Autobiographical memory in the euthymic phase of recurrent depression.
ABSTRACT The authors investigated autobiographical memory specificity in subjects who formerly had depression. In 122 euthymic patients with at least two previous major depressive episodes, memory specificity was significantly impaired compared to matched control participants but not related to residual symptoms and illness characteristics, was not differentially affected by cognitive therapy, and was also not predictive of relapse/recurrence during the 2-year follow-up. However, memory specificity was associated with age, education, and immediate and delayed memory recall. The results suggest that memory specificity may reflect a global cognitive impairment that remains in patients who (formerly) had depression but does not constitute a trait marker for vulnerability for relapse/recurrence.
- [Show abstract] [Hide abstract]
ABSTRACT: This study investigated whether overgeneral autobiographical memory (OGM) predicts the course of symptoms of depression and anxiety in a community sample, after 5, 6, 12 and 18 months. Participants (N = 156) completed the Autobiographical Memory Test and the Depression Anxiety Stress Scales-21 (DASS-21) at baseline and were subsequently reassessed using the DASS-21 at four time points over a period of 18 months. Using latent growth curve modelling, we found that OGM was associated with a linear increase in depression. We were unable to detect changes over time in anxiety. OGM may be an important marker to identify people at risk for depression in the future, but more research is needed with anxiety.Cognition and Emotion 01/2014; · 2.52 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: Depression is a debilitating mental health problem that tends to run a chronic, recurrent course. Even when effectively treated, relapse and recurrence rates remain high. Accordingly, interventions need to focus not only on symptom reduction, but also on reducing the risk of relapse by targeting depression-related disturbances that persist into remission. We are addressing this need by investigating the efficacy, acceptability and feasibility of a MEmory Specificity Training (MEST) programme, which directly targets an enduring cognitive marker of depression - reduced autobiographical memory specificity. Promising pilot data suggest that training memory specificity ameliorates this disturbance and reduces depressive symptoms. A larger, controlled trial is now needed to examine the efficacy of MEST. This trial compares MEST to an education and support (ES) group, with an embedded mechanism study.Trials. 07/2014; 15(1):293.
- [Show abstract] [Hide abstract]
ABSTRACT: Depressed individuals have difficulty remembering specific autobiographical events. These deficits often persist after recovery of mood symptoms, but the mechanisms underlying impaired memory specificity in recovered depressed individuals remain unclear. Here, we sought to examine whether performance on two cognitive measures might be related to deficits in autobiographical memory retrieval in individuals with a history of depression. Twenty-four recovered depressed women (12 with more than one previous episode) and 24 never depressed women completed two cognitive measures (Digit Span and a Number Generation Task) and tests of autobiographical memory recall. Overall, the recovered depressed women did not show deficits in autobiographical retrieval. However, those with more than one previous episode had impaired retrieval of categorical autobiographical memories. Moreover, depression history moderated the relationship between Digit Span and retrieval of categoric autobiographical memories such that within the whole recovered depressed group (but not the never depressed group), those with lower Digit Span also had poorer retrieval of categorical autobiographical memories. Our sample size was small and included only women. Moreover, order effects may have been a significant factor. These findings support the notion that working memory is an important factor in impairing autobiographical memory in those who have recovered from depression, but suggest a complex relationship with autobiographical recall.Journal of Behavior Therapy and Experimental Psychiatry 12/2013; 45(2):260-266. · 2.23 Impact Factor
Autobiographical Memory in the Euthymic Phase of Recurrent Depression
Claudi L. H. Bockting, Aart H. Schene, and
Maarten W. J. Koeter
University of Amsterdam
Elizabeth M. Wekking
Leiden University and University of Amsterdam
J. Mark G. Williams
The DELTA Study Group
University of Amsterdam
The authors investigated autobiographical memory specificity in subjects who formerly had depression.
In 122 euthymic patients with at least two previous major depressive episodes, memory specificity was
significantly impaired compared to matched control participants but not related to residual symptoms and
illness characteristics, was not differentially affected by cognitive therapy, and was also not predictive of
relapse/recurrence during the 2-year follow-up. However, memory specificity was associated with age,
education, and immediate and delayed memory recall. The results suggest that memory specificity may
reflect a global cognitive impairment that remains in patients who (formerly) had depression but does not
constitute a trait marker for vulnerability for relapse/recurrence.
Keywords: autobiographical memory, depression, relapse/recurrence, cognitive therapy, executive
Autobiographical memory can be seen as a type of declara-
tive memory (Squire, 1995) and more specifically as a form of
episodic memory (Tulving, 1972). Autobiographical memory
will be defined here as declarative, explicit memory for a
specific event in time and place with experiential awareness of
self in the experience (“autonoesis”) (Wheeler, Stuss, & Tulv-
ing, 1997). It has been found that autobiographical memory
performance in patients with a depressive disorder is impaired
in comparison to control individuals without depression. More
specifically, patients with depression are characterized by an
inability to recall specific individually experienced events and
answer with overgeneral or categoric descriptions of situations
instead (see van Vreeswijk & de Wilde, 2004; Williams, 1995,
for a meta-analysis or review of studies).
Up till now only a few relatively small studies have assessed
autobiographical memory performance in patients in remission
from recurrent depression in comparison with patients with de-
pression (Nandrino, Pezard, Poste, Reveillere, & Beaune, 2002;
Park, Goodyer, & Teasdale, 2002) or normal control subjects
(Mackinger, Pachinger, Leibetseder, & Fartacek, 2000; Park et al.,
2002). The study results suggest that autobiographical memory
impairment persists after recovery and is also characteristic of
patients in remission.
Because it is rather uncommon to find a cognitive marker for
vulnerability for depression without stress/mood induction, it
seems worthwhile to replicate these findings in a larger and
more representative group of euthymic patients with recurrent
depression. Such a replication would have potential clinical
significance. Although it is possible that autobiographical mem-
ory impairment only represents a cognitive epiphenomenon
associated with past or present depression, there are some data
suggesting that it may constitute a factor involved in the onset
Philip Spinhoven, Departments of Psychology and Psychiatry, Leiden
University, Leiden, the Netherlands; Claudi L. H. Bockting, Aart H.
Schene, Maarten W. J. Koeter, and the DELTA Study Group, Department
of Psychiatry, Academic Medical Center, University of Amsterdam, Am-
sterdam, the Netherlands; Elizabeth M. Wekking, Department of Psychol-
ogy, Leiden University, Leiden, the Netherlands, and Department of Psy-
chiatry, Academic Medical Center, University of Amsterdam, Amsterdam,
the Netherlands; J. Mark G. Williams, Department of Psychiatry, Oxford
University, Oxford, England.
This study was supported by grants from the Health Research Devel-
opment Counsel (ZON), the Department Prevention Program, and the
National Foundation for Mental Health (NFGV).
We are most grateful to the participants in our study. In addition, we
express our appreciation to the staff at the participating psychiatric sites for
their recruitment efforts and to the following therapists for conducting the
cognitive therapy: Willemijn Scholten, Swanny Wierenga, Mieke van der
Rijken, Ruud Kol, Birgitta Schalken, Els Loeb, Vera van der Kraan, and
Annemarie Fouwels. We also thank our interviewers and independent
raters and specifically Irene Visch for assistance with data management
and support and Anoek Maan for assistance with data analysis. Finally, we
thank Ismay Kremers and Rimke Haringsma for collecting the Autobio-
graphical Memory Test (AMT) data in control subjects and Chantal van
Duyn, Kim Beukenhorst, and Martin de Pruyssenaere for scoring the AMT
data. The following colleagues contributed to the DELTA study: Hanneke
Assies, Claudi Bockting, Jochanan Huyser, Maarten Koeter, Anja Lok,
Guido Nabarro, Aart Schene, Philip Spinhoven, Ieke Visser, Elly Wekking,
and Luuk Wouters.
Correspondence concerning this article should be addressed to Philip
senaarseweg 52, 2333 AK Leiden, the Netherlands. E-mail: spinhoven@
Journal of Abnormal Psychology
2006, Vol. 115, No. 3, 590–600
Copyright 2006 by the American Psychological Association
or recurrence of a depressive episode. In several studies it has
been found that autobiographical memory impairment is asso-
ciated with poor problem-solving skills (e.g., Goddard,
Dritschel, & Burton, 1996) and also with difficulties in imag-
ining specific situations in the future (e.g., Williams et al.,
1996). It is conceivable that in daily life situations stress-
ful events or mood changes will more easily induce a depres-
sive episode when these cognitive coping skills are im-
paired. Evidence for an association of lack of memory
specificity with known clinical predictors of relapse/recur-
rence (such as number of prior depressive episodes, residual
or subthreshold symptoms, age of onset, and months of recov-
ery since the last episode; Ormel, Oldehinkel, & Vollebergh,
2004) would yield support for the supposition that autobio-
graphical memory specificity constitutes a mechanism involved
in mediating relapse/recurrence in patients with recurrent
Another question in need of further exploration is whether
autobiographical memory impairment reflects a broader deficit
in cognitive functioning. Such an association has been studied
in older people (Phillips & Williams, 1997; Sampson, Kinder-
man, Watts, & Sembi, 2003), adolescents with major depression
(de Decker, Hermans, Raes, & Eelen, 2003; Park et al., 2002),
patients with traumatic brain injury (Williams, Williams, &
Ghadiali, 1998), patients with a borderline personality disorder
(Arntz, Meeren, & Wessel, 2002), and normal individuals (Wil-
liams & Dritschel, 1992). In most of these studies, a moderately
strong relationship of memory specificity with general cogni-
tive abilities (e.g., semantic fluency) and memory functioning
(e.g., immediate and delayed recall of verbal material) has been
found. Because memory impairment is consistently found in
individuals with clinical depression versus those without de-
pression (Burt, Zembar, & Niederehe, 1995) and seems to
remain stable over time (Paradiso, Lamberty, Garvey, & Rob-
inson, 1997), it is especially worthwhile to investigate whether
overgeneral autobiographical memory in patients who formerly
had depression reflects a broader deficit in executive control or
In summary, the first part of the present study was cross-
sectional, and we investigated whether specificity of autobio-
graphical memory in euthymic patients with recurrent depres-
sion in the past is impaired in comparison with healthy control
individuals without depression. It was hypothesized that spec-
ificity of autobiographical memory in this group would be
lower than that of normal control individuals. Moreover, the
association of autobiographical memory specificity with clini-
cal variables predictive of relapse/recurrence was investigated.
It was hypothesized that reduced levels of memory specificity
would be found especially in patients with a higher number of
previous depressive episodes. Finally, we also investigated
whether lack of memory specificity reflected a more global
cognitive impairment as often found in patients who (formerly)
had depression. It was predicted that reduced memory specific-
ity would be especially associated with impairments in memory
retrieval. In the second part of the study, a prospective study
that assessed the effect of group cognitive therapy on memory
specificity and the predictive value of memory specificity on
relapse/recurrence is described.
Participants were recruited as part of a randomized, controlled trial of the
preventive effect of a group cognitive therapy on relapse/recurrence of
depression in a group of patients at high risk for recurrence in comparison
with treatment as usual (Bockting et al., 2005). To be eligible individuals
had to meet the following criteria: (a) at least two major depressive
episodes in the last 5 years, as defined according to the Diagnostic and
Statistical Manual of Mental Disorders (4th ed.; DSM–IV; American
Psychiatric Association, 1994) and assessed by the Structured Clinical
Interview for DSM–IV (SCID-I; First, Spitzer, Gibbon, & Williams, 1997)
by trained evaluators; (b) current remission status, according to DSM–IV
criteria, for longer than 10 weeks and no longer than 2 years ago; (c) a
current score on the Hamilton Rating Scale for Depression (Hamilton,
1960) of ?10. Exclusion criteria were current mania or hypomania or a
history of bipolar illness, any psychotic disorder (current and previous),
organic brain damage, current alcohol or drug abuse or dependence,
predominant anxiety disorder, recent electroconvulsive therapy, recent
cognitive treatment or receiving cognitive therapy at the start of the study,
or current psychotherapy with a frequency of more than 2 times a month.
There was no restriction in using pharmacotherapy.
Clinical participants were recruited at psychiatric centers and through media
announcement in the Netherlands. After complete description of the study to
protocol was approved by the institutional ethics review committees.
The healthy control group was recruited through advertisements in a
local newspaper, in community centers, and on the Internet. Participants
with a DSM–IV Axis I disorder as determined by a shortened version of the
SCID-I, the Mini International Neuropsychiatric Interview (MINI; Sheehan
et al., 1998), or a previous depressive episode were excluded. Moreover,
because trauma exposure may be associated with memory specificity
(Hermans et al., 2004) participants reporting severe childhood sexual or
physical abuse were also excluded. Control participants were comparable
with the clinical participants on age, education level, and gender ratio (see
Autobiographical Memory Test (AMT).
Pitman (1995) modified the autobiographical memory paradigm, which
was used by Williams and Broadbent (1986). In this modified version,
respondents are asked to mention a specific moment at which they exhib-
ited a trait that is written on a card (e.g., “guilty” or “helpful”). A specific
memory is defined as a memory that refers to a particular event in the past
that happened on one particular day, lasting no longer than 1 day. At study
entry the following positive (?) and negative (?) words were presented:
friendly (?), guilty (?), impolite (?), honest (?), helpful (?), jealous
(?), intelligent (?), selfish (?), humorous (?), and hostile (?). Words
were read aloud and at the same time were shown on a card. The cue words
were presented after three practice words. The task was started only after
it was clear during the practice phase that the respondent understood the
purpose of the task. Respondents were allowed 60 seconds to remember an
event and if they did not recall one, the next word was presented.
Answers were recorded on audiotape. The following scoring categories
were used: (a) specific, (b) categoric, (c) extended, and (d) no memory/
omission. Category (a) was scored when the respondent’s first memory
referred to a particular event on one particular day. Category (b) was scored
when the memory referred to repeated events. Category (c) was scored
when the remembered event lasted longer than 1 day. Category (d) was
scored when the respondent did not mention a memory or did not respond.
One trained research assistant who was blind for treatment condition
McNally, Lasko, Macklin, and
scored all the tapes. Interrater agreement on about 50% of the retrieved
memories (n ? 610) indicated good reliability, 87.8% agreement, ? ?
0.77, comparable with previous studies.1
Severity of depressive symptoms.
for Depression (HRSD; Hamilton, 1960) and the 21-item Beck Depression
Inventory (BDI; Beck, Rush, Shaw, & Emery, 1979) were used to assess
patients’ levels of depressive symptomatology.
The Stroop Color–Word Test (Hammes,
1978) consists of three separate trials (reading color words, naming colors,
and naming the print of the color words, disregarding their verbal content).
In this third trial, the decrease in color-naming speed when presented with
colored words in nonmatching colored inks as compared with the color-
naming condition is called the color–word interference effect and is con-
sidered to be a measure of selective attention. The Digit Span test is a
subtest of the Wechsler Adult Intelligence Scale (Wechsler, 1981). The
subtest consists of two parts and requires the subject to repeat digits
forward and backward. The test is considered to be a measure of immediate
memory (forward) and working memory (backward). The Story Recall
(logical memory) subtest of the Rivermead Behavioral Memory Tests
(RBMT; Wilson, Cockburn, & Baddeley, 1985) aims to detect impairment
of everyday memory functioning by mimicking the demands made on
memory by normal daily life. The subject is asked to remember a short
passage, and the task consists of both an immediate and delayed recall
component. The California Verbal Learning Test (CVLT; Delis, Kramer,
Kaplan, & Ober, 1987) assesses multiple aspects of verbal learning by the
use of two shopping lists presented as the “Monday” and “Tuesday” list.
The CVLT measures, among other things, immediate and delayed recall 20
minutes after presentation of the word list. The selected tests are interna-
tionally accepted (Lezak, Howieson, & Lorin, 2004) and frequently used in
clinical neuropsychological studies measuring attention and memory.
The 17-item Hamilton Rating Scale
After telephone screening on inclusion and exclusion criteria, patients
were administered the SCID-I and HRSD. Patients who met inclusion
criteria and no exclusion criteria were tested before random assignment to
treatment condition and the start of therapy. Participants completed first the
AMT and were subsequently neuropsychologically tested. After the pur-
pose and procedures of the study were explained by phone, control par-
ticipants were invited for a single assessment session. Participants who did
not meet exclusion criteria as assessed by the MINI and some additional
questions on previous trauma and previous depressive episodes subse-
quently completed the AMT.
A total of 122 clinical subjects completed the assessment bat-
tery. The mean age of the patients was 44.7 years (SD ? 9.0), and
75.4% of the sample were women. Nineteen percent of the subjects
reported two previous depressive episodes in the past, 25.9%
reported three previous episodes, 27.6% reported four to six pre-
vious episodes, and 27.6% reported six or more previous episodes.
The prevalence of Axis I disorders in this euthymic group of
patients was quite low [one or more psychiatric disorders as
assessed with the SCID-I (First et al., 1997) were seen in 12.3% of
the patients: social phobia (n ? 7), panic disorder (n ? 4), eating
disorder (n ? 3), posttraumatic stress disorder (n ? 1), obsessive–
compulsive disorder (n ? 1), generalized anxiety disorder (n ? 1),
and specific phobia (n ? 1)]. The control group consisted of 37
participants and did not differ from the clinical group with respect
to age, education level, and gender ratio (see Table 1).
Differences in Number of Specific Memories Between
Euthymic Patients and Normal Control Participants
To investigate whether euthymic patients and normal control
participants produced different numbers of specific memories and
whether there was a different response pattern to positive and
negative cue words, a mixed-model 2 ? 2 analysis of variance was
conducted with Group (euthymic or normal) as the between-
subjects variable and Cue Type (positive or negative) as a within-
subjects variable. Age and level of education were entered as
covariates. Only a significant main effect for group was found,
F(1, 155) ? 11,628, p ? .001. Euthymic patients who previously
had depression reported significantly fewer specific memories
(M ? 5.3; SD ? 2.8) than normal control participants (M ? 7.3;
SD ? 1.8) irrespective of cue type (see Table 1). Although ad-
justed mean scores were used in the statistical analyses, the values
of nonadjusted mean scores for memory specificity are presented
in the text and Table 1 for the sake of comparison with results of
previous studies on this topic. No further significant main or
interaction effects were found.
Relationships of Specific Memories With Demographic
and Clinical Variables
In the clinical group, the association of specificity of autobio-
graphical memory with demographic and clinical risk factors was
further assessed with t tests for independent groups, analysis of
variance, or Pearson correlation coefficients if appropriate (see
Table 2). Of the demographic variables only age and level of
education were significantly correlated with the number of specific
(positive and negative) autobiographical memories with younger
and higher educated patients retrieving more specific memories.
Of the clinical risk factors (i.e., BDI and HRSD scores, age of first
onset, duration of illness, number of previous depressive episodes,
duration of last depressive episode, months of recovery since last
depressive episode, use of antidepressants (yes/no), and type of
current treatment), none was significantly correlated with memory
Relationship of Specific Memories With
In the clinical group the association of specificity of autobio-
graphical memory with various neuropsychological measures was
assessed with Pearson correlation coefficients (see Table 3). As
can be seen, the number of specific (positive and negative) mem-
ories showed a significant and positive relationship with scores for
1In the present study the number of specific memories in response to
positive or negative cues on the AMT was used to operationalize memory
specificity. This method of scoring is in accordance with most studies
conducted up to now (see van Vreeswijk et al., 2004, for a review).
Analyzing the results of the present study with the percentage of specific
positive or negative memories, taking into account the number of omis-
sions or the number of times subjects did not comply with the task,
essentially yielded the same results. Moreover, on the basis of coding
autobiographical memories in response to positive or negative cues on the
AMT as categoric or not, a comparable pattern of results was obtained,
suggesting that a reduced number of specific memories and an increased
number of categoric memories are functionally equivalent.
SPINHOVEN ET AL.
both immediate recall and delayed recall as measured with the
RBMT and CVLT. Scores for focused attention (Stroop Color–
Word Test), immediate memory for abstract digits (Digit Span:
Forwards) and working memory (Digit Span: Backwards) were not
significantly associated with scores for memory specificity.
Next, partial correlation coefficients were calculated to remove
the correlation that is due to their mutual association with the use
of antidepressants (coded as a dummy variable 0/1) and the sever-
ity of residual depressive complaints (BDI and HRSD scores) as
possible confounding factors. Results showed that all significant
and positive relationships between neuropsychological measure-
ments and number of specific (positive and negative) memories
were retained. Moreover, partial correlations were calculated to
investigate whether the association of RBMT and CVLT scores
with the number of specific (positive and negative) memories was
accounted for by age and education level. These analyses showed
that the total number of specific memories and the total number of
specific negative memories remained significantly and positively
correlated with immediate and delayed recall as measured with the
RBMT and CVLT.
Part 1 had three aims and each of these is considered in turn
in the light of the results. The first aim was to test the hypoth-
esis that the memory specificity of euthymic patients with
recurrent depressive episodes would be lower than that of
normal control participants. The results supported this hypoth-
esis even after statistically correcting for age and level of
education. Moreover, the mean number of specific autobio-
graphical memories (M ? 5.3; SD ? 2.8) is very comparable to
that of inpatients with depression (M ? 5.2; SD ? 2.2) (Her-
mans et al., 2004) and outpatients with depression (M ? 5.4;
SD ? 2.6) (Kremers, Spinhoven, & Van der Does, 2004). These
data suggest that autobiographical memory impairment is not a
cognitive epiphenomenon of a depressed state and is a charac-
teristic of patients in remission.
The second aim of the study was to investigate the associa-
tion of autobiographical memory impairment with several clin-
ical risk factors for relapse/recurrence. Contrary to our expec-
tations, none of the relationships with clinical characteristics
proved to be significant. Because only euthymic patients with at
least two depressive episodes were included, we cannot exclude
the possibility that different results would have been obtained in
patients in remission from a first episode. Results of the studies
of Park et al. (2002) in adolescents with a first depressive
episode indicate, however, that already after a first depressive
episode memory specificity in patients in remission is compa-
rable to that of adolescents with depression (however, see
Nandrino et al., 2002, for slightly different findings). Taken
together, our results seem to imply that although autobiograph-
ical memory impairment can be observed in patients in remis-
sion, it is not associated with known clinical risk factors for
relapse/recurrence, such as number of previous episodes and
level of residual depressive symptoms.
The relationship of autobiographical memory with general
cognitive abilities and other memory tests was also explored.
Comparable with most previous studies, a moderately strong
association of memory specificity with education as a proxy for
intelligence (Arntz et al., 2002; de Decker et al., 2003; Wessel,
Meeren, Peeters, Arntz, & Merckelbach, 2001) was found.
These results concur with those of previous studies showing a
consistent association of memory specificity with general intel-
ligence (Park et al., 2002; Sampson et al., 2003; Wessel, Mer-
ckelbach, & Dekkers, 2002; Williams et al., 1998). Moreover, a
relationship was observed with indices of immediate and de-
layed memory retrieval, which remained significant over and
above the effect of age, educational level, use of antidepres-
sants, and level of residual depressive symptoms on autobio-
graphical memory. These findings are in agreement with results
of other studies (Phillips & Williams, 1997; Williams et al.,
1998; Winthorpe & Rabbitt, 1988), indicating that indices for
executive functioning and memory recall are related to auto-
biographical memory specificity independent of age and gen-
Demographic Data and Number of Specific Memories on the AMT in Euthymic (n ? 122) and
Healthy Control Subjects (n ? 37)
tpM SDM SD
Specific positive memories
Specific negative memories
AMT ? Autobiographical Memory Test.
bEducation level was divided in three categories with respect to the Dutch educational
eral intelligence (but see Wessel et al., 2002, for contradictory
results), which suggests that impaired memory specificity can
be partly accounted for by a more general neurocognitive deficit
of patients with previous depression as was also suggested by
some authors (Paradiso et al., 1997).
Brittlebank, Scott, Williams, and Ferrier. (1993) were the first to
demonstrate that in a group of depressed psychiatric inpatients
those patients with higher levels of autobiographical memory
Relationship of Demographic and Clinical Variables With Number of Specific Memories on the
AMT (n ? 122)
Level of education (0–6)
Age of first onset
Duration of illness (years)
Months since last episode
MSDM SDM SD
Duration last episode
?2 and ?8 months
Scale for Depression.
* p ? .05.
AMT ? Autobiographical Memory Test; BDI ? Beck Depression Inventory; HRSD ? Hamilton Rating
** p ? .01. *** p ? .001.
Neuropsychological Measures and Pearson’s Correlation Coefficients With Number of Specific
Memories on the AMT (n ? 122)
Stroop: Selective attention
Digit Span (WAIS)
Story Recall (RBMT)
Shopping list (CVLT)
Rivermead Behavioral Memory Test; CVLT ? California Verbal Learning Test.
* p ? .05.** p ? .01. *** p ? .001.
AMT ? Autobiographical Memory Test; WAIS ? Wechsler Adult Intelligence Scale; RBMT ?
SPINHOVEN ET AL.
specificity as assessed at admission obtained significantly better
treatment results than those with reduced levels of memory spec-
ificity. Because scores for memory specificity did not change
during the 7-month follow-up period, the authors suggested that
autobiographical memory specificity possibly bears the quality of
a trait marker. Subsequent studies have confirmed that in subjects
with dysphoria (Mackinger et al., 2004; Mackinger, Loschin, &
Leibetseder, 2000; Svaldi & Mackinger, 2003) and outpatients
with depression (Dalgleish, Spinks, Yiend, & Kuyken, 2001;
Peeters, Wessel, Merckelbach, & Boon-Vermeeren, 2002) auto-
biographical memory predicts the course of depression after sta-
tistically controlling for initial depression levels (however, see
Brewin, Reynolds, & Tata, 1999, for contradictory data).
Findings on the valence of the stimulus words are less clear and
in patients with depression both higher levels of overgeneral
memories to positive cues (Brittlebank et al., 1993; Dalgleish et
al., 2001) as well as reduced levels of specific memories to
negative cues (Peeters et al., 2002) have been found to be predic-
tive of recovery of depression. Up to now, these studies have
shown that overgenerality in memory predicts persistence of an
existing acute depression. By contrast, there are no studies in
which autobiographical memory specificity was investigated as a
possible marker for vulnerability to a new onset (relapse/recur-
rence) of depression in individuals who formerly had depression.
Recent findings (Raes, Hermans, de Decker, Eelen, & Williams,
2003) suggest that for some people, overgenerality in memory may
serve an affect regulation function and so may be adaptive in these
Although memory specificity seems to be a relatively stable
retrieval style, specific psychological interventions have been
found to modify this cognitive style. Williams, Teasdale, Segal,
and Soulsby (2000) found that mindfulness-based cognitive ther-
apy reduced the recall of categorical memories in patients in
remission from depression. No equivalent change was found in the
treatment as usual group. Because these preliminary findings sug-
gest that psychological interventions can reduce overgeneral mem-
ories, there is a need for further studies into which psychological
interventions are suitable to enhance specific memory retrieval and
modify this possible marker of vulnerability for onset, persistence,
and recurrence of depression.
In summary, the main purposes of the prospective part of the
present study were (a) to assess the effect of additional group
cognitive therapy on memory specificity in comparison with treat-
ment as usual and (b) to investigate whether memory specificity
See Part 1 for a more complete description of the recruitment of
participants. Of the 172 randomly assigned participants who started treat-
ment as usual or additional group cognitive therapy (Bockting et al., 2005),
data on autobiographical memory at pretreatment were available for 116
participants (67.4%). There were no significant differences in demographic
and clinical characteristics or dependent measures at baseline among
participants included in the present study (n ? 116) and participants in the
original sample but not included in the present study (n ? 56), neither were
there any differences among participants in one of the two conditions of the
present study and participants in the same condition of the original sample
but not included in the present study (all p values ? .10).
From the assessment battery used in the original study, the following
measures were selected for outcome assessment in the present study.
To assess relapse/recurrence the SCID-I (First et
al., 1997) was used. At baseline and at three follow-up assessments (3, 12,
and 24 months), current and past depressive episodes were checked by
interviewers (psychologist/research assistants) blind to treatment condi-
tion. All interviews were audiotaped. Two independent experienced psy-
chiatrists who were blind to treatment condition evaluated all occasions of
participants meeting the DSM–IV criteria for major depression. In cases of
disagreement, the DSM–IV ratings of the psychiatrists were used for further
analyses. The kappa for interrater agreement between the interviewers and
psychiatrist on categorization of a relapse/recurrence or no relapse/recur-
rence was .96, indicating high agreement.
In addition to relapse/recurrence as a primary outcome measure, the
mean severity of all relapses (as assessed by the SCID-I: light, ?6
symptoms; moderate, 6–7 symptoms; severe, 8–9 symptoms), the number
of relapses/recurrences, and the duration of the depressive episodes during
the follow-up period were used as secondary outcome measures. To ac-
count for differences in follow-up time the observed number of times a
patient had a relapse or the observed duration of the depressive episodes
were also converted to number of relapses or weeks of depressive episodes,
respectively, per 2 years at risk. Moreover, it can be argued that the
pretreatment measurement of remission is randomly located in the time
period of remission, resulting in an unreliable measurement of time in
remission. Therefore, retrospective data on time in remission at baseline
and prospective data were combined by also computing the total number of
months in remission from the last depressive episode onward.
Severity of depressive symptoms.
ilton, 1960) and the 21-item BDI (Beck et al., 1979) at baseline and the
three follow-up assessments were used to analyze the trajectory of depres-
A parallel version of the AMT different from the version at
baseline was administered at 3 months of follow-up (a complete list of the
cue words, instructions, and procedure are available from the first author).
Scores on the 17-item HRSD (Ham-
With random permutated blocks by use of computer-generated cards
stored as concealed assignment codes in consecutively sealed envelopes
under the responsibility of an independent research associate, patients were
randomly assigned to (a) treatment as usual (TAU), or (b) TAU ? group
cognitive therapy (CT). Randomization was stratified by study location and
the type of aftercare (family doctor, psychiatric center, or no aftercare). The
CT in the experimental condition involved eight weekly 2-h group ses-
sions. Each CT session followed a fixed structure, with agenda setting,
review of homework, explanation of rationale of each session, and assign-
ment of homework. The CT started with identification of negative thoughts
(Session 1) and dysfunctional attitudes, (Sessions 1–3) and then proceeded
to target change (Sessions 3–7) of these attitudes using different cognitive
techniques such as Socratic questioning and identification of positive
attitudes. Moreover, patients were encouraged to practice with alternative
attitudes (Sessions 6–8). In addition, patients were asked to keep a diary
of positive experiences to enhance specific memories of positive experi-
ences instead of overgeneral memories (Sessions 4–6). Further specific
relapse/recurrence prevention strategies were formulated in the last three
sessions. TAU involved “naturalistic” care, that is, standard treatment, as
typically provided by the referring agencies, or no aftercare at all.
Using a preliminary Cox regression analysis to examine whether
the effect of additional group CT was moderated by either of the
stratification variables used in randomization, no effect of site or
type of aftercare on relapse/recurrence over 24 months was found.
Further analyses without these stratifying variables showed that
adding CT to TAU resulted in a significant increasing protective
effect with an increasing number of previous depressive episodes
experienced. For patients with five or more previous episodes
(41% of the sample), CT significantly reduced relapse/recurrence
from 71% to 46% (see Bockting et al., 2005, for a more detailed
description of the efficacy results of the trial).
Effect of Group CT on Proportion of Specific Memories
First, we investigated whether there were any statistically sig-
nificant differences for demographic and clinical characteristics or
HRSD, BDI, or AMT scores between subjects allocated to CT
(n ? 60) or TAU (n ? 56). No statistically significant differences
were observed (all p values ? .10). Of special note is the nonsig-
nificant difference in the number of specific memories at pretreat-
ment, F(1, 114) ? .001, p ? .989. Of the 60 subjects in the CT
condition, only 5 (9.3%) attended fewer than five sessions. In
analyzing whether treatment or number of previous depressive
episodes was related to memory specificity and whether there was
a different response pattern to positive and negative cue words, a
2 ? 2 x 2 mixed-model analysis of variance was conducted with
Treatment (CT or TAU) and Previous Episodes (less than five or
five or more) as between-subjects variables and Cue Type (posi-
tive or negative) as a within-subjects variable.2This analysis could
be performed for the 101 patients for which post-treatment AMT
scores were available (n ? 57 in the CT condition and n ? 44 in
the TAU condition). The main effect of treatment was not signif-
icant, F(1, 97) ? 0.049, p ? .82. The same held true for the
interaction effect of Treatment ? Cue Type, of Treatment ?
Previous Episodes, and of Treatment ? Cue Type ? Previous
Episodes. The effect size (Cohen’s d) for the difference in number
of specific memories at post-treatment between the CT and TAU
conditions was negligible (d ? 0.03). These results indicate that
additional group CT did not differentially affect memory
Autobiographical Memory as a Predictor of
Because additional group CT did not seem to influence memory
specificity, the predictive value of memory specificity on relapse/
recurrence was investigated in the total sample using a propor-
tional hazard approach to survival analysis (Cox regression) with
relapse/recurrence as the dependent variable and scores for mem-
ory specificity as independent variables.
Over the total study period of 24 months the cumulative rate for
relapse/recurrence in our sample rose to 61.2%. A Cox regression
analysis was conducted by analyzing the number of positive and
negative specific memories at baseline as predictor variables. Time
to relapse was the duration of the euthymic period from entry into
the study. This analysis did not reveal a significant effect for the
number of positive specific memories, Wald (1) ? 0.030, p ?
.863, hazard ratio ? 0.928, 95% confidence interval [CI] ?
0.397–2.167, or number of negative memories, Wald (1) ? 0.265,
p ? .607, hazard ratio ? 1.280, 95% CI ? 0.500–3.282. After the
number of previous episodes was also included in the model, only
this variable proved to be predictive of relapse/recurrence, Wald
(1) ? 5.948, p ? .015, hazard ratio ? 1.353, 95% CI ? 1.061–
1.726. The number of specific memories in patients with no
relapse/recurrence was 5.4 (SD ? 2.8) versus 5.2 (SD ? 2.7) in
patients with a relapse/recurrence.
Because autobiographical memory impairment was associated
with global cognitive impairment in the present study sample (see
Part 1 above), using Cox regression analysis we also investigated
whether any of the neuropsychological variables (i.e., immediate
and delayed recall on the RBMT and CVLT, focused attention on
the Stroop Color–Word Test, immediate memory for abstract
digits (Digit Span: Forwards), and working memory (Digit Span:
Backwards) predicted time to relapse. None of these variables
proved to be predictive of relapse/recurrence (all p values ? .10),
suggesting that cognitive impairments as observed in our sample
of patients in remission did not constitute a risk factor for
The relationship of AMT scores at pretreatment with secondary
outcome measures (i.e., time of remittance from the last episode
onward, the number of relapses/recurrences [per 2 years at risk],
the mean severity of all relapses, and the total duration of the
depressive episodes during the follow-up period [per 2 years at
risk]) were analyzed with Pearson correlation coefficients. As can
be seen in Table 4, none of the correlations of memory specificity
scores with secondary outcome measures was statistically
Next, we investigated whether the number of specific positive
and negative autobiographical memories predicted the trajectory of
depression scores by conducting mixed-effects growth curve mod-
eling using the multilevel analysis software package MLwiN 1.10
(Rasbach et al., 2004). In analyzing the repeated measurements on
the BDI from baseline (M ? 11.8; SD ? 7.5) to 3 (M ? 10.3;
SD ? 7.3), 12 (M ? 8.4; SD ? 6.8), and 24 months of follow-up
(M ? 7.8; SD ? 7.1), a quadratic model with random intercepts
and fixed linear and quadratic terms best fitted the data. Addition
of the number of specific positive or negative memories as pre-
dictors to the model did not yield significant effects (z ? ?1.40,
p ? .0808, and z ? 0.29, p ? .6141, respectively), indicating that
these variables were not able to explain the variance between
respondents in baseline BDI measurements. Cross-level interac-
tions for the number of positive and negative memories with the
linear and quadratic time variables were also added to the model,
but none of those appeared to be significant either. Hence, both
predictor variables have no influence on the average growth
In analysis of the repeated measurements on the HDRS from
baseline (M ? 3.6; SD ? 2.8) to the second (M ? 5.5; SD ? 5.7),
third (M ? 4.8; SD ? 4.1), and fourth measurements (M ? 5.1;
SD ? 4.3), a very similar quadratic model emerged. Again, adding
the number of positive (z ? ?0.85, p ? .1977) or negative
memories (z ? 0.17, p ? .5675) to the model as predictors did not
result in significant effects. Addition of the cross-level interactions
for the number of positive and negative memories with the linear
and quadratic time variables produced no significant effects either.
Thus, both predictor variables also have no influence on the
average growth trajectory.
2Because a parallel version of the AMT different from the version at
baseline was used at post-treatment, and no data on the equivalence of
these different versions are available the effect of Time was not further
SPINHOVEN ET AL.
To check whether the pattern of results was confounded by the
fact that about half of the patients had received group CT in
addition to TAU, these analyses were repeated separately for the
two conditions. The results in both groups were very similar with
none of the correlations of measures for memory specificity with
primary or secondary outcome measures reaching the level of
The prospective part of this study had two aims. The first aim
was to test the hypothesis that group CT would have a differential
effect on memory specificity in comparison with TAU. No evi-
dence for differential treatment effects was found. This result was
unexpected, because during the group training patients were ex-
plicitly asked to keep a diary of positive experiences to enhance
retrieval of specific memories of positive experiences instead of
overgeneral memories. Because compliance with homework in-
structions for diary keeping from Sessions 4 to 6 was not verified,
one possible explanation for this negative finding is that this
treatment target was not reached because of noncompliance with
homework instructions. Another possible explanation is that the
effect of simply instructing participants to retrieve specific positive
experiences is limited because participants do not know how to
achieve this goal. A training to notice elements of subjective
experience in a nonjudgmental way such as that provided in
mindfulness-based CT (Williams et al., 2000) may be necessary to
modify this retrieval process. The supposition that it is necessary
to provide subjects with an internal strategy is also consistent with
the results of experimental studies showing that the tendency to
recall categorical memories can be altered by brief experimental
manipulations such as decentering (Watkins, Teasdale, & Wil-
liams, 2000) or reducing analytical self-focus (Watkins & Teas-
dale, 2001). Simply giving homework instructions may promote
verbal–analytical thinking and insufficiently stimulate a more ex-
periential mode of processing, which may play a critical role in
reducing the tendency toward an overgeneral retrieval style.
The second aim was to test the hypothesis that memory speci-
ficity would predict relapse/recurrence, because previous studies
had focused only on the persistence of a depressive episode. This
hypothesis was not supported for this main outcome measure (i.e.,
time to relapse from entry into the study) nor for secondary
outcome measures, such as time of remission from the last episode
onward, the number of relapses/recurrences, the mean severity of
all relapses, the duration of all relapses, or the course of depressive
symptoms during the follow-up period.
What explains this pattern of data? It could be argued that lack
of memory specificity is not a deficit related to depression (“deficit
model”), but that lowered memory specificity may also constitute
a functional strategy by preventing individuals from recalling
events that might evoke painful, negative emotions (Hermans et
al., 2004; Raes et al., 2003). On the basis of such a functional
model, it could be hypothesized that lack of specificity may buffer
against the stress of daily hassles or recent life events and prolong
the period of remission. With this perspective, being less specific
may be advantageous in the euthymic phase and maladaptive in the
long term during a depressive episode (e.g., because lack of
memory specificity impedes adequate problem solving; Pollock &
Williams, 2001). In the present study, however, memory specific-
ity was unrelated to the duration of remission from entry into the
study and the last episode onward. Thus, no evidence supporting
either a deficit or a more functional model of memory specificity
was found. These findings cannot be attributed to a lack of statis-
tical power or a restriction of range in scores on relevant variables,
because our rather large well-defined patient sample showed re-
duced levels of memory specificity and also constituted a group
vulnerable for relapse/recurrence, given the fact that during the
2-year follow-up period the cumulative rate for relapse/recurrence
increased to 62%. An alternative hypothesis to account for the
present data is that reduced memory specificity constitutes one of
the persistent cognitive deficits after recovery from depression and
consequently its role in predicting relapse/recurrence may be more
limited than previously assumed. However, in the absence of
prospective studies on memory specificity as a premorbid vulner-
ability of individuals at higher risk for depression, this hypothesis
awaits further study.
The study reported here is the first to our knowledge (a) to
examine in a group of euthymic patients with a history of recurrent
depression the association of autobiographical memory impair-
Clinical Outcome Measures and Pearson’s Correlation Coefficients With Number of Specific
Memories on the AMT at Pretreatment (n ? 116)
Time of remittance from last episode
Number of relapses
Duration depressive episodes during
Mean severity of all relapses
aValues converted to number of relapses respectively weeks of depressive episodes per 2 years at risk.
AMT ? Autobiographical Memory Test.
ment with various clinical risk factors and neuropsychological
measures in a comprehensive way and (b) to investigate the
predictive value of autobiographical memory impairment for re-
lapse /recurrence. The study results suggest that patients in remis-
sion are characterized by reduced levels of memory specificity.
This memory impairment, however, does not appear to be related
to clinical risk factors for relapse/recurrence and is also not pre-
dictive of relapse/recurrence in this vulnerable patient group.
Memory specificity is associated with age, education level, and
measures for immediate and delayed memory recall. Taken to-
gether, these data suggest that autobiographical memory specific-
ity does not play a role in contributing to relapse/recurrence. These
findings also call into question the clinical utility of trying to
enhance autobiographical memory specificity in patients in remis-
sion with the purpose of preventing relapse/recurrence. Although
mindfulness-based cognitive therapy (Williams et al., 2000) seems
to be able to enhance memory specificity, it remains to be estab-
lished whether this change actually mediates the preventive effect
of this group therapy on relapse/recurrence.
Reduced memory specificity, however, does not have to be
specific for depression. In the last two decades, the retrieval of
overgeneral autobiographical memories on a cue-word task has
also been found in individuals with posttraumatic stress disorder
(McNally et al., 1995), acute stress disorder (Harvey, Bryant, &
Dang, 1998), and eating disorder (Dalgleish et al., 2003). A char-
acteristic common to all of these clinical groups is the presence of
clinically elevated levels of depression and/or a history of trau-
matic events. Moreover, memory impairments in general are also
observed in other psychiatric disorders (Burt et al., 1995). Memory
impairment might be associated with more overarching factors
common to various forms of psychopathology (e.g., severity of
illness, motivational deficits, or effortful processing deficits) (Dal-
gleish & Watts, 1990). In depression and posttraumatic stress
disorder in particular, deficits in executive functioning are prom-
inent. Executive functioning refers to cognitive processes that
control and integrate other cognitive activities such as episodic and
autobiographical memory. As has already been noted by Williams
(1995), executive dysfunction may be one of the pathways leading
to overgeneral memory retrieval. Further more comprehensive
studies into executive functioning in relation to episodic and
autobiographical memory impairments in different diagnostic
groups seem worthwhile as executive dysfunction may be one of
the components contributing to memory specificity that has been
overlooked in previous studies.
There are at least two reasons to think that the current data merit
serious consideration. First, patients with recurrent depression in a
euthymic phase form a rather unique group to study cognitive
processes unaffected by severity of symptomatology or presence
of psychiatric disorder. As a consequence, the present findings
from a large, well-defined, and homogenous group of patients with
recurrent remissions shed light on cognitive processes in patients
in remission and seem to indicate that autobiographical memory
impairment is characteristic for patients with a history of one or
more previous depressive episodes. Second, by using a prospective
design, it was possible to study the predictive value of less specific
memories for the course of the disorder during a clinically relevant
follow-up period of 24 months. This design substantiates our
conclusion that the predictive value of less specific memories for
relapse/recurrence is notably absent.
Finally, at least three limitations of this study merit consider-
ation. First, we used the AMT version of McNally et al. (1995), in
which subjects attempt to retrieve specific personal memories
exemplifying trait adjectives having either positive (e.g., friendly
or helpful) or negative valence (e.g., guilty or hostile) relevant for
self-representations in traumatized individuals. Although these
cues also seem relevant to the self-representation of patients who
previously had depression, more research of the differential effects
of different AMT cues in various clinical groups is needed. More-
over, it seems worthwhile to supplement memory cue-word tasks
such as the AMT with procedures to try to capture the occurrence
and quality of more spontaneous autobiographical memories as
typically found in depression and posttraumatic stress disorder
(Brewin, 1998). Second, in the present study patients with only one
prior depressive episode and/or those who have been euthymic for
more than 2 years were excluded. It may be possible that patients
with two or more episodes constitute a group that may be etio-
logicaly different from those with only one previous depressive
episode. In previous studies of preventive interventions in recur-
rent depression, evidence for etiologically different subgroups of
patients differing in the number of previous depressive episodes
has been found, although the number of previous depressive epi-
sodes demarcating distinct subgroups varies among studies (Bock-
ting et al., in press; Ma & Teasdale, 2004; Teasdale et al., 2000).
Third, our research design does not allow us to conclude that
memory specificity is a trait marker for vulnerability to the first
onset of depression, because we did not study memory specificity
in a group of patients without a history of depression before the
first onset of depression (Just, Abramson, & Alloy, 2001). In a
recent analogue study after controlling for baseline levels of de-
pression, negative life events were associated with increased de-
pressive symptoms only in those students high in overgeneral
memories at baseline (Gibbs & Rude, 2004). Possibly memory
specificity does not predict more autonomous recurrences of de-
pression but will predict onset of depression in response to life
events consistent with its moderating role in determining problem-
solving reactions (Williams, Barnhofer, Crane, & Beck, 2005).
These considerations underline the necessity to conduct a prospec-
tive study in which the assessment of autobiographical memory
precedes the onset of a first depressive episode.
American Psychiatric Association. (1994). Diagnostic and statistical man-
ual of mental disorders (4th ed.). Washington, DC: Author.
Arntz, A., Meeren, M., & Wessel, I. (2002). No evidence for overgeneral
memories in borderline personality disorder. Behaviour Research and
Therapy, 40, 1063–1068.
Beck, A. T., Rush, A. J., Shaw, B. F., & Emery, G. (1979). Cognitive
therapy of depression. New York: Guilford Press.
Bockting, C. L. H., Schene, A. H., Spinhoven, P., Koeter, M. W. J., Wou-
ters, L. F., Huyser, J., et al. (2005). Preventing relapse/recurrence in
recurrent depression with cognitive group therapy: A randomized con-
trolled trial. Journal of Consulting and Clinical Psychology, 73, 647–
Brewin, C. R. (1998). Intrusive autobiographical memories in depression
and post- traumatic stress disorder. Applied Cognitive Psychology, 12,
Brewin, C. R., Reynolds, M., & Tata, P. (1999). Autobiographical memory
processes and the course of depression. Journal of Abnormal Psychol-
ogy, 108, 511–517.
Brittlebank, A. D., Scott, J., Williams, J. M. G., & Ferrier, I. N. (1993).
SPINHOVEN ET AL.
Autobiographical memory in depression: State or trait marker. British
Journal of Psychiatry, 162, 118–121.
Burt, D. B., Zembar, M. J., & Niederehe, G. (1995). Depression and
memory impairment: A meta-analysis of the association, its pattern and
specificity. Psychological Bulletin, 117, 285–305.
Dalgleish, T., Spinks, H., Yiend, J., & Kuyken, W. (2001). Autobiograph-
ical memory style in seasonal affective disorder and its relationship to
future symptom remission. Journal of Abnormal Psychology, 110, 335–
Dalgleish, T., Tchanturia, K., Serpell, L., Hems, S., Yiend, J., de Silva, P.
et al. (2003). Self-reported parental abuse relates to autobiographical
memory style in patients with eating disorders. Emotion, 3, 211–222.
Dalgleish, T., & Watts, F. N. (1990). Biases of attention and memory in
disorders of anxiety and depression. Clinical Psychology Review, 10,
de Decker, A., Hermans, D., Raes, F., & Eelen, P. (2003). Autobiograph-
ical memory specificity and trauma in inpatient adolescents. Journal of
Clinical Child and Adolescent Psychology, 32, 22–31.
Delis, D. C., Kramer, J. H., Kaplan, E., & Ober, B. A. (1987). California
verbal learning test: Adult version. San Antonio, TX: Psychological
First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. W. (1997).
Structured clinical interview for DSM–IV axis I disorders (SCID-I/P).
New York: New York State Psychiatric Institute, Biometrics Research
Gibbs, B. R., & Rude, S. S. (2004). Overgeneral autobiographical memory
as depression vulnerability. Cognitive Therapy and Research, 28, 511–
Goddard, L., Dritschel, B., & Burton, A. (1996). Role of autobiographical
memory in social problem solving and depression. Journal of Abnormal
Psychology, 105, 609–616.
Hamilton, M. (1960). A rating scale for depression. Journal of Neurology,
Neurosurgery and Psychiatry, 23, 56–62.
Hammes, J. G. W. (1978). De Stroop Kleurwoord Test: Handleiding. Lisse,
the Netherlands: Swets & Zeitlinger.
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.
Hermans, D., Van den Broeck, K., Belis, G., Raes, F., Pieters, G., & Eelen,
P. (2004). Trauma and autobiographical memory specificity in depressed
inpatients. Behaviour Research and Therapy, 42, 775–789.
Just, N., Abramson, L. Y., & Alloy, L. B. (2001). Remitted depression
studies as tests of the cognitive vulnerability hypotheses of depression
onset: A critique and conceptual analysis. Clinical Psychology Review,
Kremers, I. P., Spinhoven, P., & Van der Does, A. J. W. (2004). Autobio-
graphical memory in depressed and non-depressed patients with border-
line personality disorder. British Journal of Clinical Psychology, 43,
Lezak, M. D., Howieson, D. B., & Lorin, D. W. (2004). Neuropsycholog-
ical assessment (4th ed.). Oxford: Oxford University Press.
Ma, S. H., & Teasdale, J. D. (2004). Mindfulness-based cognitive therapy
for depression: Replication and exploration of differential relapse pre-
vention effects. Journal of Consulting and Clinical Psychology, 72,
Mackinger, H. F., Leibetseder, M. F., Kunz-Dorfer, A. A., Fartacek, R. R.,
Whitworth, A. B., & Feldinger, F. F. (2004). Autobiographical memory
predicts the course of depression during detoxification therapy in alcohol
dependent men. Journal of Affective Disorders, 78, 61–65.
Mackinger, H. F., Loschin, G. G., & Leibetseder, M. M. (2000). Prediction
of postnatal affective changes by autobiographical memories. European
Psychologist, 5, 52–61.
Mackinger, H. F., Pachinger, M. M., Leibetseder, M. M., & Fartacek, R. R.
(2000). Autobiographical memories in women remitted from major
depression. Journal of Abnormal Psychology, 109, 331–334.
McNally, R. J., Lasko, N. B., Macklin, M. L., & Pitman, R. K. (1995).
Autobiographical memory disturbance in combat-related posttraumatic
stress disorder. Behaviour Research and Therapy, 33, 619–630.
Nandrino, J. L., Pezard, L., Poste, A., Reveillere, C., & Beaune, D. (2002).
Autobiographical memory in major depression: A comparison between
first-episode and recurrent patients. Psychopathology, 35, 335–340.
Ormel, J., Oldehinkel, A. J., & Vollebergh, W. (2004). Vulnerability
before, during, and after a major depressive episode: A 3-wave
population-based study. Archives of General Psychiatry, 61, 990–996.
Paradiso, S., Lamberty, G. J., Garvey, M. J., & Robinson, R. G. (1997).
Cognitive impairment in the euthymic phase of chronic unipolar depres-
sion. Journal of Nervous and Mental Disease, 185, 748–754.
Park, R. J., Goodyer, I. M., & Teasdale, J. D. (2002). Categoric overgeneral
autobiographical memory in adolescents with major depressive disorder.
Psychological Medicine, 32, 267–276.
Peeters, F., Wessel, I., Merckelbach, H., & Boon-Vermeeren, M. (2002).
Autobiographical memory specificity and the course of major depressive
disorder. Comprehensive Psychiatry, 43, 344–350.
Phillips, S., & Williams, J. M. G. (1997). Cognitive impairment, depression
and the specificity of autobiographical memory in the elderly. British
Journal of Clinical Psychology, 36, 341–347.
Pollock, L. R., & Williams, J. M. G. (2001). Effective problem solving in
suicide attempters depends on specific autobiographical recall. Suicide
and Life-Threatening Behavior, 31, 386–396.
Raes, F., 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.
Rasbach, J., Steele, F., Browne, W., & Prosser, B. (2004). A user’s guide
to MLwiN. London: Centre for Multilevel Modeling, Institute of Edu-
cation, University of London.
Sampson, M. J., Kinderman, P., Watts, S., & Sembi, S. (2003). Psycho-
pathology and autobiographical memory in stroke and non- stroke hos-
pitalized patients. International Journal of Geriatric Psychiatry, 18,
Sheehan, D. V., Lecrubier, Y., Harnett-Sheehan, K., Amorim, P., Janavs,
J., Weiller, E., et al. (1998). The MINI-International Neuropsychiatric
Interview (M.I.N.I.): The development and validation of a structured
diagnostic psychiatric interview for DSM–IV and ICD-10. Journal of
Clinical Psychiatry, 59, 22–33.
Squire, L. R. (1995). Biological foundations of accuracy and inaccuracy in
memory. In D. L. Schacter (Ed.), Memory distortions: How minds,
brains, and societies reconstruct the past (pp. 197–225). Cambridge,
MA: Harvard University Press.
Svaldi, J. J., & Mackinger, H. F. (2003). Obstructive sleep apnea syn-
drome: Autobiographical memory predicts the course of depressive
affect after nCPAP therapy. Scandinavian Journal of Psychology, 44,
Teasdale, J. D., Segal, Z. V., Williams, J. M. G., Ridgeway, V. A., Soulsby,
J. M., & Lau, M. A. (2000). Prevention of relapse/recurrence in major
depression by mindfulness-based cognitive therapy. Journal of Consult-
ing and Clinical Psychology, 68, 615–623.
Tulving, E. (1972). Episodic and semantic memory. In E. Tulving & W.
Donaldson (Eds.), Organization of memory (pp. 381–403). New York:
van Vreeswijk, M. F., & de Wilde, E. J. (2004). Autobiographical memory
specificity, psychopathology, depressed mood and the use of the Auto-
biographical Memory Test: A meta-analysis. Behaviour Research and
Therapy, 42, 731–743.
Watkins, E., & Teasdale, J. D. (2001). Rumination and overgeneral mem-
ory in depression: Effects of self-focus and analytic thinking. Journal of
Abnormal Psychology, 110, 353–357.
Watkins, E., Teasdale, J. D., & Williams, R. M. (2000). Decentring and
distraction reduce overgeneral autobiographical memory in depression.
Psychological Medicine, 30, 911–920.
Wechsler, D. (1981). Manual for the Wechsler Adult Intelligence Scale—
Revised. New York: Psychological Corporation.
Wessel, I., Meeren, M., Peeters, F., Arntz, A., & Merckelbach, H. (2001).
Correlates of autobiographical memory specificity: The role of depres-
sion, anxiety and childhood trauma. Behaviour Research and Therapy,
Wessel, I., Merckelbach, H., & Dekkers, T. (2002). Autobiographical
memory specificity, intrusive memory, and general memory skills in
Dutch-Indonesian survivors of the World War II era. Journal of Trau-
matic Stress, 15, 227–234.
Wheeler, M. A., Stuss, D. T., & Tulving, E. (1997). Toward a theory of
episodic memory: The frontal lobes and autonoetic consciousness. Psy-
chological Bulletin, 121, 331–354.
Williams, J. M. G. (1995). Depression and the specificity of autobiograph-
ical memory. In D. C. Rubin (Ed.), Remembering our past. Studies in
autobiographical memory (pp. 244–267). Cambridge, MA: Cambridge
Williams, J. M. G., Barnhofer, T., Crane, C., & Beck, A. T. (2005).
Problem solving deteriorates following mood challenge in formerly
depressed patients with a history of suicidal ideation. Journal of Abnor-
mal Psychology, 114, 421–431.
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. (1992). Categoric and extended
autobiographical memories. In M. A. Conway, D. C. Rubin, H. Spinnler,
& W. Wagenaar (Eds.), Theoretical perspectives on autobiographical
memory (pp. 391–410). Dordrecht, the Netherlands: Kluwer.
Williams, J. M. G., Ellis, N. C., Tyers, C., Healy, H., Rose, G., &
MacLeod, A. K. (1996). The specificity of autobiographical memory and
imageability of the future. Memory & Cognition, 24, 116–125.
Williams, J. M. G., Teasdale, J. D., Segal, Z. V., & Soulsby, J. (2000).
Mindfulness-based cognitive therapy reduces overgeneral autobiograph-
ical memory in formerly depressed patients. Journal of Abnormal Psy-
chology, 109, 150–155.
Williams, W. H., Williams, J. M. G., & Ghadiali, E. J. (1998). Autobio-
graphical memory in traumatic brain injury: Neuropsychological and
mood predictors of recall. Neuropsychological Rehabilitation, 8, 43–60.
Wilson, B., Cockburn, J., & Baddeley, A. D. (1985). The Rivermead
Behavioural Memory Test. Bury St. Edmunds, England: Thames Vally
Winthorpe, C., & Rabbitt, P. (1988). Working memory capacity, IQ, age,
and the ability to recount autobiographical events. In M. M. Gruneberg,
P. E. Morris, & R. N. Sykes (Eds.), Practical aspects of memory:
Current research and issues (pp. 175–179). Chichester, England: Wiley.
Received October 29, 2004
Revision received July 26, 2005
Accepted July 28, 2005 ?
SPINHOVEN ET AL.