. & Prescott, C. A. (1999). Causal relationship between stressful life
events and the onset of major depression. American Journal of Psvchiatry, 156. 837-841.
Kendler, K. S., Kuhn. 1.. & Prescott. C. A. (2004). The interrelationship of neuroticism. sex. and stressful
life events in the prediction of episodes of major depression. American Journal ofPsychiatry, 161.631-
636.
Kessler. R. c., Berglund. P.. Demler. 0 .. Jin, R.. Koretz, D., Merikangas, K. R., Rush, A. 1.. Walters. E. E..
& Wang. P. S. (2003). National Comorbidity Survey Replication. The epidemiology of major depressive
disorder: Results from the National Comorbidity Survey Replication (NCS-R). Journal of the American
Medical Association, 289, 3095-3105.
Kishimoto. T.. Radulovic, 1.. Radulovic. M.. Lin. C. R.. Schrick. C.. Hooshmand, E, Hermanson, 0 ..

248
 Rosenfeld, M. G., & Spiess, J. (2000). Deletion of crhr2 reveals an anxiolytic role for corticotropin-
releasing hormone receptor-2. Nature Genetics, 24, 415-419.
Krieg, 1. c., Lauer, C. J., Schreiber, w., Modell, S., & Holsboer, E (2001). Neuroendocrine,
polysomnographic and psychometric observations in healthy subjects at high familial risk for affective
disorders: The current state of the 'Munich vulnerability study'. Journal (if Affective Disorders, 62,
33-37.
Kronfol, Z. (2002). Immune dysregulation in major depression: A critical review of existing evidence.
International Journal (if Neuropsychopharmacology, 5, 333-343.
Ladd, C. 0., Owens, M. J., & Nemeroff, C. B. (\996). Persistent changes in corticotropin-releasing factor
neuronal systems induced by maternal deprivation. Endocrinology, 137, 1212-1218.
Lopez, 1. E, Chalmers, D. T., Little, K. Y, & Watson, S. J. (1998). A. E. Bennett Research Award.
Regulation of serotonin I A, glucocorticoid, and mineralocorticoid receptor in rat and human
hippocampus: implications for the neurobiology of depression. Biological Psychiatry, 43, 547-573.
Lyons, D. M., Martel, E L.. Levine, S., Risch, N. J., & Schatzberg, A. E (\999). Postnatal experiences and
genetic effects on squirrel monkey social affinities and emotional distress. Hormones and Behavior, 36,
266-275.
Makino, S., Hashimoto, K., & Gold, P. W. (2002). Multiple feedback mechanisms activating corticotropin-
releasing hormone system in the brain during stress. Pharmacology, Biochemistry and Behavior. 73.
147-158.
Makino, S.. Shibasaki, T., Yamauchi, N., Nishioka, T., Mimoto, T., Wakabayashi, I., Gold, P. W.. &
Hashimoto, K. (1999). Psychological stress increased corticotropin-releasing hormone mRNA and
content in the central nucleus of the amygdala but not in the hypothalamic paraventricular nucleus in
the rat. Brain Research, 850, 136-143.
McCauley, J., Kern, D. E.. Kolodner, K., Dill, L., Schroeder, A. E, DeChant, H. K., Ryden, 1., Derogatis,
L. R., & Bass, E. B. (\997). Clinical characteristics of women with a history of childhood abuse:
Unhealed wounds. Journal (If the American Medical Association, 277, 1362-1368.
McClure, D. J. (1966). The diurnal variation of plasma cortisol levels in depression. Journal (!f
Psychosomatic Research, 10, 189-195.
McQueen, G. M., Campbell, S., McEwen, 8. S., Macdonald, K., Amano, S.. Joffe, R. T., Nahmias, C.. &
Young, L. T. (2003). Course of illness, hippocampal function, and hippocampal volume in major
depression. Proceedings of the National Academy of Sciences of the United States of America. 100.
1387-1392.
Mendels, J., & Frazer, A. (1974). Brain biogenic amine depletion and mood. Archives (if General
Psychiatry, 30, 447-451.
Merali, Z.. Du, L., Hrdina, P.. Palkovits, M., Faludi, G., Poulter, M. 0., & Anisrnan, H. (2004). Dysregula-
tion in the suicide brain: mRNA expression of corticotropin-releasing hormone receptors and
GABA(A) receptor subunits in frontal cortical brain region. Journal of Neuroscience, 24, 1478-1485.
Meyers, B. S.. Alexopoulos. G. S., Kakuma, T., Tirumalasetti, E, Gabriele, M., Alpert, S.. Bowden. C.. &
Meltzer, H. Y. (\ 999). Decreased dopamine beta-hydroxylase activity in unipolar geriatric delusional
depression. Biological Psvchiatrv, 45, 448-452.
Michelson, D.. Stratakis, C.. Hill. L., Reynolds, 1.. Galliven, E., Chrousos, G.. & Gold, P. (\996). Bone
mineral density in women with depression. New England Journal (if Medicine, 335, 1176-1181.
Modell, S., Lauer, C. J.. Schreiber, W.. Huber, J., Krieg, J. c., & Holsboer, E (1998). Hormonal response
pattern in the combined DEX-CRF test is stable over time in subjects at high familial risk for affective
disorders. Neuropsychopharmacologv. 18.253-262.
Musselman, D. L., Miller, A. H.. Porter, M. R.. Manatunga, A., Gao, E, Penna, S.. Pearce, 8. D.. Landry,
1.. Glover, S., McDaniel. J. S., & Nemeroff, C. B. (2001). Patients with depression: Preliminary
findings. American Journal of Psychiatry, 158, 1252-1257.
Nemeroff, C. 8. (1992). New vistas in neuropeptide research in neuropsychiatry: Focus on corticotropin-
releasing factor. Neuropsychopharmacology, 6, 69-75.
Nemeroff. C. B. (2003). The neurobiological consequences of childhood abuse. Presentation at the 2003
meeting of the American Psychiatric Association, San Francisco, CA. May 17-22.2003.
Nemeroff, C. B.. Krishnan. K. R.• Reed. D.. Leder, R.. Beam, c., & Dunnick, N. R. (1992). Adrenal gland
enlargement in major depression: A computed tomographic study. Archives of General Psychiatry, 49,
384-387.
Nemeroff, C. 8., Owens, M. 1.. Bissette, G.. Andorn, A. C; & Stanley, M. (\988). Reduced corticotropin
releasing factor binding sites in the frontal cortex of suicide victims. Archives of General Psvchiatrv,

249
 45.577-579.
Nemeroff. C. B.. Widerlov. E.. Bissette. G.. Walleus. H.. Karlsson, I.. Eklund. K.. et al. (1984). Elevated
concentrations of CSF corticotropin-releasing factor-like immunoreactivity in depressed patients.
Science. 226. 1342-1344.
Newport. D. J.. Heim, C.. Owens. M. J.• Ritchie. 1. C.. Ramsey. C. H., Bonsall, R.. et al. (2003).
Cerebrospinal fluid corticotropin-releasing factor (CRF) and vasopressin concentrations predict
pituitary response in the CRF stimulation test: A multiple regression analysis. Neuropsychopharmaco-
logy, 28. 569-576.
Nikolarakis, K. E.• Almeida. O. F.. & Herz, A. (1986). Corticotropin-releasing factor (CRF) inhibits
gonadotropin-releasing hormone (GnRH) release from superfused rat hypothalami in vitro. Brain
Research. 377, 388-390.
Orth, D. N. (\992). Corticotropin-releasing hormone in humans. Endocrine Reviews, /3.164-191.
Oshima. A., Yamashita. S.. Owashi, T.. Murata. T.. Tadokoro, C.. Miyaoka, H., Kamijima, K., & Higuchi,
T. (2000). The differential ACTH responses to combined dexamethasone/CRF administration in major
depressive and dysthymic disorders. Journal ofPsychiatric Research, 34. 325-328.
Pacak. K.. Tjurmina. 0 .. Palkovits. M.. Goldstein. D. S.. Koch. C. A.. Hoff. T., & Chrousos, G. P. (2002).
Chronic hypercortisolemia inhibits dopamine synthesis and turnover in the nucleus accumbens: An in
vivo microdialysis study. Neuroendocrinology, 76. 148-157.
Parker. A. J.. Wessely, S.. & Cleare. A. 1. (200 I). The neuroendocrinology of chronic fatigue syndrome and
fibromyalgia. Psychological Medicine. 31. 1331-1345.
Plotsky. P. M.. & Meaney, M. J. (\993). Early. postnatal experience alters hypothalamic corticotropin-
releasing factor (CRF) mRNA. median eminence CRF content and stress-induced release in adult rats.
Brain Research. Molecular Brain Research, /8. 195-200.
Posener, J. A.. Wang, L.. Price, 1. L.. Gado, M. H.. Province. M. A.. Miller, M. I.. Babb, C. 1\1., &
Csernansky. J. G. (2003). High-dimensional mapping of the hippocampus in depression. American
JOIIl'l1al (If Psychiatry. 160.83-89.
Pratt, L. A.. Ford. D. E.. Crurn, R. M.. Armenian. H. K.. Gallo. J. J.. & Eaton, W. W. (\996). Depression,
psychotropic medication. and risk of myocardial infarction. Prospective data from the Baltimore ECA
follow-up. Circulation. 94. 3123-3129.
Pryce. C. R.. Ruedi-Bettschen, D.. Dettling. A. c.. & Feldon, J. (2002). Early life stress: Long-term
physiological impact in rodents and primates. News in Physiological Sciences, /7, 150-155.
Raadsheer. F. c.. van Heerikhuize. J. J.. Lucassen, P. 1.. Hoogendijk, W. 1.. Tilders, F. J., & Swaab, D. F.
(1995). Corticotropin-releasing hormone mRNA levels in the paraventricular nucleus of patients with
Alzheimer's disease and depression. American Journal o( Psvchiatry, 152, 1372-1376.
Raza. 1.. Massoud. A. F.. Hindmarsh. P. C.. Robinson. I. C.. & Brook. C. G. (1998). Direct effects of
corticotrophin-releasing hormone on stimulated growth hormone secretion. Clinical Endocrinology.
48.217-222.
Reichardt. H. M.. Kaestner. K. H.. Tuckermann. 1.. Kretz. 0 .. Wessely, 0 .. Bock, R.. Gass, P.. Schrnic, w.,
Herrlich, P.•Angel. P.. & Schutz. G. (1998). DNA binding of the glucocorticoid receptor is not essential
for survival. Cell. 93. 531-541.
Ressler. K. J., & Nemeroff, C. B. (1999). Role of norepinephrine in the pathophysiology and treatment of
mood disorders. Biological Psychiatry, 46. 1219-1233.
Ribeiro. S. C.. Tandon, R., Grunhaus. L.. & Greden, J. F. (1993). The DST as a predictor of outcome in
depression: A meta-analysis. American Journal (It"Psychiatry, /50. 1618-1629.
Rinne. T., de Kloet. R.. Wouters. L.. Goekoop. 1. G.. DeRijk. R. H.. & van den Brink. W. (2002).
Hyperresponsiveness of hypothalamic-pituitary-adrenal axis to combined dexamethasone/cortico-
tropin-releasing hormone challenge in female borderline personality disorder subjects with a history of
sustained childhood abuse. Biological Psychiatry, 52, 1102-1112.
Risch. S. C.. Lewine. R. J.. Kalin. N. H.. Jewart. R. D.. Risby, E. D.. Caudle, J. M.. Stipetic, M., Turner, J.,
Eccard, M. B.. & Pollard, W.E. (\992). Limbic-hypothalamic-pituitary-adrenal axis activity and
ventricular-to-brain ratio studies in affective illness and schizophrenia. Neuropsvchopharmacologv; 6.
95-100.
Rivest. S.. Laflamme. N.. & Nappi. R. E. (\995). Immune challenge and immobilization stress induce
transcription of the gene encoding the CRF receptor in selective nuclei of the rat hypothalamus. Journal
of Neuroscience, 15. 2680-2695.
Sachar, E. 1.. Hellman. L.. Fukushima. D. K.. & Gallagher. T. F. (1970). Cortisol production in depressive
illness. A clinical and biochemical clarification. Archives of General Psvchiatrv, 23, 289-298.

250
Sanchez. M. M.. Ladd, C. 0 .. & Plotsky, P. M. (2001). Early adverse experience as a developmental risk
factor for later psychopathology: Evidence from rodent and primate models. Developmental
Psychopathology, 13,419-449.
Sapolsky. R. M. (1990). Glucocorticoids, hippocampal damage and the glutamatergic synapse. Progress in
Brain Research, 86, 13-23.
Schweiger. U.. Weber, B.. Deuschle. M.. & Heuser, I. (2000). Lumbar bone mineral density in patients with
major depression: Evidence of increased bone loss at follow-up. American Journal of Psychiatrv. /57.
118-120.
Scott, L. V.. Teh, J., Reznek. R.. Martin, A.. Sohaib, A.. & Dinan, T. G. (1999). Small adrenal glands in
chronic fatigue syndrome: A preliminary computer tomography study. Psychoneurorndocrinology, 24,
759-768.
Seasholtz. A. E, Burrows, H. L., Karolyi, I. J.. & Camper, S. A. (2001). Mouse models of altered CRH-
binding protein expression. Peptides. 22, 743-751.
Smith. M. A.. Davidson, J., Ritchie, J. C., Kudler, H., Lipper, S., Chappell, P.. & Nemeroff, C. B. (1989).
The corticotropin-releasing hormone test in patients with posttraumatic stress disorder. Biological
Psychiatry, 26. 349-355.
Stein. M. B.. Yehuda, R.. Koverola, C; & Hanna, C. (1997). Enhanced dexamethasone suppression of
plasma cortisol in adult women traumatized by childhood sexual abuse. Biological Psychiatry. 42,680-
686.
Swanson. L. W., Sawchenko, P. E., Rivier, J., & Vale, W. W. (1983). Organization of ovine corticotropin-
releasing factor immunoreactive cells and fibers in the rat brain: An immunohistochemical study.
Neuroendocrinology, 36, 165-186.
Swanson, L. w., & Simmons, D. M. (1989). Differential steroid hormone and neural intluences on peptide
mRNA levels in CRF cells of the paraventricular nucleus: A hybridization histochemical study in the
rat. Journal ofComparative Neurology, 285, 413-435.
Tennant, C. (2002). Life events. stress and depression: A review of recent findings. Australian and New
Zealand Journal of Psychiatry, 36, 173-182.
Thakore, J. H., Richards, P. J.. Reznek, R. H.. Martin, A., & Dinan, T. G. (1997). Increased intra-abdominal
fat deposition in patients with major depressive illness as measured by computed tomography.
Biological Psychiatry, 41, 1140-1142.
Timpl. P.. Spanagel, R.. Sillaber, I., Kresse. A.. Reul, J. M. H. M., Stalla, G. K.. Blanquet, V.. Steckler. T.,
Holsboer, F.. & Wurst. W. (1998). Impaired stress response and reduced anxiety in mice lacking a
functional corticotropin-releasing hormone receptor I. Nature Genetics, /9, 162-166.
Tronche, F.. Kellendonk, C; Kretz. 0 .. Gass, P.. Anlag, K., Orban, P. C, Bock, R., Klein, R., & Schutz, G.
(1999). Disruption of the glucocorticoid receptor gene in the nervous system results in reduced anxiety.
Nature Genetics. 23. 99-103.
Vaillant. G. E. (1998). Natural history of male psychological health. XIV: Relationship of mood disorder
vulnerability to physical health. American Journal "f Psychiatry, /55. 184-191.
Vakili, K.. Pillay, S. S.. Later. B., Fava. M.. Renshaw. P. E, Bonello-Cintron. C. M.• & Yurgelun-Todd, D.
A. (2000). Hippocampal volume in primary unipolar major depression: A magnetic resonance imaging
study. Biological Psychiatrv, 47, 1087-1090.
Vale. w.. Spiess. J., Rivier, C; & Rivier, J. (1981). Characterization of a 4 l-residue ovine hypothalamic
peptide that stimulates secretion of corticotropin and beta-endorphin. Science, 2/3, 1394-1397.
Vazquez. D. M. (1998). Stress and the developing Iimbie-hypothalamic-pituitary-adrenal axis.
Psvchoneurorndocrinology; 23, 663-700.
Villafuerte. S. M.. Del-Favero. J., Adolfsson, R.. Souery, D., Massat, I., Mendlewicz, J., Van Broeckhoven,
C.. & Claes, S. (2002). Gene-based SNP genetic association study of the corticotropin-releasing
hormone receptor-2 (CRFR2) in major depression. American Journal of Medical Genetics, /14, 222-
226.
Vythilingam, M., Heim, C .. Newport, J., Miller, A. H., Anderson, E., Bronen, R., Brummer, M., Staib, L.,
Vermetten, E.. Charney, D. S.. Nemeroff, C. B., & Bremner, 1. D. (2002). Childhood trauma associated
with smaller hippocampal volume in women with major depression. American Journal l!f Psychiatry,
/59,2072-2080.
Watson, S.. Gallagher, P.. Ritchie, J. C, Ferrier, I. N., & Young, A. H. (2004). Hypothalamic-pituitary-
adrenal axis function in patients with bipolar disorder. British Journal of Psychiatry, /84, 496-502.
Yehuda, R. (2001). Biology of posttraumatic stress disorder. Journal of Clinical Psychiatry, 62 (Suppl /7),
41-46.

251
Yehuda, R., Boisoneau, D., Lowy, M. T., & Giller, E. L., Jr. (1995). Dose-response changes in plasma
cortisol and lymphocyte glucocorticoid receptors following dexamethasone administration in combat
veterans with and without posttraumatic stress disorder. Archives of General Psychiatry, 52, 583 ..593.
Yehuda, R., Boisoneau, D., Mason, J. W, & Giller, E. L. (1993). Glucocorticoid receptor number and
cortisol excretion in mood, anxiety, and psychotic disorders. Biological Psychiatry, 34, 18-25.
Zobel, A. W, Nickel, T., Klinzel, H. E., Ackl, N., Sonntag, A., Ising, M., & Holsboer, F. (2000). Effe~ts of
the high-affinity corticotropin-releasing hormone receptor I antagonist R 121919 in major depression:
The first 20 patients treated. Journal (if Psychiatry Research, 34, 171-181.
Zobel, A. W., Yassouridis, A., Friesboer, R. M., & Holsboer, F. (1999). Prediction of medium-term outcome
by cortisol response to the combined Dexamethasone-CRF test in patients with remitted depression.
American Journal of Psychiatry, 156, 949-951.

252
 Epilogue

Towards Integration in the Theory and Treatment

of Depression?

The Time is Now

Patrick Luyten, Sidney J. Blatt, & Jozef Corveleyn

The chapters in this book represent a wide diversity of approaches to research on

depression, including epidemiological, cognitive-behavioral, psychodynamic,
developmental psychopathological, and neurobiological approaches. Some readers
may wonder whether there is any common ground between these approaches. We
believe and hope to demonstrate that there is.
In what follows, we will first discuss what we believe to be five major common
themes that come to the fore in many, if not most, chapters of this volume. These
are: (I) a general dissatisfaction with the DSM approach of depression, (2) a focus
on personality dimensions and on mental representations or cognitive-affective
structures, (3) the importance of interpersonal and (4) environmental factors, and
finally (5) the need for a developmental perspective on depression. One of the
primary goals of this discussion is to foster a dialogue that attempts to stimulate
integration among various theoretical and methodological approaches to research
on depression. Along the way, we will also propose an etiologically-based,
transdiagnostic dynamic interactionism model as an alternative for the DSM view
of depression. In this model, we will emphasize the centrality of maladaptive
mental representations of self and other in both the development as well as the
treatment of depression and other disorders. Although our focus will primarily be
on psychodynamic, cognitive-behavioral and neurobiological approaches, we will
attempt to integrate aspects from other approaches as well, including psychiatric
genetics, developmental psychopathology, social cognition, and attachment theory.
Finally, we will consider future research tasks and the clinical implications that
follow from the proposed dynamic interactionism model.

Avenues For Further Integration

Dissatisfaction with the DSM: Plea for a Transdiagnostic Perspective

Dissatisfaction with DSM. Most frequently emphasized in the various chapters

and perhaps the most important finding of this book, is the general dissatisfaction
with the categorical view of depression in the DSM. Time and again, research from
perspectives as diverse as epidemiology, cognitive-behavioral and psychodynamic
models of depression, developmental psychopathology, and neurobiology, point to
serious problems with the DSM taxonomic approach. These issues with DSM are
of central concern because the development of a valid taxonomic system is central
to all research on psychopathology, including depression.
First, DSM distinguishes several categories of depression (e.g., major depression
versus minor depression or dysthymia) mainly on the basis of differences in
severity. Research, however, shows that a dimensional view, which conceptualizes
depression as situated on a continuum from mild dysphoria to full-blown clinical
depression, is more valid and useful than the DSM categorical view (see
Demyttenaere, Van Oudenhove, & De Fruyt, Chapter I, this volume). DSM appears
not to "carve nature at its joints" but rather introduces sharp distinctions where they
do not, in reality, exist (Kendler & Gardner, 1998; Solomon, Haaga, & Arnow,
200 I). This reliance on a weak categorical system seriously impedes all research
efforts. One of the many negative consequences of the creation of these "pseudo-
entities" (Parker, 1999, p. 102) is the underestimation of the importance of
"subclinical" depressive symptoms and syndromes (such as minor depression,
subthreshold depression) (Kessler et aI., 2003). Recent research has clearly
demonstrated both the clinical importance of such subclinical mood disturbances in
predicting later psychopathology and severe psychosocial impairment (e.g., Kessler
et aI., 2003; Judd, Akiskal, & Paulus, 1997), as well as identifying the continuity
between biopsychosocial factors implied in both "subclinical" and "clinical" mood
disturbances (e.g., Ormel, Oldehinkel, & Brilman, 2001).
Second, under the influence of the DSM, depression has been increasingly
considered as a disease with a specific etiology and pathogenesis, and as a state
which is clearly distinct from "normality", from normal sadness and from other

254
psychiatric disorders. This has had serious unfortunate consequences. Research has
been overly focused on trying to establish the "unique" etiology and characteristics
of depression (see also Van Praag, de Kloet, & van as, 2004). From a treatment
perspective, psychiatrists and clinical psychologists have started treating disorders
and their symptoms instead of individuals. This is further amplified by the DSM
assumption that Axis I disorders and personality characteristics are completely
independent because DSM assumes that Axis I and Axis II are orthogonal and thus
independent. This view has led to an increasingly confusing literature on the
comorbidity of depression (and other Axis I disorders) and Axis II disorders. If
there is one conclusion that can be drawn about this literature, apart from the fact
that Axis II itself suffers from serious validity problems (Shedler & Westen, 2004),
is that depression and personality (disorders) are closely associated and even
causally linked. Thus, the DSM view that depression and personality are orthogonal
is clearly incorrect. Moreover, apart from this issue of the relation of personality
disorders to depression, the DSM approach has contributed to investigators
underestimating the importance of subclinical personality problems (e.g., enduring
problems concerning intimacy and relatedness), which are present in many if not
most depressed patients (Blatt, 2004; Morrison, Bradley, & Westen, 2003). As
reviewed in Chapter 4, research has for example clearly identified the causal
relationship between the personality dimensions of Dependency/Sociotropy and
Self-Critical Perfectionism/Autonomy and depression. These cognitive-affective
styles appear to be relatively unresponsive to brief treatments, and probably require
more intensive treatment (Blatt, 2004; see also Chapter 4, this volume).
DSM has also promoted the view that individuals with Axis I disorders such as
depression are "hosts" for a certain "pathogen". Yet, as Westen and ShedIer (2000)
have pointed out, while this disease analogy might be appropriate for some forms
of psychopathology (e.g., Posttraumatic Stress Disorder after severe trauma), for
many difficulties it is clearly not. Congruent with research findings reviewed in this
book this "analogy quickly breaks down because 'host' and 'pathogen' are not so
neatly distinguishable in the psychological realm. Humans often seek, evoke, or
elicit the environmental pathogens to which they are then exposed" (Westen &
Shedler, 2000, p. 113; see also Blatt & Zuroff, 1992). In sum, DSM has introduced
an arbitrary and unproductive distinction between axis I and II, thereby neglecting
the important role of personality factors in psychopathology. Both in diagnosis and
in treatment, much is to be gained by a person rather than a symptom focus - that
it is important to evaluate the life experiences and the meaning structures that
determine how people experience their world, both in normal as well as in disrupted
functioning.
A person focus is also consistent with the conceptualization of depression in this
volume as a deviation of "normal" development, which can vary in intensity (e.g.,
see Chapters 3, 5, 7, and 8 in this volume). Depression, according to this view, is a
more or less "normal" phenomenon that is intimately tied to the human condition
and not so much a disease that is present in those unfortunate individuals who have
a particular, although as yet unknown, (biological) vulnerability. The ubiquity of
depression has been demonstrated, for example, by Monroe and Simons (1991).
These investigators have shown that while the risk of becoming clinically depressed

255
after one severe negative life event is relatively small, the likelihood of becoming
depressed rises considerably with each new negative life event until, under
increasing stress, almost everyone becomes clinically depressed. In fact, perhaps
with the exception of some extremely resilient individuals (Bonnano, 2004), after
experiencing three or more severe negative life events in a one year period almost
everyone becomes depressed (Monroe & Simons, 1991). Findings such as these
seriously challenge the view that depression is a disease that is only present in some
unfortunate few. Instead, these findings favor the view that depression is a disorder
that is intrinsically associated with human development and life experiences (see
also Chapters 3 and 7, this volume).
Third, the DSM promotes a disease model of depression that defines depression
by a number of symptoms, and the outcome of treatment with an almost exclusive
focus on symptom reduction. It has become increasingly clear, however, that
treatments that focus on symptom reduction alone are insufficient. On average, two
thirds of patients relapse after brief, symptom-focused treatments of depression
(Westen & Morrison, 200 I; Westen, Novotny, & Thompson-Brenner, 2004a; Zuroff
& Blatt, 2002). Moreover, as we will discuss in detail below, although depressive
symptoms may quickly disappear in symptom focused treatments, underlying
vulnerabilities, including impaired and disrupted interpersonal relationships, are
likely to persist. Hence, treatment needs to focus on underlying vulnerability
factors to prevent future relapse, a theme that is echoed in many chapters in this
volume.
Fourth, one of the major limitations of DSM is its focus on manifest
symptomatology to define psychopathology. This is particularly true concerning the
diagnosis of major depression, the main depression category in DSM, which can be
diagnosed based on almost any combination of a number of symptoms. Because of
this a-etiological view, individuals who receive a DSM diagnosis such as major
depression are a very heterogeneous group in terms of etiology and pathogenesis
(Parker, 2000). This is detrimental for both research as well as clinical practice (Van
Praag et al., 2004). For example, often a world of difference exists between
someone who becomes depressed after losing a partner in a car accident, versus an
overly dependent individual that becomes depressed after his or her partner ends
their relationship after twenty long years of relational conflicts, struggles, and
frustrations. Although both would receive a DSM diagnosis major depressive
disorder, first episode, it is very likely that the course and prognosis in both
individuals will be markedly different. Moreover, congruent with findings that
depression, personality, and environmental factors reciprocally influence one
another, it is also important to take into account the active role of individuals in
creating their own "depressogenic" environment, as well as both intrapersonal and
interpersonal changes as a consequence of being depressed (see Chapter 4 and
Chapter 5, this volume).
An alternative: An etiologically-based, dynamic interactionism approach. To
summarize, although the DSM approach to mood disorders has led to important
insights and discoveries, it is clearly limited. Despite much dissatisfaction with this
approach in clinicians and researchers, the DSM approach continues to dominate
both research and treatment guidelines. However, the time appears ripe for

256
developing a more valid, etiologically-based diagnostic system (Blatt & Levy,
1998) that takes into account the dynamic interactionism between person,
environment, and depression (Blatt, 2004; Kendler, Kuhn, & Prescott, 2004; Zuroff,
Mongrain, & Santor, 2004), and that considers depression as a disorder that is
linked to normal development (Blatt & Levy, 1998). Ideally, such a diagnostic
system would consist of several components or axes:
(I) We believe that assessment should start with a detailed descriptive diagnosis
that documents depressive symptoms as well as their severity. This enables the
clinician to appreciate the nature of a patient's complaints, as well as their influence
on daily functioning. Moreover, a descriptive diagnosis should also contain
important leads concerning etiological and pathogenetic factors. Although
historically many unsuccessful attempts have been made to identify distinct
depressive subtypes based on symptoms, future research might identify distinct
clusters of symptoms that delineate specific depressive disorders. For example,
there is increasing evidence for the existence of a melancholic subtype of
depression, which is mainly characterized by pronounced psychomotor retardation
(Parker, 2000; Pier, Hulstijn, & Sabbe, 2004). Likewise, anxious and hostile or
irritable subgroups of non-melancholic depression may exist, which are
presumably linked to temperament and personality features (Parker et al., 1998,
I 999a, I 999b ) such as Dependency/Sociotropy versus Self-Critical Perfectionism!
Autonomy respectively. Next, future studies should investigate whether these
subtypes, if they exist, are associated with a different course, prognosis, etiology,
pathogenesis, and treatment responsivity. For instance, evidence suggests that the
melancholic subtype of depression (and psychomotor retardation in particular) is
associated with a positive response to antidepressant treatment (Pier et al., 2004)
and electroconvulsive therapy (Parker, 2000). In addition, both clinicians and
researchers as well as insurance companies might find it often more convenient to
speak in terms of categories (see also Chapter 1, this volume). Furthermore, the fact
that depression generally is situated on a continuum does not necessarily exclude
the existence of some discrete categories of depression (such as the melancholic
subtype) (see also Parker, 2000). The (methodological) challenge for future
research might lie in integrating continuous/dimensional and categorical
viewpoints with the hope that this would improve the validity of diagnostic
distinctions.
(2) The next step in diagnosis should consist of a consideration of depression
from a dimensional perspective. From a clinical point of view, many patients
present with depressive problems that warrant clinical attention, but cannot be
adequately diagnosed by DSM. Many of these patients receive diagnoses such as
"minor depression, "subclinical depression", "subthreshold depression", or
"depression not otherwise specified". As reviewed in Chapter 1, no real "cut-off'
exists between these "subclinical" syndromes and clinical depression. While some
distinct symptom-defined subtypes of depression, qualitatively different from each
other and/or from "subclinical" depression or dysphoria may exist, as stated before,
depression appears to be best situated on a continuum going from mild dysphoria
to full-blown clinical depression (Blatt, 1974, 2004). In addition, as stated earlier,
the biopsychosocial processes implied in the etiology and pathogenesis of "normal"

257
(e.g., normal sadness), "subclinical" (often referred to as distress) and "clinical"
depression are essentially the same. Hence it is important to assess mood and mood
fluctuations across the life span. For example, when did the first signs of mood
change appear? Are we dealing with a person who has suffered from continuous
bouts of depression? Or are depressive moods always associated with severe events,
and, if so, what is the nature of these events? Consideration of such issues almost
automatically leads to a third component in the process of diagnosing - the factors
involved in the etiology and pathogenesis.
(3) The next step is thus to consider factors involved in the etiology and
pathogenesis. It is very unlikely that clinical symptoms alone are sufficient to
identify etiological and pathogenetical factors implied in subclinical and clinical
depression (Blatt, 2004; Parker, 2000). And it is likely that several etiological and
pathogenetic pathways exist towards subclinical and clinical depression. As various
chapters in this book emphasize, careful attention should be given to environmental
factors such as parental style, early life stress, life events, chronic stress, and daily
hassles, and cognitive-affective schemas or personality styles such as
Dependency/Sociotropy and Self-Critical Perfectionism/Autonomy. In addition, the
interaction between these factors in the developmental history of the individual
should be carefully mapped (see Chapter 4 and Chapter 5, this volume; Ormel et
aI., 200 I). Furthermore, apart from the identification of vulnerabilities, strengths or
resilience should be taken into account, as they provide important clues with regard
to both biological and psychosocial processes implied in depression (e.g., Bonnano,
2004; Charney, 2004). Moreover, for too long, diagnosis has been exclusively
aimed at documenting deficits, and neglecting strengths and possibilities.
These considerations concerning etiology and pathogenesis are likely to lead to
a more relativistic view of diagnosis because the psychosocial (e.g., Chapters 3 and
6, this volume) and biological (e.g., Chapter 6, this volume) processes implied in
depression are likely to be related to many other disorders. Hence, we believe that
a transdiagnostic perspective should be adopted in diagnosing different forms of
psychopathology including depression. In contrast to DSM, research findings
suggest that it is very unlikely that each psychopathological disorder has its unique
and specific etiology and pathogenesis. The same holds for what is now currently
considered as subthreshold psychiatric conditions (e.g., Lewinsohn, Shankman,
Gau, & Klein, 2004). For example, the effects of early life stress on HPA axis
functioning are likely to playa role in several psychopathological (e.g., depression,
PTSD), functional (e.g., Chronic Fatigue Syndrome), as well as medical disorders
(see Chapter 8, this volume; see also Segerstrom & Miller, 2004; Van Houdenhove
& Egle, 2004). Likewise, there appear to be few life events that specifically predict
depression (Paykel, 2003). And finally, personality factors such as Dependency/
Sociotropy and Self-Critical Perfectionism/Autonomy have been shown to playa
role in various other disorders than depression (e.g., Blatt, 2004; Blatt & Shichman,
1983; Shafran & Mansell, 2001), and may explain the high comorbidity between
various Axis I and Axis II disorders (Bieling, Summerfeldt, Israeli, & Antony, 2004;
Blatt & Levy, 1998).
Hence, research and assessment should concentrate on charting the various
biopsychosocial processes implied in normal development, and in deviations from

258
normal development, and how these deviations can be expressed in various (clusters
of) disorders (Blatt, 1995; Blatt & Levy, 1998; Blatt & Shichman, 1983). Here, the
developmental psychopathological principles of equifinality and multifinality may
assist in clarifying this argument (Cicchetti & Cohen, 1995). The principle of
equifinality, rather than assuming that one single etiology exists for a disorder, as is
currently implied by DSM, holds that various etiological pathways are possible
towards one disorder. In contrast, the principle of multifinality refers to the fact that
the same etiological factors may result, depending on other influences, in a variety
of disorders. Thus, various etiological pathways may exist towards depression, as
suggested by recent psychodynamic and cognitive-behavioral formulations (see
Chapter 3, this volume), whereas the same etiological factors may results in various
disorders, as suggested by neurobiological research on stress and the HPA axis (see
Chapter 8, this volume). Hence, patients with the same clinical presentation may
have a very different etiological background and patients with similar etiological
backgrounds, depending on other factors, may express their problems in different
ways. Thus, assessment should not be solely aimed at diagnosing a particular
disorder at a particular moment in time, but should include a developmental
assessment of the underlying vulnerabilities as well as strengths of an individual,
which should then inform treatment options for that individuaL. Much can be
gained, in our view, from this person, rather than symptom or disorder-focused
approach.
For example, early in adolescence, because of emotional neglect and abuse, a
young boy may start to experience problems in relating to peers, teachers, and
parents. As is often the case, he may express these difficulties primarily in school
and conduct problems. Later on, he may start experimenting with drugs to
compensate for his feelings of emptiness. Then, faced with the realities of adult life,
he may start to show the first signs of a mixed anxious/depressed condition, which
is subdued when he believes he has met the girl of his life. However, because he is
overly dependent on her, he is very jealous and clinging, and she ends the
relationship, and he becomes severely depressed. From a DSM perspective, this
individual would have had several clinical and subclinical diagnoses. Although
perhaps correct from a descriptive perspective, it would be incorrect to assume that
he has several "diseases" (e.g., conduct disorder, substance abuse disorder,
dysthymia), each associated with a particular vulnerability, and each needing a
different treatment approach. Rather, at various points in life, his problems,
influenced by many psychosocial and biological variables, are expressed in
different ways (see also Harrington, 2001). At various points in life, his life
could have taken different paths (e.g., what if a teacher or an athletic coach had
taken a particular interest in him, or if he had met another girl, or moved to another
part of the country, or had found a job that gave him some satisfaction and
recognition?).
Ample evidence supports the principles of equifinality and multifinality in the
development of psychological disturbances (Cicchetti & Cohen, 1995). Concerning
depression, several studies have shown that etiological and pathogenetic factors
implied in depression are also found in a variety of other disorders (multifinality),
which is reflected in the high comorbidity between depression and other Axis I,

259
such as anxiety disorders (Nemeroff, 2002) and Axis II disorders. On the other
hand, there are several etiological and pathogenetic pathways to depression
(equifinality) (e.g., Blatt, 2004; Claes, 2003; Gilmer & McKinney, 2003; Kasen et
al., 200 I; Klein, Lewinsohn, Seeley, & Rohde, 200 I; Lewinsohn, Rohde, Seeley,
Klein, & Gotlib, 2003; Weissman et aI., 1999a, 1999b).
(4) Finally, descriptive/categorical, dimensional and etiological/pathogenetic
considerations should be combined in making a systematic statement concerning
the likely course and prognosis of the disorder, as well as the treatment possibilities.
We will return to this issue further below.
Although such a transdiagnostic, multiaxial dynamic interactionist model of
diagnosing depression is not currently used in mainstream research, many impor-
tant aspects of this interactional model have been fairly clearly documented by
research over recent decades, as discussed in the various chapters in this volume.
For example, research has clearly shown how cognitive-affective styles such as
Dependency/Sociotropy and Self-Critical Perfectionism/Autonomy interact with
(congruent) life events in predicting depression (see Chapter 4 and 5, this volume).
Likewise, the role of early adversity, and its impact on both neurobiological and
psychosocial factors implied in depression, has now been clearly established (see
Chapter 8, this volume). Other pieces of the puzzle (e.g., how genetic factors
determine temperament and subsequent personality development, and how these
factors interact with specific parental styles) can be, at least in principle,
investigated (see also Van Praag et al., 2004). Moreover, in clinical practice,
regardless of theoretical orientation, many clinicians (implicitly) use such an
etiologically-based diagnostic model, as both common sense and research suggests
(e.g., Morrison et aI., 2003). Yet, as long as the DSM categorical disease view
continues to dominate depression research and treatment guidelines, and research
funding continues to depend largely on the adoption of this approach, there is little
reason to believe that research programs including treatment research will seek to
develop an alternative diagnostic system (Westen et al., 2004a).

Personality and Mental Representations or Cognitive-Affective Schemas

A second area of convergence in this volume concerns the emphasis on

personality and mental representations or cognitive-affective schemas and
depression. This appears to be a particularly fruitful area for dialogue and
integration among the various approaches to depression because the notion of
mental representation or cognitive-affective schema is a central construct not only
in psychodynamic object-relational and cognitive-behavioral formulations, but also
in cognitive science, social psychology, developmental psychopathology, and
attachment theory (Blatt, Auerbach, & Levy, 1997), and thus may provide a
common language for many investigators and clinicians.
Research from various theoretical perspectives has contributed to our knowledge

260
concerning how mental representations of self and others develop throughout life
from the early infant-caregiver relationship to senescence, and how they organize,
shape and guide our cognitions, affects, and behaviors (Blatt et aI., 1997). More-
over, our knowledge of both adaptive and maladaptive pathways in the development
of mental representations has increased substantial1y, particularly with the advent
of object-relational psychoanalytic theory, attachment theory, and developmental
psychopathology. It has now become clear that specific forms or clusters of
psychopathology are associated with specific disturbances in the content and/or
structure of mental representations (Blatt, 1995; Blatt et aI., 1997). For instance,
various psychodynamic (e.g., Blatt, 1974, 1995, 2004; Westen, 2002) and cogni-
tive-behavioral (Beck, 1999; Linehan, 1993; Young, 1999) investigators link
psychopathology with specific disturbances in the content and/or structure of
mental representations. In fact, it can be argued that all major psychological
theories of depression focus on aspects of mental representations as a major factor
in depression and a major factor in the treatment process (Blatt, 2004; Blatt,
Stayner, Auerbach, & Behrends, 1996). Most theorists, like Beck (1999), focus on
the content of these representations, while psychodynamic formulations often focus
more on the structural (cognitive or developmental) organizations of these
representations (e.g., Blatt, 1995; Blatt et al., 1996; Westen, 2002) because these
reflect more implicit or procedural aspects of these cognitive structures as well as
important developmental characteristics.
It is interesting to note that theoretical formulations have increasingly linked
these disturbances in mental representations with broader personality dimensions.
This is not only the case in psychodynamic formulations, but also in cognitive-
behavioral theory, as is exemplified by the work of Beck (1999), Linehan (1993),
and Young (1999). As Beck (1999) explains, considerations concerning personality
and personality structure once were considered to be typical of psychodynamic
formulations but cognitive-behavioral approaches have shifted from an exclusive
focus on the symptomatic expressions of depression to considerations concerning
personality and personality structure as vital to an understanding of psycho-
pathology (Beck, 1999). Despite symptomatic improvement, many patients, parti-
cularly those with personality disorders, relapse because of continuing vulnerability
associated with often highly resistant cognitive-affective schemas that are deeply
embedded in the personality of an individual. This turn towards personality in
cognitive-behavioral research has resulted primarily from the disappointing results
of "traditional" cognitive-behavior therapy with many patients, and the realization
that treatment should not be exclusively aimed at "surface" cognitions and
symptom relief, but at reducing underlying vulnerabilities such as rather stable and
relatively resistance cognitive-affective schemas to prevent relapse. Interestingly,
this change has also been accompanied by the growing importance of
developmental considerations within the cognitive-behavioral movement. Young
(1999), for example, has been a strong advocate of introducing the notion of "Early
Maladaptive Schernas" (EMS), i.e., maladaptive schemas that develop early in life,
into cognitive-behavioral theory.
Thus, the construct of mental representations may provide a bridge between
clinicians and investigators from various theoretical orientations. What these

261
various theoretical formulations concerning mental representations have in
common, is that they not only propose that representations of self and other are
relatively stable characteristics that organize and guide the individual's affects,
cognitions, and behaviors, but that treatment can be seen as a process aimed at
changing the content and/or structure of these representations (Bateman & Fonagy,
2004; Blatt et al., 1996; Blatt et al., 1997). An important question for further
research will be whether these mental representations or cognitive-affective sche-
mas can be changed (that is, become developmentally more mature and adaptive)
via cognitive suggestions that focus on the content of representations (e.g., to think
differently about yourself and others) or whether changes in these cognitive-
affective schemas can most effectively occur, with strength and stability, primarily
as they are experienced in an intense interpersonal relationship such as the thera-
peutic relationship, where one begins to relinquish prior maladaptive, inappro-
priate, and developmentally more primitive representations through experiencing
and becoming aware of their repetitive distortion of life experiences, and replacing
these impaired representations with new, more adaptive representations of self and
others as they are experienced in the intensity of the therapeutic relationship,
implying a focus on both content and structural characteristics (Blatt et al., 1996).
Furthermore, a change in mental representations can range from the micro-level
(e.g., targeting overgeneral autobiographical memories, see Chapter 2, this
volume), traditionally the level at which cognitive-behavior therapy is aimed, to the
macro-level (e.g., discussing broader issues concerning personality or cognitive-
affective schemas associated with Dependency/Sociotropy and Self-Critical
Perfectionism/Autonomy (Blatt, 1974; see Chapter 3, this volume), traditionally the
level on which psychodynamic psychotherapy focuses (Westen, 2000). Hence, the
concept of mental representation may link micro- and macro-processes in
psychopathology and foster further dialogue between treatment strategies that are
mainly directed at micro-processes (e.g., CBT) versus macro-processes (e.g.,
psychodynamic psychotherapy). In addition, because various theoretical formula-
tions concerning mental representations include a developmental perspective (e.g.,
cognitive developmental psychology, attachment theory, object-relational psycho-
dynamic formulations), these concepts may also foster integration with develop-
mentally based assessment and treatment (see also Chapter 7, this volume).
Research on mental representations may also provide a bridge between different
areas of research. In Chapter 2, Hermans, Raes and Eelen, for example, illustrate
how theoretical and methodological approaches in cognitive science may inform
clinical research on the role of (autobiographical) memory in depression. They
show how mental representations in depressed patients tend to be overgeneral and
how memory tends to be selective, resulting in a rapid onset and the eventual
deepening of depressed mood. Investigation of relations between these memory
disturbances in depression and broad cognitive-affective schemas associated with
Dependency/Sociotropy and Self-Critical Perfectionism/Autonomy (e.g., Blatt,
1974; Blatt, Wein, Chevron, & Quinlan, 1979; Moore & Blackburn, 1993; Nunn,
Mathews, & Trower, 1997), could foster the interchange between cognitive science
and clinical research.
Finally, an important issue for further research is the neurobiology of mental

262
representations. As Herman et al. suggest in Chapter 2, new techniques such as
brain imaging are likely to play a vital role in furthering our knowledge of
the relationship between the psychology and neurobiology of mental represen-
tations (Beutel, Stern, & Silbersweig, 2003; Goldapple et aI., 2004; Laakso et al.,
2003).

A Focus on the Interpersonal Context of Depression

Psychological formulations concerning depression often debate the question

whether depression has mainly intrapsychic versus interpersonal origins. It is clear
from the various contributions to this volume that this is unproductive. Both
theoretical formulations and empirical research demonstrate that intrapsychic and
interpersonal factors are intertwined (e.g., Joiner, 200 I; Shahar, Blatt, Zuroff,
Krupnick, & Sotsky, 2004; see also Chapter 3, this volume), and that both are often
two sides of the same coin. Again, a focus on mental representations may help
bridge the intrapersonal-interpersonal gap, because developmental research has
clearly shown how interactions with significant others, from early on, form the
basis of representations of self and other - the building blocks of relatively stable
cognitive-affective schemas or internal working models that organize and guide the
individual (Blatt, 1991, 1995; Blatt et al., 1997; Bowlby, 1988; Main, Kaplan, &
Cassidy, 1985; Stern, 1985).
Aside from these theoretical considerations, research reviewed in this volume
(e.g., see Chapter 4, 5 and 8, this volume) clearly shows the central importance of
interpersonal processes in the pathogenesis, etiology, maintenance, relapse, and
treatment of depression (see also Beach, 2001; Brown & Harris, 1978; Coyne,
Burchill, & Stiles, 1991; Joiner, Metalsky, Katz, & Beach, 1999; Segrin, 2000; Tse
& Bond, 2004). Depression is not just a "personal pathology", but also a "relational
pathology" (Hammen, 2003). Again, these findings caution against considering
depression as a disease that should be treated by a symptom-focused approach.
Moreover, the individual is not merely a host to a pathogen, to which he or she
passively responds. To the contrary, as reviewed in Chapter 4 and 5, specific
interpersonal styles not only play a role in the etiology and pathogenesis of
depression, but being depressed also leads to further relational problems (e.g.,
lowered social skills, contempt and rejection by others, etc.), leading to a vicious
cycle (Wachtel, 1997), so that after the depression has remitted, a "depressogenic"
interpersonal style, with its associated conflicts, may persist. Consider, for
example, an overly dependent woman who is admitted to hospital because of severe
depression after her husband left her. Although she may be feeling much better after
a couple of weeks, when she returns home she is not only faced with the fact that
she probably lost her husband, and therefore has to re-arrange her life completely,
but others in her environment have not forgotten how depressed she was and how
she, for instance, has threatened to kill herself in front of her children. Perhaps her

263
children are now afraid of her, and prefer to live with their father. In addition,
although she may be feeling better, her dependent, clinging relational style may
persist. Thus, often, the individual remains at considerable risk for relapse, parti-
cularly because such vicious cycles, resulting from a dysfunctional transactional
style (see Chapter 4, this volume), are easily activated because of the bias in the
encoding and perception typical of depressed patients. As Hermans et al. argue in
Chapter 2, for depressed individuals "everything seems to come back" very quickly,
in part because of their biased perception and underlying fears which they tend to
confirm by their interpersonal style (see Chapter 4, this volume).
Thus, an important area for future research is the integration of current formu-
lations concerning interpersonal processes in depression, including psychodynamic
(e.g., Blatt, 2004) and cognitive-behavioral (e.g., Beck, 1983; Safran & Segal,
1990), with approaches such as systems and family approaches (see Beach, 2001).
From a clinical perspective, the central importance of interpersonal processes in the
etiology and pathogenesis of depression suggests an urgent need to incorporate
interventions more explicitly aimed at relational processes in the treatment of
depression (Hammen, 2003) as expressed in Interpersonal Therapy (lPT>, a
treatment that focuses on contemporary interpersonal relationships, and, more
generally, in a relational emphasis in individual psychotherapy as expressed in
understanding transference enactments of repetitive interpersonal patterns. In
particular, the therapeutic relationship provides the basis for the patient to observe,
revise, and eventually relinquish maladaptive representations of self and other, and
begin to build and internalize more mature, integrated, and adaptive representations
of self and others (e.g., Blatt et aI., 1996). Despite growing evidence for the efficacy
of couple and family approaches in the treatment of depression (e.g., Denton,
Golden, & Walsh, 2003), this area is often neglected, particularly in existing
guidelines for the treatment of depressed patients (e.g., American Psychiatric
Association, 2000). Again, treatment guidelines appear to lag behind the findings
of empirical research. In this context, much may also be gained from the
experiential perspective which has long emphasized the importance of the
therapeutic alliance and interpersonal factors in treating depressed individuals. The
increasing dialogue between experiential, cognitive (Ottens & Hanna, 1998), and
psychodynamic (e.g., Greenberg & Watson, 1998) approaches is a step in the right
direction, but much more needs to be done in this effort.

Environmental Factors and the Nature versus Nurture Debate

Historically, research on depression has long been dominated by the question

whether there are two subtypes of depression, one more biological in origin (e.g.,
endogenous depression), the other one more environmentally determined (e.g.,
reactive or neurotic depression). This discussion reflects the familiar issue of nature
versus nurture. It has become clear, however, that the nature versus nurture debate

264
is unproductive, especially given the increasing evidence for gene-environment
correlations and interactions (Kendler, 2001; Rutter, 2002; Rutter et aI., 1997; Van
Praag et aI., 2004), and recursive interactions between biological and psychosocial
factors (Raison & Miller, 2003; Tsigos & Chrousos, 2002; Uchino, Cacioppo, &
Kiecolt-Glaser, 1996).
Gene-environment correlations in the context of depression refer to the notion
that genes may affect the exposure to certain environmental stressors (or vice versa)
associated with depression, while gene-environment interactions refer to the
synergistic interaction of genes and environment in predicting depression (e.g., a
certain gene may increase the sensitivity for life stressors). Again, the construct of
cognitive-affective schemas (or personality) may play a central role here, as
temperament and personality factors are currently viewed as the most likely crucial
factors that mediate the effect of genes on environment (e.g., Kendler &
Karkowski-Shuman, 1997; Saudino, Pedersen, Lichtenstein, McClearn, & Plomin,
1997). Broad personality traits such as neuroticism, for example, have shown to be
associated with the occurrence of dependent (fateful) life events, thus increasing the
risk of depression (e.g., Kendler, Gardner, & Prescott, 2003; Kendler et aI., 2004;
Kendler, Neale, Kessler, Heath, & Eaves, 1993; Ormel et aI., 200 I; Ormel, Stewart,
& Sanderman, 1989; Van as & Jones, 1999; Van as, Park, & Jones, 2001). Also,
evidence for gene-environment interactions suggests that certain genes may lead to
increased sensitivity to life stress associated with depression (e.g., Caspi et aI.,
2003).
Unfortunately, most research has not yet investigated recursive interactions
between genes, individuals, their environment, and depression, and has mainly
focused on broad personality traits such as neuroticism which may be less
informative. I As discussed in Chapter 3, theoretical formulations concerning gene-
environment correlations and interactions are congruent with our proposed
dynamic interactionism model. Future studies should adopt a full dynamic
interactionism perspective, focusing on more specific personality dimensions such
as Dependency/Sociotropy and Self-Critical Perfectionism!Autonomy and their
interactions with congruent events (see Chapter 4 and Chapter 5, this volume), and
try to take into account the complex recursive interactions between individuals,
their environment, and depression. These studies are likely to transform current
treatment strategies for depression because they clearly emphasize the importance
of the patient's (social) environment when treating depression (Zuroff & Blatt,
2002). Again, we believe that systems and family approaches may have much to
offer in this context as well as intensive, non-symptom focused, individual
psychotherapy in which the patient-therapist relationship is a foremost factor in the
therapeutic process.
I This research has also been complicated by the fact that although there is clear evidence for the role of

family transmission in depression. no genes have been definitely associated with depression (Van Praag et
aI., 2(04). In addition, although the human genome project may lead to the identitication of certain genes
associated with depression, we consider it unlikely that one particular gene or set of genes will be
associated with depression. Rather, congruent with our transdiagnostic dynamic interactionism view, we
expect that particular genes might be involved in the etiology of several disorders, particularly through the
effects they have on temperamental and personality factors (see also Kendler, 200 I and Chapter 8, this
volume).

265
 In addition, evidence has increasingly indicated the importance of recursive
interactions between psychosocial and neurobiological processes implied in various
disorders, including depression (Claes, 2003; Gold & Chrousos, 2003). As
reviewed in Chapter 8 by Claes and Nemeroff, research, particularly on the effects
of early adversity on the psychobiological reactions to stress, is rapidly increasing
our knowledge concerning interactions between neurobiological and psychosocial
factors. As Claes and Nemeroff point out, research on the HPA axis, in particular,
might playa vital role in bridging the gap between biological and psychological
approaches. Although Claes and Nemeroff indicate that this integrative research has
already begun, truly interdisciplinary and integrative research is still a rarity,
although several theoretical models have been proposed in this area (e.g., Kandel,
1999; Cloninger, Svrakic, & Przbybeck, 1993).
To summarize, the nature-nuture debate in depression (what part of depression is
genetically or environmentally determined) has become obsolete. The time has
come to give up the outmoded Cartesian dualism that characterizes much of the
literature on life stress and depression, and depression in general. For example,
estimates of heritability of what is often considered to be a purely "environmental"
factor such as social support have ranged between 40% and 80% (e.g., Henderson,
1998; Kendler, 1997). Again, personality dimensions such as Neuroticism and
Dependency/Sociotropy and Self-Critical Perfectionism/Autonomy are likely
mediators in the relationship between genetic factors and social support (Kendler,
1997). In this context, Kendler (200 I) distinguishes between "within-the-skin"
physiological pathways (i.e., the direct effect of genes on attitudes, affects, and
behaviors), and "outside-the-skin" behavioral pathways (i.e., the effects of genes on
the [social] environment, such as exposure to stress or social support) (see also
Silberg et aI., 1999; Thapar, Harold, & McGuffin, 1998). According to Kendler
(200 I, p. 1005), studies of both pathways provide "an important complement to and
context for current efforts to localize individual susceptibility genes". In general,
based on a meta-analysis (Sullivan, Neale, & Kendler, 2000), the heritability of
depression is estimated at 30-40%. How much of this heritability is due to a
"within-the-skin" or "outside-the-skin" effect is largely unknown, and may differ
depending on environmental factors. What is clear, however, is that research is
moving beyond the old dichotomies such as the nature versus nurture dichotomy
(Rose, 200 I) to more complex models that emphasize dynamic, recursive
interactions between biological and psychosocial factors, and between genes and
environment. As Rose (200 I, p. 3) has put it: "Organisms are in constant interaction
with their environment: that is, organisms select environments just as environments
select organisms. Like organisms, environments evolve and are homeodynamic
rather than homeostatic; both 'genome' and 'envirome' are abstractions from this
continuous dialectic". One of the great challenges of future research will be to
model and unravel these dialectic interactions. This brings us to our last point,
namely the need to consider depression from a developmental perspective.

266
The Need for a Developmental Perspective

Strikingly, all chapters in this volume emphasize the importance of a

developmental perspective. In Chapter I and 6, epidemiological research is cited
that clearly shows that depression is not a static and isolated disorder, but a highly
recurrent condition, which has its onset more often than not in childhood or
adolescence. In Chapter 2, Hermans and colleagues link disturbances in
autobiographical memory from a cognitive perspective to (early) trauma, thereby
implying a developmental perspective in the etiology of depression. In addition,
psychodynamic and cognitive-behavioral (see Chapter 3, 4, 5, and 7, this volume),
as well as neurobiological research (see Chapter 8, this volume), suggest that an
understanding of developmental factors is crucial to an understanding of
depression. This emphasis on developmental factors is striking because, as we
noted in the introduction of this volume, until recently many even doubted whether
depression in childhood even existed.
Congruent with the proposed etiologically-based, transdiagnostic alternative to
the DSM symptom-based categorization of depression and psychopathology in
general, these findings strongly suggest that research on the assessment and
treatment of depression should be based on a thorough knowledge of
developmental pathways that lead to expressions of depression (see also Ormel &
de long, 1999). Stated otherwise: the classification, assessment, and treatment of
psychopathology should be linked to normal developmental processes and to
disruptions in these processes (Blatt & Homann, 1992; Blatt & Levy, 1998). Such
a developmental perspective could provide a strong theoretical foundation and a
basis for integrating many of the issues that various approaches to depression, and
psychopathology more generally, have in common. Briefly, future research should
investigate how personal genetic predispositions correlate and interact with and
shape environmental events and how these interactions result in the construction of
mental representations of self and other in both normal and pathological
development. Psychopathology can then be conceptualized as disturbances in the
process of the development of representations of self and other (Blatt, 1995), such
that these representational structures become locked in relatively immature forms
that are relatively ineffective in later phases of the life cycle. These impairments of
developmental processes forestall subsequent development in some individuals and
thus these individuals need constructive reparative interpersonal experiences (e.g.,
in psychotherapy) that enable them to recognize the repetitive and maladaptive
nature of their mode of experiencing their interpersonal world and to be able to
revise this mode, based on experiences with a constructive caring other.
Unfortunately, as reviewed in detail in Chapter 6, few longitudinal studies exist
that have studied developmental pathways towards depression. In addition, these
studies have mostly concentrated on the relationship between DSM diagnoses, and
between these diagnoses and clinical variables, rather than on theoretically derived
hypotheses about possible etiological pathways. However, as illustrated in Chapter
6, 7, and 8, several interesting and promising areas of research exist, such as

267
research on care giving patterns of non-clinical and depressed parents and the
attachment styles of the children of these parents, and the role of early life stress on
neurobiological factors implied in depression and other disorders.
Although currently few truly interdisciplinary developmental studies of
depression exist, a true dialogue and integration between psychosocial and bio-
logical approaches of depression may develop under the influence of develop-
mental psychopathological research (Rutter, 2002). For example, theoretical
notions such as gene- and person-environment correlations and interactions are
broad enough to accommodate many different theoretical approaches, while at the
same time being specific enough to lead to clear and testable hypotheses concerning
the complex, recursive interactions between nature and nurture, between psyche
and soma. In addition, with the continuing development and growing availability of
complex statistical analyses methods such as Structural Equation Modeling (SEM),
survival analysis, and growth-curve modeling, we may be able to begin to answer
complex questions concerning the multiple intertwining of the biopsychosocial
factors implied in depression and other disorders.

The Limits of Dialogue and Integration

Before discussing research and clinical implications of the emerging dialogue in

depression research, it would be well to pause and ask if we have not overestimated
the possibilities for further dialogue and integration. And indeed, we believe that
some cautionary caveats are needed.
First, whereas many recognize limitations of DSM, there is no reason to believe
that the next edition of DSM will be fundamentally different. Plans for DSM- V, to
be published in 2012, mention no fundamental changes in the DSM approach to
mood disorders. In addition, although the recent Strategic Plan for Mood Disorders
Research by the National Institute of Mental Health (NIMH, 2003) recognizes
some limitations of DSM, and aims at funding interdisciplinary research on
depression, it has no plans to fund alternative research programs that aim at
developing an etiology-based classification of mood disorders.
Second, it must be clear that, despite considerable common ground, differences
do exist between various theoretical approaches of depression. These differences
are not only limited to differences concerning the DSM approach of depression,
but, as noted in Chapter 3, they often involve differences in their concept of human
nature that are not easily surmounted, particularly because these views are not
always based on empirical data, but also on ideological preferences.
Third, although dialogue and integration should ideally lead to more
comprehensive theories and ultimately more effective treatments, it is unlikely that
dialogue and integration have only advantages. In this context, one should carefully
distinguish between various forms of integration (Arkowitz, 1997). On a first level,
dialogue and integration often begins, and also but unfortunately ends, with

268
technical integration (i.e., incorporating techniques used in other theoretical
approaches). Hence, technical integration should be accompanied by theoretical
integration - by changes in theoretical formulations. Otherwise, the result could be
an "empty" eclecticism, without a firm grounding in overarching, comprehensive
theories. Messer (1986), for example, proposes the notion of "assimilative inte-
grationism", which refers to the fact that concepts, hypotheses, and therapeutic
techniques borrowed from other theoretical orientations are assimilated in a
theoretically meaningful way in an existing theoretical framework.
Finally, we believe that a certain tension between theoretical orientations is
healthy and even vital for further development in the field. Theoretical and
methodological pluralism often prevent monolithic thinking, which always risks
creating a one-sided, false sense of certainty. Hence, tension among various
approaches may be constructive, leading to new discoveries and paradigms.
But, as indicated in the various chapters in this volume, there appears to be much
common ground between various approaches of depression which should limit the
often artificial separation and competition between different theoretical approaches.
In addition, in clinical practice there is often much more similarity in clinical
approaches than is suggested by theoretical writings. For instance, Morrison,
Bradley and Westen (2003), studying a group of randomly selected psychiatrists
and clinical psychologists, found that the average clinician, regardless of theoretical
orientation (including CBT and IPT), treated depressed patients much longer than
is currently suggested in treatment guidelines. More importantly, they found,
regardless of theoretical orientation, that more than two thirds of the clinicians
indicated that the depressed patient they were describing had clinically significant
problems with relatedness or commitment in relationships, and with the expression
of anger, that needed treatment. Thus, most clinicians, regardless of theoretical
orientation, attend to similar issues. Likewise, in another study of patients with
eating disorders, Thompson-Brenner and Westen (2004) found that clinicians,
regardless of theoretical orientation, used more structured or more open clinical
interventions, depending on circumstances. For example, in patients with comorbid
(personality) pathology, clinicians, including cognitive-behavioral clinicians, used
more typical psychodynamic interventions. Ablon and Jones (2002), consistent
with these findings, found that treatment process in both IPT and CBT in the NIMH
Treatment of Depression Collaborative Research Program actually resembled most
the prototype for CBT, a surprising finding given that these therapists received
considerable training and supervision to adhere as closely as possible to their
therapeutic orientation.

Recommendations for Future Research

As Ingram and Price (200 I, p. ix) have noted, after an era in which research was
mainly focused on diagnosis and symptoms, "there is little doubt that the future of

269
clinical research and treatment efforts lies in the study of vulnerability processes."
This is particularly true for depression research. Moreover, given the high relapse
figures associated with depression, even when treated with so-called Empirically
Supported Treatments (ESTs) (Westen et aI., 2004a), an additional area of
increasing importance is the study of the different biopsychosocial factors
influencing relapse, particularly in first versus subsequent episodes of depression
(see Chapter I, this volume).
Thus, a refinement of diathesis-stress theories is clearly needed (Coyne &
Whiffen, 1995; Monroe & Simons, 1991). The lack of more comprehensive and
integrated theoretical models is currently one of the main impediments to further
dialogue and fuller integration of research findings concerning depression.
Ultimately, the integration of research findings should provide the basis for an
etiologically-based diagnostic alternative to the current a-theoretical approach of
DSM.
Methodological pluralism holds considerable promise to articulate and test
comprehensive, integrative theories. Using various methodologies, ranging from
N=I designs, to experimental and naturalistic studies in both animals and humans,
to epidemiological studies of representative community samples as well as studies
of high-risk samples (e.g., children of depressed mothers) should provide the basis
for more comprehensive theoretical formulations. In particular, we need inter-
disciplinary, developmentally informed, multiwave research programs to study
theoretically derived hypotheses concerning the dynamic interactions between
various biopsychosocial factors in normal and psychopathological development
(including depression). One of the most urgent problems in such studies is the need
to operationalize the construct of depression. As stated above, depression research
is currently divided over whether there are only quantitative or also qualitative
differences between normal dysphoria and clinical depression, although
increasingly the bulk of the evidence seems to support a continuum view. New
statistical analyses methods may help to settle this issue (Solomon et aI., 200 I).
Concerning treatment research, we believe that it is also time for a paradigm
change, both theoretically and methodologically (Beutler, Clarkin, & Bongar, 2000;
Guthrie, 2000; Westen et aI., 2004a). As Blatt, Shahar and Zuroff (2002) and
Westen et aI. (2004a) have argued, it is time to move beyond the dichotomous
judgment concerning treatments (or treatment packages) of depression as supported
versus unsupported, as is currently the case in the treatment guidelines for
identifying empirically supported treatments (e.g., APA, 2000), to identifying the
mechanisms of therapeutic change (Blatt et aI., 2002) and testing of specific
theories of change which then can be subsequently integrated in empirically
informed treatments (Westen et aI., 2004a). This will also necessitate the use of
research designs other than the randomized clinical trial (Blatt, Berman, Cook, &
Ford, 1998; Blatt & Zuroff, 2004).
As we noted in the Introduction to this volume, meta-analytic reviews of ;,0-
called ESTs provide a much more nuanced and often more pessimistic view
concerning the efficacy of these treatments (Westen et aI., 2004a). Concerning
depression, response rates in the short run are about 50%, with many patients
remaining symptomatic (Westen & Morrison, 2001). The few available studies on

270
the long term effects of ESTs for depression show that at best only about one third
of the patients remain improved after two years. In addition, as noted in the
Introduction to this volume, solid evidence for theoretically hypothesized effects of
ESTs for depression is largely lacking (e.g., Parker, Roy, & Evers, 2003).
Faced with these disappointing results, more extended forms of CBT and IPT
have been developed to provide continuation and/or maintenance treatment.
Although continuation and maintenance treatments have been shown to be effective
in reducing relapse, Westen et a\. (2004a) have correctly argued that this approach,
that is, extending brief treatments, may be highly problematic. In fact, it is ironic to
note that current guidelines for the treatment of depression now tend to promote
long-term versions of brief treatments. As Westen et a\. (2004a) suggest (see also
Blatt & Shahar, 2004; Blatt et a\., 2002), perhaps it is time to change our strategy,
and to ask ourselves from the start which patients need what kinds of treatment
(e.g., brief versus long-term). Hence, what is needed is the identification of
treatment relevant patient dimensions (Beutler et a\., 200 I). This also implies that
more research is needed on the efficacy and effectiveness of other, often long-term,
currently practiced treatments for depression, besides the brief treatments that are
currently included in lists of ESTs, especially in view of the growing research
literature that documents the effectiveness and efficacy as well as necessity of long-
term treatments such as experiential (Elliott, Greenberg, & Lietaer, 2004) and
psychodynamic psychotherapies (Bateman & Fonagy, 2001; Blatt, 1992; Blatt &
Shahar, 2004; Blomberg, Lazar, & Sandell, 2001; Leichsenring & Leibling, 2003).
Since the NIMH Strategic Plan for Mood Disorders Research (NIMH, 2003)
includes plans for funding outcome research of long-term treatments, more
information on the efficacy and effectiveness of these treatments should become
available in the near future.
Furthermore, we need methodological pluralism in treatment research. The
randomized clinical trial (RCT) is clearly not the only gold that glitters in outcome
research (Blatt & Zuroff, 2004; Slade & Priebe, 200 I; Westen et al., 2004a). For
example, because of strict exclusion criteria, it is often unknown how results from
RCTs can be translated to our daily clinical practice. More importantly, Westen et
a\. (2004a) have shown that RCTs, currently considered by many as the "gold
standard", are often inadequate for outcome research because many of the metho-
dological assumptions of RCTs (e.g., that psychopathology is highly malleable, that
most patients have only one diagnosis or disorder, etc.) are violated in the treatment
of various disorders. For example, RCTs are based on the assumption that a patient
typically has only one disorder, and can therefore be treated for that one disorder
with a manualized treatment. However, most patients show high rates of comor-
bidity between Axis I disorders and between Axis I and Axis II disorders, violating
this assumption. For research on depression, most assumptions of RCTs appear to
be violated, and thus alternative designs should be used or developed (for a detailed
discussion, see Westen et a\., 2004a).
From the various contributions to this volume, it also becomes clear that
interdisciplinary outcome research is needed that not only includes carefully chosen
symptomatic measures, but also broader outcome measures, including personality
vulnerabilities, level of psychosocial functioning, and the capacity to cope with new

271
life stressors ("resilience"; Blatt & Zuroff, 2004; Westen et aI., 2004a; Zuroff, Blatt,
Krupnick, & Sotsky, 2003). In addition, such studies should include multiwave,
clinically meaningful, follow-up intervals (e.g., from 2 to 5 years) to assess the
impact of treatment (Zuroff & Blatt, 2002; Westen et aI., 2004a). Given the high
relapse rates associated with brief treatments for depression, future psychotherapy
research should also investigate which patients need longer treatment and should
differentiate between patients who may be more responsive to supportive inter-
ventions and patients who may be more responsive to exploration and interpretation
(Blatt, 1992; Blatt & Shahar, 2004). In addition, studies have shown that (brief)
treatment of depression may only temporarily deactivate maladaptive cognitive-
affective schemas. Hence, psychotherapy studies should include measures that tap
into latent vulnerabilities, including Dependency/Sociotropy and Self-Critical
Perfectionism!Autonomy, that influence the onset and/or relapse of depression.
Studies need to address the role of these personality vulnerabilities in the etiology,
clinical course, and treatment of depression (Blatt et aI., 2002; Widiger &
Anderson, 2003). Finally, outcome research should routinely include both psycho-
social and biological outcome measures to unravel the relationship between process
and outcome at both the biological and the psychosocial level. The work of Kandel
(1999) suggests that psychosocial interventions may not only lead to
neurobiological changes, but also to changes in the expression of certain genes.
Hence, a truly interactive model also implies that psychosocial interventions may
lead to lasting biological changes, and that biological treatment may influence
either directly or indirectly the psychosocial factors implicated in depression (see
also Fonagy, Gergely, Jurist, & Target, 2004; Gabbard, 2000).
As Westen, Novotny and Thompson-Brenner (2004b) argue, this will necessitate
funding research that is not restricted to categorical DSM diagnoses and research
that is more closely in touch with clinical interventions as they are practiced in the
community. For example, research teams as well as research funding organizations
should include practicing clinicians in their review panels in efforts to reduce the
current gap between research and clinical practice.

Clinical Implications: Bridging the Gap between Research and Clinical Practice

With the consideration of the clinical implications of the emerging dialogue in

depression research, we have come full circle. As noted in the Introduction to this
volume, the efficacy and effectiveness of so-called ESTs for depression is quite
limited, and much remains unknown about the long term effects of currently
practiced ESTs and other forms of treatment for depression. Not surprisingly,
therefore, many clinicians are dissatisfied with current guidelines for the treatment
of depression. They have the feeling that these guidelines are simplistic and do not
do justice to the complexities of clinical practice. In fact, the study by Morrison et
al. (2003) clearly demonstrates that clinicians, including clinicians using CBT and

272
IPT, treat the average patient at least twice as long as prescribed by current
guidelines. In addition, most clinicians are of the opinion that the majority of
depressed patients need also treatment for clinical or subclinical personality
problems. And as Blatt and colleagues (Blatt, Sanislow, Zuroff, & Pilkonis, 1996)
demonstrated, more effective therapists believe that depressed patients need longer
treatment, primarily treatment that has a psychosocial focus. Thus, many clinicians
experience a wide gap between clinical guidelines promulgated by government
agencies and HMOs and what they encounter in their clinical practice.
The chapters in this volume clarify aspects of this gap between theory and
research on the one hand and clinical practice on the other. Let us recapitulate
briefly how this gap developed, and what should be done to close this gap. Initially,
it was believed that depression was a relatively benign disorder, which was
expected to respond well to brief pharmacological and psychosocial treatments. In
the meantime, epidemiological research, however, has pointed out that depression
is a highly recurrent disorder, which often has its onset in childhood or adolescence
(see Chapter I and Chapter 6, this volume). Moreover, depression shows high
comorbidity with other Axis I disorders (particularly anxiety disorders) and many
Axis II disorders. Many depressed patients also show subclinical personality
problems, such as problems concerning autonomy and identity, and intimacy and
relatedness (Blatt, 2004; Morrison et al., 2003). Also, depressed individuals often
continue to experience a range of subclinical personality and interpersonal
problems after brief treatment because of underlying, relatively stable and treat-
ment resistant, maladaptive cognitive-affective schemas or personality dimensions,
such as Dependency/Sociotropy and Self-Critical Perfectionism/Autonomy.
Therefore, they often remain vulnerable to future episodes, in part because of their
distinct cognitive biases (see Chapter 2, this volume) that lead them to generate
their own "depressogenic" environment (see Chapter 3, 4 and 5, this volume). In
addition, vulnerability associated with depression is probably also embedded in
relatively stable neurobiological disturbances, such as disturbances in HPA axis
functioning (Gutman & Nemeroff, 2003; see also Chapter 8, this volume). Not
surprisingly, therefore, brief treatments for depression have been relatively
ineffective.
The first response to these issues was to extend brieftreatments into maintenance
and continuation treatments. This approach certainly has its merits because it
results in much lower relapse rates. Another approach has been to develop
alternative treatments, particularly concerned about changing underlying maladap-
tive cognitive-affective schemas or personality dimensions (e.g., Beck, 1983, 1999;
Blatt, 1974,2004; McCullough, 2003; Linehan, 1993; Segal, Williams, & Teasdale,
2002; Young, 1999). Not surprisingly, many of these approaches originated from
experiences with the treatment of personality disorders (e.g., Beck, 1999; Linehan,
1993; Young, 1999), and often involve treatments that are considerable longer than
the typical 16-sessions of "traditional" CBT or IPT. This should not be surprising.
In fact, a general consensus has emerged that indicates that pharmacological
treatment for depression needs to be extended in many patients up to 6 months or
even a year or more (Geddes, Carney, Davies, Furukawa, Kupfer, Frank, &
Goodwin, 2003; Kupfer & Frank, 2001). Thus, perhaps the pendulum is starting to

273
swing in the other direction, after an era in which the predominant view was that
brief treatments were effective for all patients. Also, studies have been done and are
underway to determine which depressed patients benefit most from what kinds of
treatment (e.g., Blatt, 1992; Blatt & Shahar, 2004; Beutler et aI., 2000).
Where does this leave the clinician? Until more data become available, research
reviewed in this volume suggests the following preliminary guidelines, guidelines
that reflect our belief that assessment and treatment should be intrinsically linked
with each other and with basic research.
Assessment of depression. In accordance with the proposed dynamic inter-
actionism model of depression discussed earlier, assessment of the depressed
patient should consist of:
(1) Detailed assessment of depressive symptoms and the severity of depression
(2) A consideration of depression from a dimensional view, preferably including a
life-span perspective
(3) Charting of etiological and pathogenetic factors involved, with special attention
for the interaction between genetic factors (such as temperament), parental
styles, early adversity, subclinical (e.g., Dependency/Sociotropy and Self-
Critical Perfectionism/Autonomy) and clinical (personality disorders)
personality problems, and life stress. For example, course and treatment
prospects will look dramatically different when one is dealing with a depressed
patient functioning at the neurotic level, versus a depressed patient with
borderline features. However, current guidelines hardly account even such
obvious distinctions.
(4) A systematic statement concerning likely course and prognosis, together with
assessment of treatment possibilities for the patient. This should also include an
assessment of the patient's treatment preferences. Here the question is: what
treatment, if any treatment is needed, is likely to be most effective for which
patient? For example, recent research clearly suggests that dependent/socio-
tropic and self-critical/autonomous patients need a different therapeutic
approach that focuses on relatedness versus autonomy and self-definition,
respectively (Blatt, 1992; Blatt et aI., 2002; Blatt & Shahar, 2004). In addition,
self-critical/autonomous patients typically need more intensive treatment. More
research on the influence of these and other personality dimensions on process
and outcome in the treatment of depressed patients is needed. In particular,
congruent with our dynamic interactionism model, we believe that a crucial
factor in treatment is addressing the structural organization, content, and
development of maladaptive and often rigid cognitive-affective structures in the
context of the therapeutic relationship.

Several instruments are available to document more objectively steps one to

three listed above. Concerning step four, the clinician has currently precious little
information, precisely because etiological factors are often excluded in outcome
research. However, recent work is increasingly filling this gap (e.g., Beutler et al.,
2000; Blatt, 1992; Blatt & Shahar, 2004).

274
 Treatment of depression. In general, clinicians should benefit from a growing
dialogue and integration between various theoretical approaches. It is clear that
earlier exaggerated differences among various forms of psychotherapy are starting
to diminish. For example, psychodynamic approaches of depression are becoming
more flexible and directive and are incorporating techniques from other
approaches, while cognitive-behavioral approaches incorporate several techniques
and conceptualizations from other theoretical approaches, including a fuller
appreciation of childhood experiences and the interpersonal dimensions of the
treatment process, mainly from the psychodynamic and experiential tradition
(Bemporad, 1995; Blatt, 2004; Bateman & Fonagy, 2004; Jones & Pulos, 1993;
Kwon, 1999).
Cognitive-behavioral approaches are not only increasingly focusing on
underlying cognitive-affective schemas, but also are beginning to consider therapist
competence (Shaw et aI., 1999), and the therapeutic relationship and transference
in particular as an important source of information and vehicle for interventions
(e.g., Linehan, 1993; Waddington, 2002). Mindfulness Based Cognitive Therapy
(MBCT; Segal et aI., 2002; Teasdale et aI., 2000), a treatment package specifically
aimed at reducing relapse in depression, in tum, focuses on developing a meta-
perspective in depressed patients, which entails patients observing their own
(ruminative) thoughts without judging or suppressing them (see also Chapter 2, this
volume). This comes very close to the psychoanalytic emphasis on exploration, free
association, and interpretation, and particularly to the notion of mentalization or
reflective functioning, which is the basis of Mentalization Based Treatment (MBT),
a psychoanalytic, evidence-based long-term treatment for borderline personality
disorder (Bateman & Fonagy, 2004). As Hermans and colleagues note in Chapter
2, mindfulness based treatment is about "providing patients with life-long skills for
adequately dealing with negative emotion in the future". In psychodynamic terms,
this refers to the internalization of the "analytic function", i.e., the capacity for
insight and mentalization (Blatt & Behrends, 1987). Interestingly, patients treated
with MBT continued to improve during an 18-month follow-up (Bateman &
Fonagy, 200 I), which is dramatically different than the typical high relapse rates
after brief treatments for most disorders. Findings such as these not only illustrate
the need for long-term follow-up in outcome research, but at the same time they
also illustrate that longer treatment may result in substantial and sustained change.
These integrative trends in psychosocial treatments clearly reflect the fact that
brief treatments, which focus on "surface" cognitions and symptoms, are inef-
fective in many patients. Instead, the focus has shifted towards underlying relatively
stable cognitive-affective schemas, or personality dimensions, in order to prevent
relapse. Further dialogue and integration concerning the treatment of depression
could be achieved by focusing on the five areas of convergence earlier discussed,
i.e., (I) taking into account etiological and pathogenetic factors in the classification
of mental disorders, (2) focusing on mental representations and underlying
cognitive-affective schemas or personality dimensions, (3) taking into account the
interpersonal dimensions of depression, and (4) more generally, the role of
environmental factors and their interaction with depression and cognitive-affective
schemas, and (5) focusing on developmental factors implied in depression.

275
 Concerning treatment more specificaIly, findings reviewed in this volume point
to the fact that current guidelines for the treatment of depression have the following
limitations:
(I) They underestimate the recurrent nature of depression
(2) They largely neglect patient, therapist, and therapeutic aIliance factors in favor
of specific techniques, and
(3) They overemphasize the importance of symptoms to the neglect of the role of
etiological and pathogenetic factors implied in depression (see Blatt, 2004).

It is clear that these issues are linked to the fact that current guidelines are closely
linked to the DSM approach of depression. If, from a DSM perspective, depression
is a relatively benign Axis I disorder (a state disorder, unrelated to personality), that
can be diagnosed and treated based on an assessment of symptoms, one can develop
and manualize, analogous to the development of a medication for a disease, a
treatment package specificaIly designed for this particular "disease". When
symptoms disappear, the disease has been "cured". The disease model and the drug
metaphor of the DSM (e.g., see Stiles & Shapiro, 1989; Wampold, 1997) has clearly
impacted on the current guidelines for the treatment of depression. As the findings
reviewed in this volume illustrate, all these DSM assumptions have been refuted by
empirical research (Blatt & Zuroff, 2004). Therefore, until more data become
available, clinicians should attend particularly to the foIlowing issues when treating
depressed patients:

(I) Depression is often a highly recurrent disorder, and should be seen as a

distortion of normal development. Many patients treated for depression have had
several episodes of depression and experience significant psychosocial difficulties,
often beginning in childhood or adolescence. Because many depressed patients also
have considerable subclinical or clinical personality problems, most often reflected
in interpersonal difficulties (e.g., problems concerning relatedness), it is extremely
unlikely that such lifelong difficulties will be solved, or even diminished, by
focusing on symptoms alone. While long-term treatment may not be indicated for
many patients, all treatments should be concerned about reducing relapse. This may
be achieved by different means, ranging from psycho-education, to using treatment
packages that are specificaIly aimed at reducing relapse (e.g., Segal et al., 2002), to
more extended treatment (Blatt et aI., 2002). Because personality dimensions or
cognitive-affective structures playa central role in the development of depression,
as stated above, treatment should focus on the development of the structural
organization and content of these cognitive-affective structures in the context of the
therapeutic relationship. Moreover, because the importance of relational factors in
depression, both as an antecedent and as a consequence, treatment should also pay
particularly attention to interpersonal issues, both in the focus of the treatment and
in the consideration of the need for couple and/or family therapy.

276
 (2) More weight should be given to patient, therapist, and alliance factors
(Beutler et al., 2001; Blatt & Zuroff, 2004; Lambert & Ogles, 2004). As stated
before (see also Chapter 4, this volume), it is ironic that current guidelines for the
treatment of depressed patients have almost exclusively focused on particular
therapeutic techniques or treatment packages given the fact that more than four
decades of psychotherapy research have pointed to the importance of patient (Blatt,
2004; Clarkin & Levy, 2004), and therapist (Beutler et aI., 2004) variables, as well
as the therapeutic alliance (Blatt & Zuroff, 2004; Division 29 Task Force on
Empirically Supported Therapy Relationships, 2002). This neglect of personality
and interpersonal factors in the therapeutic process is all the more ironic given the
fact that, with some important exceptions, for many disorders different forms of
treatment appear to have similar effects (the famous "Dodo bird verdict").
Therefore, we would like to make a strong plea to include consideration of patient,
therapist, and alliance factors in establishing guidelines for the treatment of
depressed patients.
(3) Etiological and pathogenetic factors should inform treatment. Depressed
patients are a heterogeneous group from an etiological and pathogenetic
perspective. Current guidelines, however, treat depressed patients as uniformly
responsive to a variety of psychosocial (e.g., CBT, IPT) and pharmacological
treatments. This assumption of uniformity is clearly incorrect. Again, because these
treatment guidelines were not informed by basic science (i.e., research on etiologi-
cal pathways to depression), the influence of etiological factors on treatment res-
ponse has not been given sufficient attention. Therefore, considerations concerning
etiological and pathogenetic factors should be included in developing treatment
guidelines. In particular, special attention should be given to the role of genetic
factors, which can exert their influence both "inside" and "outside" the skin
(Kendler, 200 I), neurobiological disturbances associated with HPA axis
functioning, and their interaction with psychosocial factors such as parental style,
early adversity, life stress, and relatively stable personality dimensions or cognitive-
affective schemas.

Conclusions

Our work on this volume has made us acutely aware of the increasing oppor-
tunities of further dialogue and integration in depression research. In particular, an
etiologically-based, dynamic interactionism view of depression, emphasizing
interactions among genetics, early adversity, current life stress, and relatively stable
cognitive-affective schemas or personality dimensions, emerges as a model that
may facilitate the integration of various theoretical, methodological, and clinical
approaches to depression. At the same time, much work remains to be done. The
future of research on depression therefore promises to be exciting and will
undoubtedly lead to many new advances in our knowledge of this disabling

277
disorder. It is our sincere belief that ultimately the quality of clinical services will
benefit from these advances. In particular, the proposed etiologically-based dyna-
mic interactionism model not only promises to lead to important theoretical insights
in depression that can subsequently inform treatment of depression, but it may also
assist in developing theory-driven and empirically-based prevention programs that
are aimed at etiological and pathogenetic factors implied in depression.
In closing, we would like to re-emphasize in this context the importance of
maladaptive representations in the development and treatment of depression, that is,
how maladaptive representations of self and other, that have developed starting in
early life as a result of interactions between biological and psychosocial factors,
make certain individuals vulnerable to particular depressive experiences and other,
related subclinical and clinical problems, such as anxiety, substance abuse, and
perhaps also somatic and functional disorders. Hence, the notion of mental
representations may not only build bridges between various approaches towards the
etiology and pathogenesis of depression, it may also provide a unified view 011 the
treatment of depression. A focus on mental representations implies, regardless of
theoretical orientation, that the treatment of depression should primarily consist in
making these individuals aware of their maladaptive representations and help them
to build and internalize more adaptive representations of self and others.

References

Ablon, 1. S., & Jones, E. E. (2002). Validity of controlled clinical trials of psychotherapy: Findings from
the NIMH Treatment of Depression Collaborative Research Program. American Journal of Psychiatry,
/59, 775-783.
American Psychiatric Association (2000). Practice guidelines for the treatment of patients with major
depression. Retrieved September 3. 2004, from https://www.psych.org/psych_pract/trealg/pg/
Depression2e.book.cfm
Arkowitz, H. (1997). lntegrative theories of therapy. In P. Wachtel & S. Messer (Eds.), Theories of
psychotherapy: Origins and evolution (pp. 227-288). Washington. DC: American Psychological
Association Press.
Bateman, A., & Fonagy, P. (2001). Treatment of borderline personality disorder with psychoanalytically
oriented partial hospitalisation: An 18-month follow-up. American Journal of Psychiatry, /58, 36-42.
Bateman, A., & Fonagy, P. (2004). Psychotherapvfor borderline personality disorder. Mentalizatlon-based
treatment. Oxford: Oxford University Press.
Beach, S. R. H. (Ed). (2001). Marital and familv processes in depression. A scientific foundation for
clinical practice. Washington, DC: American Psychological Association.
Beck, A. T. (1983). Cognitive therapy of depression: New perspectives. In P. J. Clayton & J. E. Barrett
(Eds.), Treatment of depression: Old controversies and new approaches (pp. 265-290). New York:
Raven Press.
Beck, A. T. (1999). Cognitive aspects of personality disorders and their relation to syndromal disorders: A
psychoevolutionary approach. In C. R. Cloninger (Ed.), Personality and psychopathology (pp. 411-
429). Washington, DC/London: American Psychiatric Press.
Bernporad, J. R. (1995). Long-term analytic treatment of depression. In E. E. Beckham & W. R. Leber
(Eds.), Handbook of depression (2nd ed.) (pp. 391-403). New York: Guilford Press.
Beutel, M. E., Stern, E., & Silbersweig. D. A. (2003). The emerging dialogue between psychoanalys sand
neuroscience: Neuroimaging perspectives. Journal (!f the American Psychoanalytic Association, 5/,
773-801.

278
Beutler, L. E., Clarkin, 1. E, & Bongar, B. (2000). Guidelines for the systematic treatment of the depressed
patient. New York/Oxford: Oxford University Press.
Beutler, L. E., Malik, M., Alimoharned, S., Harwood, T. M., Talebi, H., Noble, S., & Wong, E. (2004).
Therapist variables. In M. 1. Lambert (Ed.), Bergin and Garfield's handbook of psychotherapy and
behavior change (5th ed.) (pp. 227-306). New York: John Wiley & Sons.
Bieling, P. 1., Summerfeldt, L. 1., Israeli, A. L., & Antony, M. M. (2004). Perfectionism as an explanatory
construct in comorbidity of Axis I disorders. Journal of Psychopathology and Behavioral Assessment,
26, 193-201.
Blatt, S. 1. (1974). Levels of object representation in anaclitic and introjective depression. The
Psychoanalytic Study of the Child, 29, 107-157.
Blatt, S. 1. (1991). Depression and destructive risk-taking behavior in adolescence. In L. P. Lipsitt & L. L.
Mitnick (Eds.), Self-regulatory behavior and risk-taking: Causes and consequences (pp, 285-309).
Norwood, Nl: Ablex Press.
Blatt, S. 1. (1992). The differential effect of psychotherapy and psychoanalysis with anaclitic and
introjective patients: The Menninger psychotherapy research project revisited. Journal of the American
Psvchoanalvtic Association, 40, 691-724.
Blatt, S. 1. (1995). Representational structures in psychopathology. In D. Cicchetti & S. L. Toth (Eds.),
Emotion. cognition, and representation (Rochester Symposium on developmenal psvchopathology Vol.
6) (pp, 1-33). New York: University of Rochester Press.
Blatt, S. 1. (2004). Experiences ofdepression: Theoretical, clinical and research perspectives. Washington,
DC: American Psychological Association.
Blatt, S. J., Auerbach, 1. S., & Levy, K. N. (1997). Mental representation in personality development,
psychopathology, and the therapeutic process. Review of General Psychology, 1,351-374.
Blatt. S. 1.. & Behrends, R. S. (1987). Internalization, separation-individuation, and the nature of
therapeutic action. International Journal o] Psycho-Analysis, 68. 279-297.
Blatt, S. J., Berman, W. H., Cook, B., & Ford. R. Q. (1998). Effectiveness of long-term. intensive. inpatient
treatment for seriously disturbed young adults: A reply to Bein. Psychotherapy Research, 8, 42-53.
Blatt. S. 1.• & Homann, E. (1992). Parent-child interaction in the etiology of dependent and self-critical
depression. Clinical Psychology Review. 12.47-91.
Blatt. S. 1., & Levy, K. N. (1998). A psychodynamic approach to the diagnosis of psychopathology. In 1.
W. Baron (Ed.), Making diagnosis meaningful. Enhancing evaluation and treatment o] psychological
disorders (pp. 73-109). Washington. DC: American Psychological Association.
Blatt. S. 1.. Sanislow, C. A.. Zuroff, D. C.. & Pilkonis, P. A. (1996). Characteristics of effective therapists:
Further analyses of data from the NIMH TDCRP. Journal (if Consulting and Clinical Psvchologv, 64,
1276-1284.
Blatt, S, J., & Shahar, G. (2004). Psychoanalysis: With whom, for what. and how'? Comparisons with
psychotherapy. Journal of the American Psychoanalytic Association, 52. 393-447.
Blatt, S. J., Shahar, G.• & Zuroff, D. C. (2002). Anaclitic/Sociotropic and Introjective/Autonomous
dimensions. In J. C. Norcross (Ed.), Psychotherapy relationships that work. Therapist contributions
and responsiveness to patients (pp, 315-333). Oxford: Oxford University Press.
Blatt, S. 1., & Shichman, S. (1983). Two primary configurations of psychopathology. Psychoanalysis and
Contemporary Thought. 6. 187-254.
Blatt, S. J., Stayner, D., Auerbach. 1. S.. & Behrends. R. S. (1996). Change in object and self repre-
sentations in long-term, intensive, inpatient treatment of seriously disturbed adolescents and young
adults. Psychiatry: Interpersonal and Biological Processes, 59. 82-107.
Blatt, S. L Wein, S. 1., Chevron. E., & Quinlan. D. M. (1979). Parental representations and depression in
normal young adults. Journal ofAbnormal Psychology, 88, 388-397.
Blatt, S. J., & Zuroff, D. C. (1992). Interpersonal relatedness and self-definition: Two prototypes for
depression. Clinical Psychology Review, 12. 527-562.
Blatt. S. 1.. & Zuroff, D. C. (2004). Empirical evaluation of the assumptions in identifying evidence-based
treatments in mental health. Manuscript submitted for publication.
Blomberg. 1.. Lazar, A.. & Sandell, R. (2001). Long-term outcome of long-term psychoanalytically
oriented therapies: First findings of the Stockholm Outcome of Psychotherapy and Psychoanalysis
Study. Psychotherapy Research. 11.361-382.
Bonanno, G. A. (2004). Loss. trauma, and human resilience. Have we underestimated the human capacity
to thrive after extremely aversive events') American Psychologist, 59. 20-28.
Bowlby. 1. (1988). A secure base: Clinical applications tif attachment theory. London: Routledge.

279
Brown, G. W, & Harris, T. O. (1978). Social origins of depression. Routledge: London.
Caspi, A., Sugden, K., Moffitt, T. E., Taylor, A., Craig, I. W, Harrington, H., McClay, J., Mill, J., Martin,
J., Braithwaite, A., & Poulton, R. (2003). Influence of life stress on depression: Moderation by a
polymorphism in the 5-HTT gene. Science, 301,386-389.
Charney, D. S. (2004). Psychobiological mechanisms of resilience and vulnerability: Implication; for
successful adaptation to extreme stress. American Journal of Psychiatry, 161, 195-216.
Cicchetti, D., & Cohen, D. J. (\995). Perspectives on developmental psychopathology. In D. Cicchetti &
J. Cohen (Eds.), Developmental psychopathology (pp. 3-20). New York: Wiley.
Claes, S. (2003). Corticotropin-releasing hormone (CRH) in psychiatry: From stress to psychopathology.
Annals of Medicine, 35, 1-12.
Clarkin, J., & Levy, K. N. (2004). The influence of client variables on psychotherapy. In M. J. Lambert
(Ed.), Bergin and Garfield's handbook ofpsychotherapy and behavior change (5th ed.) (pp. 194-226).
New York: John Wiley & Sons.
Cloninger, C. R., Svrakic, D. K.. & Przbybeck, T. R. (1993). A psychobiological model of temperanent
and character. Archives of General Psychiatry. 50. 975-990.
Coyne, J. C; Burchill, S. A. L., & Stiles. W B. (\991). An interactional perspective on depression. In C.
R. Snyder & D. O. Forsyth (Eds.), Handbook ofsocial and clinical psychology: The health perspective
(pp. 327-349). New York: Pergamon.
Coyne, J. C; & Whiffen, V. E. (\995). Issues in personality as diathesis for depression: The ease of
sociotropy-dependency and autonomy-self-criticism. Psychological Bulletin, 118,358-378.
Denton, W. H., Golden, R. N.. & Walsh. S. R. (2003). Depression, marital discord, and couple therapy,
Current Opinion in Psychiatry. 16. 29-34.
Division 29 Task Force on Empirically Supported Therapy Relationships (2002). Conclusions and
recommendations of the Division 29 Task Force. In J. C. Norcross (Ed.), Psychotherapy relationships
that work (pp. 441-443). Oxford: Oxford University Press.
Elliott, R., Greenberg, L. S.. & Lietaer, G. (2004). Research on experiential psychotherapies. In M. J.
Lambert (Ed.), Bergin and Garfield's handbook ofpsychotherapy change (5th ed.) (pp. 493-539). New
York: John Wiley & Sons.
Fonagy, P., Gergely, G., Jurist, E. L.. & Target. M. (2004). Affect regulation, mentalization, and the
development of the self London/New York: Karnac.
Gabbard, G. O. (2000). A neurobiologically informed perspective on psychotherapy. British Journal of
Psychiatry, 177,117-122.
Geddes, J. R., Carney, S. M.. Davies, c.. Furukawa. T. A., Kupfer, D. J., Frank, E., & Goodwin, G. M.
(2003). Relapse prevention with antidepressant drug treatment in depressive disorders: A systematic
review. The Lancet, 361, 653-661.
Gilmer, W. S., & McKinney, W T. (2003). Early experience and depressive disorders: Human and non-
human primate studies. Journal ofAffective Disorders. 75.97-113.
Gold, P. W., & Chrousos, G. P. (2002). Organization of the stress system and its dysregulation in
melancholic and atypical depression: High vs low CRH/NE states. Molecular Psychiatry, 7, 254-275.
Goldapple, K., Segal, Z., Garson, C; Lau, M.. Bieling, P., Kennedy, S., & Mayberg, H. (2004). Modulation
of cortical-limbic pathways in major depression. Archives of General Psychiatry, 61, 31-41.
Greenberg, L., & Watson. J. (\998). Experiential therapy of depression: Differential effects of client-
centered relationship conditions and process experiential interventions. Psychotherapy Research, 8.
210-224.
Guthrie, E. (2000). Psychotherapy for patients with complex disorders and chronic symptoms. The need
for a new research paradigm. British Journal of Psychiatry, 177. 131-137.
Gutman, D. A.• & Nemeroff, C. B. (2003). Persistent central nervous system effects of an adverse early
environment: Clinical and preclinical studies. Physiology & Behavior, 79,471-478.
Hammen. C. (2003). Interpersonal stress and depression in women. Journal of Affective Disorders. 74,
49-57.
Harrington, R. C. (2001). Childhood depression and conduct disorder: Different routes to the same
outcome? Archives of General Psychiatry, 58. 237-238.
Henderson. A. S. (1998). Social support: Its present significance for psychiatric epidemiology. In B. P.
Dohrendwend (Ed.), Adversity, stress, and psychopathology (pp. 390-397). New York: Oxford
University Press.
Ingram, R. E., & Price, J. M. (2001). Preface. In R. E Ingram & J. M. Price (Eds.), vulnerabiliry 10
psychopathology. Risk across the lifespan (p. ix-xii), New YorkILondon: The Guilford Press.

280
Joiner, T. E., Jr. (200 I). Nodes of consilience between interpersonal-psychological theories of depression.
In S. R. H. Beach (Ed), Marital and family processes in depression. A scientific foundation for clinical
practice (pp. 129-138). Washington, DC: American Psychological Association.
Joiner. T. E., Jr., Metalsky, G. I., Katz, 1., & Beach, S. R. H. (1999). Depression and excessive reassurance-
seeking. Psychological Inquiry, 10, 269-278.
Jones, E. E., & Pulos, S. M. (1993). Comparing the process of psychodynamic and cognitive-behavioral
therapies. Journal of Consulting and Clinical Psychology, 61, 306-316.
Judd, L. L., Akiskal, H. S., & Paulus, M. P. (1997). The role and clinical significance of subsyndromal
depressive symptoms (SSD) in unipolar major depressive disorder. Journal of Affective Disorders, 45,
5-17.
Kandel, E. R. (1999). Biology and the future of psychoanalysis: A new intellectual framework for
psychiatry revisited. American Journal of Psychiatry, 156, 505-524.
Kasen, S.. Cohen, P., Skodol, A. E., Johnson, J. G., Smailes, E., & Brook, J. S. (2001). Childhood
depression and adult personality disorder. Alternative pathways of continuity. Archives of General
Psychiatry, 58, 231-236.
Kendler, K. S. (1997). Social support: A genetic-epidemiologic analysis. American Journal of Psychiatry,
154, 1398-1404.
Kendler, K. S. (200 I). Twin studies of psychiatric illness. An update. Archives of General Psychiatry, 58,
1005-1014.
Kendler, K. S., & Gardner, C. O. (1998). Boundaries of major depression: An evaluation of DSM-IV
criteria. American Journal of Psychiatry, 155. 172-177.
Kendler, K. S., Gardner, C. 0., & Prescott, C. A. (2003). Personality and the experience of environmental
adversity. Psychological Medicine. 33, 1193-1202.
Kendler, K. S.• & Kurkowski-Shuman, L. (1997). Stressful life events and genetic liability to major
depression: Genetic control of exposure to environment? Psychological Medicine, 27, 539-547.
Kendler, K. S.• Kuhn, J., & Prescott. C. A. (2004). The interrelationship of neuroticism, sex, and stressful
life events in the prediction of episodes of major depression. American Journal of Psychiatry, 161.631-
636.
Kendler, K. S., Neale. M., Kessler, R., Heath, A., & Eaves, L. (1993). A longitudinal twin study of
personality and major depression in women. Archives of General Psychiatry, 50, 853-862.
Kessler, R. c., Merikangas, K. R., Berglund, P., Eaton, W W, Koretz, D. S., & Walters, E. E. (2003). Mild
disorders should not be eliminated from the DSM-Y. Archives of General Psychiatry, 60. 1117-1122.
Klein. D. N.. Lewinsohn, P. M.. Seeley. J. R.. & Rohde, P. (2001). A family study of major depressive
disorder in a community sample of adolescents. Archives of General Psvchiatrv, 58, 13-20.
Kupfer. D. J., & Frank, E. (2001). The interaction of drug- and psychotherapy in the long-term treatment
of depression. Journal ofAffective Disorders. 62, 131-137.
Kwon, P. ( 1999). Attributional style and psychodynamic defense mechanisms: Toward an integrative model
of depression. Journal (!{ Personality. 67. 645-658.
Laakso. A., Wallius, E.. Kajander, J., Bergman, J.. Eskola, 0., Solin, 0., Ilonen, T., Salokangas, R. K. R..
Syvalahti, E.. & Hietala, J. (2003). Personality traits and striatal dopamine synthesis capacity in healthy
subjects. American Journal ofPsychiatrv, 160,904-910.
Lambert, M. J., & Ogles, B. M. (2004). The efficacy and effectiveness of psychotherapy. In M. J. Lambert
(Ed.), Bergin and Garfield's handbook ofpsychotherapy and behavior change (5th ed.) (pp, 139-193).
New York: John Wiley & Sons.
Leichsenring, D., & Leibling, E. (2003). The effectiveness of psychodynamic therapy and cognitive
behavior therapy in the treatment of personality disorders: A meta-analysis. American Journal of
Psychiatry, 160, 1223-1232.
Lewinsohn, P. M., Rohde, P.. Seeley, J. R.. Klein, D. N., & Gotlib, I. H. (2003). Psychosocial functioning
of young adults who have experienced and recovered from major depressive disorder during
adolescence. Journal (if Abnormal Psychology. 112,353-363.
Lewinsohn, P. M., Shankman, S. A.. Gau, J. M., & Klein, D. N. (2004). The prevalence and co-morbidity
of subthreshold psychiatric conditions. Psychological Medicine, 34, 613-622.
Linehan. M. M. (1993). Cognitive-behavioral treatment of borderline personality disorder. New York: The
Guilford Press.
Main. M., Kaplan, N., & Cassidy, J. (1985). Security in infancy, childhood, and adulthood: A move to the
level of representation. In I. Bretherton & E. Waters (Ed.), Growing points in attachment theory and
research. Monographs (if the Society-for Research in Child Development, 50. 66-104.

281
McCullough, J. P. (2003). Treatment for chronic depression: Cognitive Behavioral Analysis System of
Psychotherapy (CBASP). London: The Guilford Press.
Messer, S. (1986). Behavioral and psychoanalytic perspectives at therapeutic choice points. American
Psychologist, 41, 1261-1272.
Monroe, S. C., & Simons, A. D. (1991). Diathesis-stress theories in the context of life stress research:
Implications for the depressive disorders. Psychological Bulletin, 110,406-425.
Moore, R. G., & Blackburn, I.-M. (1993). Sociotropy. autonomy and personal memories in depression.
British Journal of Clinical Psychology, 32. 460-462.
Morrison, K. H., Bradley, R., & Westen. D. (2003). The external validity of controlled clinical trials of
psychotherapy for depression and anxiety: A naturalistic study. Psychology and Psychotherapy: Theory.
Research and Practice, 76. 109-132.
National Institute of Mental Health (2003). Breaking ground, breaking through: The Strategic Plan for
Mood Disorders Research. Retrieved September 3. 2004. from https://www.nimh.nih.gov/strategic/
mooddisorders.pdf
Nemeroff, C. B. (2002). Editorial. Co morbidity of mood and anxiety disorders: The rule, not the exception?
American Journal of Psychiatry, 159. 3-4.
Nunn, J. D., Mathews, A., & Trower. P. (1997). Selective processing of concern-related information in
depression. British Journal of Clinical Psychology. 36. 489-503.
Ormel, 1.. & de Jong, A. (1999). On vulnerability to common mental disorders: An evidence-based plea for
a developmental perspective. In M. Tansella & G. Thornicroft (Eds.), Common mental disorders in
primary care (pp. 34-52). London: Routledge.
Orrnel, 1.. Oldehinkel. A. 1.. & Brilman, E. I. (2001). The interplay and etiological continuity of neuro-
ticism, difficulties, and life events in the etiology of major and subsyndromal, first and recurrent
depressive episodes in later life. American Journal of Psychiatry. 158, 885-891.
Orrnel, 1., Stewart, R.• & Sanderman. R. (1989). Personality as modifier of the life change-distress
relationship: A longitudinal modelling approach. Social Psychiatry and Psychiatric Epidemiology. 24,
187-195.
Ottens, A. J., & Hanna, F. J. (1998). Cognitive and existential therapies: Toward an integration.
Psychotherapy: Theory. Research. Practice. Training, 35. 312-324.
Parker, G. (1999). Clinical trials of antidepressant medications are producing meaningless results. British
Journal of Psychiatry. 183. 102-103.
Parker, G. (2000). Classifying depression: Should paradigms lost be regained? American Journal of
Psychiatry, 157, 1195-1203.
Parker, G., Hadzi-Pavlovic, D.. Roussos. J.. Wilhelm. K.. Mitchell. P.• Austin, M.-P.. Hickie, I., Gladstone,
J.• & Eyers, K. (1998). Non-melancholic depression: The contribution of personality, anxiety ami life
events to subclassification. Psychological Medicine. 28. 1209-1219.
Parker. G., Roy, K., & Eyers, K. (2003). Cognitive behavior therapy for depression? Choose horse; for
courses. American Journal of Psychiatry. 160.825-834.
Parker, G., Roy, K., Wilhelm. K.. Mitchell. P., Austin. M.-P.. Hadzi-Pavlovic, D., & Little, C. (1999a). Sub-
grouping non-melancholic depression from manifest clinical features. Journal of Affective Disorders,
53, 1-13.
Parker, G., Roy, K., Wilhelm. K.. Mitchell. P., Austin, M.-P.. & Hadzi-Pavlovic, D. (l999b). Subgrouping
non-melancholic major depression using both clinical and aetiological features. Australian and Nell'
Zealand Journal of Psychiatry, 33. 217-225.
Paykel, E. S. (2003). Editorial. Life events: Effects and genesis. Psychological Medicine, 33, 1145-1148-
Pier, M. P. B. I., Hulstijn, w.. & Sabbe, B. (2004). Differential patterns of psychomotor functionirg in
unmedicated melancholic and nonmelancholic depressed patients. Journal of Psychiatric Research, 38.
425-435.
Raison, C. L., & Miller, A. H. (2003). When not enough is too much: The role of insutlicient glucocorticoid
signaling in the pathophysiology of stress-related disorders. American Journal of Psychiatry. 160.
1554-1565.
Rose, S. (200 I). Moving on from old dichotomies: Beyond nature-nurture towards lifeline perspective.
British Journal (if Psychiatrv, 178 (SuppI40), s3-s7.
Rutter, M. (2002). The interplay of nature. nurture. and developmental influences. The challenge ahead for
mental health. Archives (if General Psychiatry, 59, 996-1000.
Rutter, M., Dunn, J., Plomin, R., Simonoff, E., Pickles. A., Maughan, B., Ormel, 1., Meyer, 1., & Eaves, L.
(1997). Integrating nature and nurture: Implications of person-environment correlations and

282
 interactions for developmental psychopathology. Development and Psychopathology, 9, 335-364.
Safran, J. D., & Segal, L. S. (1990). Interpersonal process in cognitive therapy. New York: Basic Books.
Saudi no, K. 1., Pedersen, N. L., Lichtenstein, P., McClearn, G. E., & Plomin, R. (1997). Can personality
explain genetic influences on life events'! Journal of Personality and Social Psychology, 72, 196-206.
Segal, Z. v., Williams, 1. M. G., & Teasdale, 1. D. (2002). Mindfulness-based Cognitive Therapy for
Depression: A new approach to preventing relapse. New York: The Guilford Press.
Segerstrom, S. C, & Miller, G. E. (2004). Psychological stress and the human immune system: A meta-
analytic study of 30 years of inquiry. Psychological Bulletin, 130,601-630.
Segrin, C (2000). Social skills deficits associated with depression. Clinical Psychology Review, 20, 379-
403.
Shafran, R., & Mansell, W. (2001). Perfectionism and psychopathology: A review of research and
treatment. Clinical Psychology Review, 21, 879-906.
Shahar, G., Blatt, S. 1., Zuroff, D. C, Krupnick, J. L., & Sotsky, S. M. (2004). Perfectionism impedes social
relations and response to brief treatment for depression. Journal of Social and Clinical Psychology, 23,
140-154.
Shaw, B. E, Elkin, I., Yamaguchi, J., Olmsted, M., Vallis. T. M., Dobson, K. S., Lowery, S., Sotsky, S. M.,
Watkins, J. T., & Imber, S. D. (1999). Therapist competence ratings in relation to clinical outcome in
cognitive therapy of depression. Journal of Consulting and Clinical Psychology, 60, 441-449.
Shedler, J., & Westen, D. (2004). Refining personality disorder diagnosis: Integrating science and practice.
American Journal of Psychiatry, 161, 1350-1365.
Silberg, J., Pickles, A., Rutter, M., Hewitt, J., Simonoff, E., Maes, H., Carbonneau, R., Murrelle, L., Foley,
D., & Eaves, L. (1999). The influence of genetic factors and life stress on depression among adolescent
girls. Archives of General Psychiatry, 56, 225-232.
Slade, M., & Priebe, S. (2001). Editorial. Are randomised controlled trials the only gold that glitters'!
British Journal (if Psychiatry, 179. 286-287.
Solomon, A., Haaga, D. A. E, & Arnow, B. A. (200 I). Is clinical depression distinct from subthreshold
depressive symptoms'! A review of the continuity issue in depression research. Journal of Nervous and
Mental Disease, 189,498-506.
Stern, D. N. (1985). The interpersonal world of the infant: A view from psychoanalysis and developmental
psychology. New York: Basic Books.
Stiles, W. B., & Shapiro, D. A. (1989). Abuse of the drug metaphor in psychotherapy process-outcome
research. Clinical Psychology Review, 9, 521-543.
Sullivan, P. E, Neale, M. C, & Kendler, K. S. (2000). The genetic epidemiology of major depression:
Review and meta-analysis. American Journal of Psychiatry, 157, 1552-1562.
Teasdale, J. D., Segal, Z. v.. Williams, J. M. G., Ridgeway, V. A." Soulsby, 1. M., & Lau, M. A. (2000).
Prevention of relapse/recurrence in major depression by mindfulness based cognitive therapy. Journal
(if Consulting and Clinical Psychology, 68, 615-623.
Thapar, A., Harold, G., & Mcfluffin, P. (1998). Life events and depressive symptoms in childhood - shared
genes or shared adversity'!: A research note. Journal of Child Psychology and Psychiatry, 39, 1153-
1158.
Thompson-Brenner, H., & Westen, D. (2004). A naturalistic study of psychotherapy for bulimia nervosa:
Comorbidity, outcome, and therapeutic interventions in the community. Unpublished manuscript,
Boston University, Boston.
Tse, W. S., & Bond, A. J. (2004). The impact of depression on social skills. A review. Journal of Nervous
and Mental Disease, 192,206-268.
Tsigos, C., & Chrousos, G. P. (2002). Hypothalamic-pituitary-adrenal axis, neuroendocrine factors and
stress. Journal of Psychosomatic Research, 53, 865-871.
Uchino, B. N., Cacioppo, J. T., & Kiecolt-Glaser, J. K. (1996). The relationship between social support and
physiological processes: A review with emphasis on underlying mechanisms and implications for
health. Psychological Bulletin, 119,488-531.
Van Houdenhove, B., & Egle, U. T. (2004). Fibromyalgia: A stress disorder'! Piecing the biopsychosocial
puzzle together. Psychotherapy and Psychosomatics, 73,267-275.
Van as, 1., & Jones, B. P. (1999). Early risk-factors and adult person-environment relationships in affective
disorders. Psychological Medicine, 29, 1055-1067.
Van as, 1., Park, S. B., & Jones, P. B. (2001). Neuroticism, life events and mental health: Evidence for
person-environment correlation. British Journal of Psychiatry, 178 (SuppI40), s72-s77.
Van Praag, H. M., de Kloet, R., & van as, J. (2004). Stress, the brain and depression. Cambridge:

283
 Cambridge University Press.
Wachtel, P. L. (1997). Psychoanalysis, behavior therapy, and the relational world. Washington, DC:
American Psychological Association.
Waddington, L. (2002). The therapy relationship in cognitive therapy: A review. Behavioural and Cognitive
Psychotherapy, 3D, 179-191.
Wampold, B. E. (\ 997). Methodological problems in identifying efficacious psychotherapies.
Psychotherapy Research, 7, 21-43.
Weissman, M., Wolk, S., Goldstein, R. B., Moreau, D., Adams, P., Greenwald, S., Klier, C. M., Ryan, N.
D., Dahl, R. E., & Wickramaratne, P. (1999a). Depressed adolescents grown up. Journal of the
American Medical Association, 281, 1707-1713.
Weissman, M., Wolk, S., Wickramaratne, P., Goldstein, R. 8., Adams, P., Greenwald, S., Ryan, N. D.. Dahl,
R. E., & Steinberg, D. (1999b). Children with prepubertal-onset major depressive disorder and anxiety
grown up. Archives of General Psychiatry. 56, 794-801.
Westen, D. (2000). Integrative psychotherapy: Integrating psychodynamic and cognitive-behavioral t ieory
and technique. In C. R. Snyder & R. E. Ingram (Eds.). Handbook of psychological change (pp. 217-
242). New York: Wiley.
Westen, D. (2002). Manual for the Social Cognition and Object Relations Scales (SCaRS). Emory
University, Atlanta, GA.
Westen, D., & Morrison, K. (2001). A multidimensional meta-analysis of treatments for depression, panic,
and generalized anxiety disorder: An empirical examination of the status of empirically supported
therapies. Journal of Consulting and Clinical Psychology, 69, 875-899.
Westen, D.. Novotny, C. M.. & Thompson-Brenner, H. (2004a). The empirical status of empirically
supported psychotherapies: Assumptions, findings, and reporting in controlled clinical trials.
Psychological Bulletin, 130, 631-663.
Westen, D., Novotny, C. M.. & Thompson-Brenner, H. (2004b). The next generation of psychotherapy
research: Reply to Ablon and Marci (2004), Goldfried and Eubanks-Carter (2004), and Haaga (2004).
Psychological Bulletin, 130,677-683.
Westen, D., & Shedler, 1. (2000). A prototype matching approach to diagnosing personality disorders:
Toward DSM- V, Journal (!f Personality Disorders, 14, 109-126.
Widiger, T, A., & Anderson, K. G. (2003). Personality and depression in women, Journal of Affective
Disorders, 74, 59-66.
Young, 1. E. (\999). Cognitive therapy for personality disorders: A schema-focused approach (3th ed.).
Sarasota, FL: Professional Resource Exchange.
Zuroff, D. C; & Blatt, S, J. (2002). Vicissitudes of life after the short-term treatment of depression: Roles
of stress, social support, and personality. Journal ojSocial and Clinical Psychology, 21, 473-496.
Zuroff, D. C; Blatt, S. J., Krupnick, J. L., & Sotsky, S. M. (2003). Enhanced adaptive capacities after brief
treatment for depression. Psychotherapy Research, 13, 99-115.
Zuroff, D. C., Mongrain, M., & Santor, D. (2004). Conceptualizing and measuring personality vulnerability
to depression: Comment on Coyne and Whiffen (1995). Psychological Bulletin, 130,489-511.

284
 Author Index

Abraham, K., 197-200 Eelen, P., 10, 53, 262

Aube, J. A., III, 148 Egle, U. T., 258
Auerbach,1., 154,260-261 Erikson, E. H., 138

Bagby, R M., 75, 96, 98, 101, 120 Fazaa, N., 145
Bateman, A., 262, 271, 275 Fairbairn, W. R. D., 200, 211
Beck, A. T., 4, 8, 10, 44, 60, 67-69, 72- Fichman, L., 98, 104, 108, III, 113,
83, 85-90, 95-98, 100-10 I, 103-104, 149,174
108-110,112-115,117-123,142, Fieve, R R., 119
145,173,261,264,273 Firth-Cozens, 1., 100-102
Bemporad, 1. R., 4, 68, 76-86, 90, 121- Fitzpatrick, D., 111-112, 145, 148-149
122, 142,275 Flett, G. L., 75, 87, 96, 105-113, 120,
Besser, A., 103, 108, III, 113, 141, 147 145
Beutel, M. E., 13, 141,263 Fonagy, P., 13, 118, 150, 193-194,209,
Beutler, L. E., 84, 123, 270-271, 274, 217,262,271-272,275
277 Fontaine, J. R J., 96, 114
Bibring, E., 88, 141,204,206 Frank, S. 1., 101, 141
Blankstein, K. R, 104-105, 108, 145, Freud, A., 190-193
147 Freud, S., 80, 87, 122, 139-141, 151,
Blatt, S. J., 2, 8-9, 10-11,60,67-71,75, 194,196-199,217,219,243
82-85, 95, 103-104, 110, 115-118, Friedhoff, A. J., 104
138-139, 143, 153, 164, 168, 170- Friedman, M. A., III, 113
174, 183, 194, 200, 205-212, 218, Frost, R. 0., 96, 121
220, 255, 257-258, 260-263, 267,
271-272,274-277 Gabbard, G. 0., 272
Bornstein,R.E,31,97, 107, III, 149 Gallagher, L. E, 146
Bowlby, J., 4, 68, 76-81, 85-86,90,121, Gitlin, M., 104-105, 109
142, 193,263 Goodman, S. H., 5, 140, 164, 167-168,
Brown, G. w., 263 176-184
Gotlib, I. H., 5, 46, 96, 99, 112, 140,
Cicchetti, D., 164, 166, 169, 175-179, 142, 164, 167-168, 174, 176, 178-
189-190,205,217-218,259 184,260
Claes, S. J., II, 238-239, 260, 266 Gunderson.J, G., 8,68
Clarkin, 1. E, 13, 84, 123, 270, 277
Cluckers, G., 11, 213 Hammen, C. L., 55, 98, 100, 104-106,
Corveleyn, 1., 2, 3, 6, 10, 13,67, 114 109,147,165,172,174,263-264
Coyne, J. c, 75, 87, 105, 108-109, 147, Harris, T. 0., 112,263
263,270 Hermans, D., 10, 53-54, 58, 262, 264,
267,275
D' Afflitti, 1. P., 8,96, 140,207 Hewitt, P. L., 75, 96, 105, 108, 110, 120
De Fruyt, J., 9, 254
Demyttenaere, K., 9, 26, 36, 120,254 Igreja, J., 10I

285
Joffe, W. G., 88 Nietzel , M. T., 98, 105, 144
Joiner, T. E., 101, 104, 107, 113, 147,
263 Ormel, J., 254, 258, 265, 267
Ouimette, P. c, 98-10 I
Kandel, E., 29, 89, 266, 272 Overholser, J. C., 98, 100-10 1
Kazdin, A., 219
Kendler, K. S., 18, 27-28 , 35-36, 228, Parke~G. ,6,98, 103,254,256-258,271
254,257,265-266,277 Peselow, E. D., 119-120
Kessler, R., 5, 43, 78, 103.227,254,265 Priel, 8., 101, 103-106, 113, 141, 144,
Klein, D. N., 6, 67, 98, 99, 100, 101, 146-147 , 149
102, 146, 258, 260
Klein, M., 197-20[ , 203 Quinlan, D. M., 8, 71, 80, 96, 98, 115-
Koestner, R., 98, 101, 104, 108, III, 116, 119, 121, 140-141, 143, 145,
149, 174 148,173
Krupnick, J. L., 106. 116, 122, 148, 263,
272 Raes, E, 53, 59
Kuperminc, G. P., 101-102, 143-144, Robins, C. J., 8, 68, 76, 78, 96, 98·99,
146,183 101, 104-106, III. 119, 123, 142,
Kutcher, G. S., 96 147
Kwon, P., 8, 67, 96, 105.275 Rutter, M., 88. 110, 171, 205, 265, 268

Leadbeater, B. J., 101-102, 143-144, Sabbe, B., 68, 257

146,183 Safran, J. D., 85-86 , 114,264
Lehman, A. K., 98 Sanfil ipo, M. P., 119
Levy, K. N., 70, 73-74, 77, 89, 257-260, Sanislow, C. A., 7, 99, 115-116, 145,
267-277 148,273
Lewinsohn, P. M., 169, 173-174 , 258, Santor, D., 70, 98, 105, 114, 149,257
260 Segal, Z. v.. 5-6, 10, 55, 62-63, 99-10 I,
Luyten , P., 6, 9-10, 13,67-68.85,96,99, 104,117,120,273,275-276
114-141 Shahar, G., 70, 72,82,96,101-102,104-
107, 113, 116-119, 141, 143-153,
Mahler, M., 3, 200, 202 263,270-274
Maroudas, c ., 8,67,69,71 ,73,77,79- Shea, M. T., 6-8, 67-68, 115-116 , 145
81,83-85,87-89, 116, 142 Shedler, J., 89, 255
Meurs, P., 164,211,213 Soenens, B., 96, 114
Mongrain, M., 70, 98, 101, 104, 106, Spitz, R. A., 88. 192-193, 211, 220
109, 111-114, 143, 145, 147-149, Stern, D. N.• 183, 193,211,263
257
Monroe, S. M.• 28, 106, 108, 173, 255- Teasdale, J. D., 10, 49, 54, 62-63, 273,
256,270 275
Morrison. K. H., 6-7, 67,84,87-88,111-
112, 122, 255-256, 260, 269-270, Van Houdenhove, B., 258
272-273 Van Os, J., 255, 265
Van Oudenhove, L., 254
Nemeroff, C. 8. , 228-229, 232-233, Van Praag, H., 74, 89, 255-256, 260, 265
245,260,273 Vliegen, N., 11,211

286
Wachtel, P. L., 51, 67, 114, 149,263 Wiseman, H., III
Wallerstein, S., 1-4
Westen, D., 6-7, 13,67,74,84-89, 122, Yehuda, R., 236-237
255-256,260-262,269-272
Whiffen, V. E., 75, 87, 104-105, 108- Zohar, A. H., 143-144, 149
109, III, 113, 115, 147,270 Zuroff, D. c, 7-8, 68, 70-73, 75-78, 80-
Whisman, M. A., 105, III, 113 82, 89, 91, 96, 98-102, 104-117,
Williams, M., 7, 10,44,46,52-59,61- 120-123, 140-142, 145, 147-150,
62,273 173-174, 207, 255-257, 263, 265,
Winnicott, D. w., 198,200,209,211 270-273, 276-277

287
 Subject Index

Adolescence/adolescent, II, 54, 218, Cortisol, 29, 33, 230-245

236,259,267,273,276 Countertransference, 82-83, 122
Adolescent depression, 163-188 DAS (See Dysfunctional Attitudes
Adversity, 4, 89, 154, 156, 174, 260, Scale)
266,274,277 Dependency, 8, 10, 45-51, 255-262,
Aggression, 71, 76-78, 80-82, 88, III, 265-266,272-274
167, 171, 195-207,211-213 Dependent depression (See Anaclitic
Ambivalence, 80, 83, 141, 196-204, depression)
208,212 Depressed mothers, 167-168, 172-173,
Anaclitic depression 3, 68, 77, 98, 139, 179-183, 270
167, 194,205,208-211 Depression
Antidepressants 7,17,22,25,30,32-35, adolescence in (See Adolescent
121,229 depression)
Antisocial behavior, 171 atypical (See Atypical depression)
Attachment, 4, 8, 68, 72, 81, 84, 112, bipolar (See Bipolar depression)
142, 155, 167, 178, 183, 193-194, clinical (See Clinical depression)
212,242,254,260-262,268 dependent (See Anaclitic depres-
Atypical depression, 236-237, 244-245 sion)
Autonomous depression, 79, 142 dysthymic (See Dysthymic disorder)
Autonomy, 3, 8, 10-11, 60-61, 68-75, endogenous, 89, 109, 119, 264
79, 81-88, 95-123, 138-143, 156, childhood in (See Childhood depres-
206-207, 255-266, 272-274 sion)
introjective (See Introjective (self
Beck Depression Inventory (BOI), 49, critical) depression)
56, 117, 119 major (See Major Depressive
BOI (See Beck Depression Inventory) Disorder)
Bipolar depression, 235 minor, 34, 254, 257
narcissistic (See Narcissistic depres-
Childhood depression, 163-166. 169- sion)
170, 173-176, 181,213 seasonal (See Seasonal Affective
Clinical depression, 2, 52, 74, 89, 103, Disorder)
105, 109, 140, 168, 254, 257-258, sociotropic (See Sociotropic depres-
270 sion)
Cognitive-affective schernas, 4, 258- subthreshold (See Subthreshold
265, 273-277 depression)
Cognitive behavioral therapy, 6, 35, 86- postpartum (See Postpartum depres-
87, 115, 148 sion)
Compulsive caregiving, 77 relational, 206-214
Congruency hypothesis, 52, 76, 78, 81, Depressive Experiences Questionnaire
103-105,108-110,147,173-174 (DEQ), 96-97, 100, 142-152, 157,
Connectedness, 3 171,207
Corticotropin, 227-239 DEQ (See Depressive Experiences

289
 Questionnaire) Interpretation, 44, 49, 82-83, 96, 108,
Developmental psychopathology, II, 117-119,192,216,235,272,275
74,88,163-165,168,184-185,189- Introjective (self critical) depression, 60,
196, 200-205, 208-210, 216-220, 68,70,90,194,206-212
254,260 IPT (See Interpersonal therapy)
Diagnostic and Statistical Manual of
Mental Disorders (DSM), I, 2, 10, Life events, 8, 27-28, 76, 78, 81,95,98,
12, 18-19,23,25,34-37,70,74,89- 103-109, 112, 146-147, 171-173,
90,99,166,168,227,244-245,253- 184,228,245,256,258,260,265
260,267-268,270,272,276 Life stress
Drug treatment, 7, 119 current, 4, 89, 108, 277
DSM (See Diagnostic and Statistical early, 89, 228-229, 238, 240, 242-
Manual of Mental Disorders) 245,258,268
Dynamic interactionism, I, 4, 11-12, Loss, 2-3, 8, 29-33, 36, 62, 72-73, 76-
104, 106, 108, 110, 153-154, 253- 78, 81-82, 88, 112, 140-144, 147,
257,260,265,270,274,277-278 149, 154, 166, 168-169, 173
Dysfunctional Attitudes Scale (DAS),
142, 162
Major Depressive Disorder (MDD), 17-
Dysthymia (See Dysthymic Disorder)
26, 29-37, 53-54, 74, 99, 165, 172,
Dysthymic disorder, 34, 99, 168, 170-
227-229,232-245,256
172
MDD (See Major Depressive Disorcer)
Mediating model, 107-108
Efficacy factor, 141, 145-146, 149
Melancholia, 2, 104-141, 195
Endogenous depression, 89, 109, 119,
Memory (autobiographic), 10, 17, 20,
264
23,29-33,43-63,169,227,262,267
Mental representations (See also Object
Gender differences, 71, 98, 183
representations), 253, 260-263, 267,
Gene-environment correlations and
275,278
interactions, 265
Mindfulness based cognitive therapy,
Genes, 11, 238-239, 243, 265-266, 272
10,62,275
Glucocorticoids, 33, 239, 241
Mortality, 17, 20-23, 36-37, 165, 227,
Guilt, 3, 68, 70, 77-79, 82, 98, 138, 140-
234
145, 153, 166-167, 173, 198-199,
202,204,207,212
Narcissistic depression, 196, 201-202,
Helplessness, 3, 68, 76, 141, 143-144, 206,208
201-207,211 National Institute of Mental Health
Hopelessness, 3, 51, 58, 145 (NIMH), 5-8, 24-25, 27, 115, 119-
HPA axis, 33, 89, 229-245, 258-259, 122,148, 165,268-271
266,273,277 Neuroticism, 117, 228, 265-266
NIMH (See National Institute of Mental
Interpersonal relationships, 69, 73, 81, Health)
84,107,117-118,122,139-151,174,
206-207,256,262,264 Object loss, 78, 140,207
Interpersonal therapy (lPT), 6, 67, 115, Object relationship, 20 I
122,148,264,269,271,273,277 Object representations, 78, 81, 152

290
Personality, 2, 4, 6, 8, 10-11,44,67-89, Separation, 3, 70, 78, 80, 84, 102, I 12,
95-96, 99-123, 137-143, 149-156, 144, 167, 173, 191, 193, 196-197,
165, 173, 181, 203-217, 228, 235, 202,206,217,240,269
237,242-245,253-277 Shame, 79, 138, 143,204
Personality disorder, 6, 8, 67-74, 116, Social support, II, 107-108, 112-113,
139, 165, 208, 216-217, 235, 237, 118,146-147, 149,266
243,255,261,273-275 Sociotropic depression, 76, 79-80, 142
Pharmacotherapy (See also Antidepres- Sociotropy, 8, 10-11, 60-61, 255-266,
sants, Drug treatment), 8, 119-120, 272-274
123,235,243 Sociotropy-autonomy Scale (SAS), 61,
Placebo, 7,89, 115, 119, 148 96,142
Postpartum depression, 108, 182-183 Somatic symptoms, 207, 257
Precipitating events (See also Proximal Subthreshold depression, 254, 257
antecedents), 79-81 Suicide, 21-22, 52, 55, 76, 79, 153, 165,
Prior episodes of depression, 25-27, 31 197,228,233
Proximal antecedents, 150, 218 Superego, 3, 80, 88, 141, 196-197,201-
203,206,209,216
Recurrence, 24-27, 34, 43-44, 63, 116, TAT (See Thematic Apperception Test)
165,178 TDCRP (See Treatment of Depression
RDC (See Research diagnostic criteria) Collaborative Research Program)
Relapse, 5-7, 10, 24-26, 34-35, 43-46, Temperament, 88, 120, 123, 167,
51, 54, 56, 58, 60-63, 84, 87, 102, 171,176, 178, 183, 257, 260, 265,
115-116, 122, 170, 176, 183-184, 274
227,234,256,261-264,270-276 Thematic Apperception Test (TAT), 154,
Relapse prevention, 10, 44, 61, 63 156
Research diagnostic criteria (RDC), 119 Therapeutic relationship, 82-83, 87,
Riggs-Yale study, 118 115-116, 118-119, 122, 148, 164,
214,262,264,274-277
SAD (See Seasonal Affective Disorder) Therapist factor, 123
SAS (See Sociotropy-autonomy Scale) Transference, 82-83, 115, 118, 121-122,
Seasonal Affective Disorder (SAD), 56, 264,275
236 Transdiagnostic, 12,253-254,258,260,
Schizophrenia, 70, 208, 210, 239 267
Scar effects, 102-103 Treatment of depression, I, 5-7, 44, 63,
Scar hypothesis, 30, 100 67-69, 84, 101, 110, 115, 121-123,
Self-critical depression (See Introjec- 140, 142, 148, 253-254, 263-264,
tive depression) 267,269,271-272,275-278
Self-critical perfectionism, 8, 10-11, 68, brief, 6-8, 67, 115, 117, 120, 122, 150,
70, 75, 79-82, 95-123, 141, 148, 255, 271-275
255-266, 272-274 long-term, 63, 67, 84, 271, 275-276
Self-definition, 3, 11,69, 73, 83-84, 87, Treatment of Depression Collaborative
97,138-139,142-146,151-152,156- Research Program (TDCRP), 6-7,
157,205-207,274 101,115-116,120-122,148,269

291