Initial psychometric properties of the experiences questionnaire: validation of a self-report measure of decentering.
ABSTRACT Decentering is defined as the ability to observe one's thoughts and feelings as temporary, objective events in the mind, as opposed to reflections of the self that are necessarily true. The Experiences Questionnaire (EQ) was designed to measure both decentering and rumination but has not been empirically validated. The current study investigated the factor structure of the EQ in both undergraduate and clinical populations. A single, unifactorial decentering construct emerged using 2 undergraduate samples. The convergent and discriminant validity of this decentering factor was demonstrated in negative relationships with measures of depression symptoms, depressive rumination, experiential avoidance, and emotion regulation. Finally, the factor structure of the EQ was replicated in a clinical sample of individuals in remission from depression, and the decentering factor evidenced a negative relationship to concurrent levels of depression symptoms. Findings from this series of studies offer initial support for the EQ as a measure of decentering.
Initial Psychometric Properties of the Experiences Questionnaire:
Validation of a Self-Report Measure of Decentering
David M. Fresco, Michael T. Moore, Manfred H.M. van Dulmen
Kent State University
Zindel V. Segal
University of Toronto
S. Helen Ma, John D. Teasdale
Medical Research Council Cognition and Brain Sciences Unit, Cambridge, UK
J. Mark G. Williams
Decentering is defined as the ability to observe one’s
thoughts and feelings as temporary, objective events in the
mind, as opposed to reflections of the self that are
necessarily true. The Experiences Questionnaire (EQ) was
designed to measure both decentering and rumination but
has not been empirically validated. The current study
investigated the factor structure of the EQ in both under-
graduate and clinical populations. A single, unifactorial
decentering construct emerged using 2 undergraduate
samples. The convergent and discriminant validity of this
decentering factor was demonstrated in negative relation-
ships with measures of depression symptoms, depressive
rumination, experiential avoidance, andemotion regulation.
Finally, the factor structure of the EQ was replicated in a
clinical sample of individuals in remission from depression,
and the decentering factor evidenced a negative relationship
to concurrent levels of depression symptoms. Findings from
this series of studies offer initial support for the EQ as a
measure of decentering.
S A F R A N A N D S E G A L (1 990) define decentering as
the ability to observe one’s thoughts and feelings as
temporary, objective events in the mind, as opposed
to reflections of the self that are necessarily true. In
a decentered perspective, “…the reality of the
moment is not absolute, immutable, or unalter-
able…” (Safran & Segal, 1990, p. 117). For exam-
ple, an individual engaged in decentering would
say, “I am thinking that I feel depressed right now”
instead of “I am depressed.” Decentering is present-
focused and involves taking a nonjudgmental and
accepting stance regarding thoughts and feelings.
Decentering, or the capacity to take adetached view
of one’s thoughts and emotions, is a concept that
has held some importance early on within the
cognitive-behavioral tradition. Ingram and Hollon
(1986, p. 272) posit that “cognitive therapy relies
on helping individuals switch to a controlled,
effortful mode of processing that is metacognitive
in nature and focuses on depression-related cogni-
tion” and that “the long-term effectiveness of cog-
nitive therapy may lie in teaching patients to initiate
this process in the face of future stress.” Safran and
Segal (1990) emphasize decentering as an effortful
activity as well as an important potential mechan-
ism of change in cognitive therapy, proposing that
interpersonal processes in the course of psychother-
apy provide the opportunity to experientially dis-
confirm dysfunctional interpersonal schema. By
identifying how their beliefs influence their interac-
tions with other people, clients recognize how they
Behavior Therapy 38 (2007) 234–246
Address correspondence to David M. Fresco, Department of
Psychology, 226 Kent Hall Annex, Kent State University, Kent, OH
44242, USA; e-mail: firstname.lastname@example.org.
© 2007 Association for Behavioral and Cognitive Therapies. Published by
Elsevier Ltd. All rights reserved.
actively shape their reality and therefore how their
thoughts and feelings do not necessarily reflect
Insofar as the ability to decenter is thought to be
necessary for healthy cognitive, psychological, and
social development, the lack of this ability is
thought to be a general vulnerability factor for
psychological and social dysfunction. Research into
depression, in particular, has illustrated that de-
pressed individuals are characterized by a bias for
attending to their own experiences (Lyubormirsky
& Nolen-Hoeksema, 1995; Lyubormirsky, Tucker,
Caldwell, & Berg, 1999; Morrow & Nolen-
Hoeksema, 1990). This egocentric preoccupation
with their own experiences also seems to impair
depressed individuals’ ability to function socially
(Evans, Williams, O’Loughlin, & Howells, 1992;
Sidley,Whitaker, Calam, & Wells, 1997). Clinically,
gains in decentering correspond with reductions in
this bias in depressed individuals (Watkins, Teas-
dale, & Williams, 2000), suggesting that decenter-
ing is a concept of importance in maintaining
Initial efforts to reliably measure decentering
arose within the assessment of a related construct,
megacognitive awareness, which is operationalized
as “the process of experiencing negative thoughts
and feelings within a decentered perspective…”
(Teasdale et al., 2002, p. 276). Metacognitive
awareness is assessed using the Measure of
Awareness and Coping in Autobiographical Mem-
ory (MACAM; Moore, Hayhurst, & Teasdale,
1996). In the MACAM, participants are presented
with descriptions of eight mildly depressing situa-
tions via tape recording and asked to imagine
themselves in these situations. Interviewers then
utilize a semistructured interview to elicit specific
memories that have been brought to mind by the
taped vignette. The participant rates the feelings
associated with each of the eight memories (one for
each vignette) on a 0-to-100 scale, and the
interviewer rates the degree of metacognitive
awareness on a 1-to-5 scale. Teasdale et al.
(2002) conducted a comprehensive study to estab-
lish the construct validity of this measure of meta-
cognitive awareness by first comparing 40 patients
in partial remission from major depression to 20
never-depressed healthy controls. The never-
depressed group possessed higher metacognitive
awareness (Cohen’s (1988) d=1.08), even after
controlling for individual differences in levels of
Teasdale and his colleagues (2002) also evaluated
the degree to which metacognitive awareness might
mediate the reduction in relapse that results from
Mindfulness-Based Cognitive Therapy for depres-
sion (MBCT; Segal et al., 2002). MBCT is group
psychotherapy for the prevention of relapse in
which formerly depressed patients are provided
with specific training in mindfulness skills, which,
in part “… teach individuals to become more aware
of thoughts and feelings and to relate to them in a
wider, decentered perspective…” (Teasdale et al.,
2000, p. 618). In the Teasdale et al. (2002) study,
100 participants currently in remission or recovery
from major depression were randomized to receive
MBCT or treatment as usual (TAU). As predicted,
MBCT training resulted in larger increases in
decentering than TAU, highlighting its potential as
a possible mediator. Previous findings have indi-
cated that MBCT reduces risk of relapse or
recurrence of major depressive disorder-particu-
larly for patients with a history of depressive
episodes (Ma & Teasdale, 2004; Teasdale et al.,
2000). In the first study to evaluate the efficacy of
MBCT (Teasdale et al., 2000), MBCT significantly
reduced rates of relapse for participants who had
experienced three or more prior episodes of
depression (40% of patients relapsed in this
group) when compared to TAU (66% relapse rate
in this group). Ma and Teasdale (2004) replicated
the finding that MBCT reduced rates of relapse for
individuals with a history of depression when
compared to TAU using an identical methodology
(36% of patients relapsed in MBCT versus 78%
relapse rate in a group receiving TAU).
Teasdale and colleagues (2002) also examined
158 patients from two treatment sites. All partici-
pants, who had recently suffered from major
depression, were partially remitted following treat-
ment with antidepressant medication and rando-
mized to first receive antidepressant medication and
clinical management alone or together with cogni-
tive therapy for 20 weeks, and then receive follow-
up continuation and maintenance medication for
48 weeks. Findings indicated that lower levels of
baseline metacognitive awareness predicted earlier
relapse across both treatment groups. There was
also a larger increase in metacognitive awareness in
the cognitive therapy group as compared to the
clinical management group.
The measure of metacognitive awareness used
by Teasdale et al. (2002) was specifically designed
to assess decentering. However, it is time-consum-
ing and requires participants to listen to eight
taped vignettes and a trained individual to admin-
ister a semistructured interview to the participant.
Despite the strengths of such a thorough metho-
dology, it would be prohibitive to use such a
measure in more practice-oriented settings. Thus,
as an alternative to this more time-consuming
assessment, the Experiences Questionnaire (EQ)
was developed by Teasdale as a means of
operationalizing the changes assumed to occur
during MBCT. One predicted change was the
ability to adopt a decentered perspective, which, in
turn, may be applicable to other psychological
treatments of depression. Insofar as an important
aspect of mindfulness training was believed to be
teaching patients how to “decenter” from depres-
sive thinking patterns, it would be important to
have a measure of the extent to which patients
actually did so. The EQ items were generated by
Teasdale, in consultation with Segal and Williams,
to assess, in a face-valid way, three facets of
decentering (Z. V. Segal, personal communication,
July 14, 2006). These facets are the ability to view
one’s self as not synonymous with one’s thoughts
(e.g., “I can separate myself from my thoughts and
feelings”), the ability not to habitually react to
one’s negative experiences (e.g., “I can observe
unpleasant feelings without being drawn into
them”), and the capacity for self-compassion
(e.g., “I am better able to accept myself as I
am”). The full scale was constructed to have two
subscales, one measuring the changes assumed to
occur in MBCT, including decentering, and a
second measuring rumination. The EQ rumination
subscale was included as a control against response
bias. The purpose of the current investigation was
to study the factor structure of the EQ and refine
its psychometric properties in a manner consistent
with the theory that guided its creation.
In summary, decentering is a construct singled
out early in the history of cognitive therapy as
an active ingredient in producing lasting treat-
ment effects, but which has gained greater
currency in our field during the recent explosion
of interest in mindfulness and acceptance. In this
paper, we present the findings of three studies
that aim to provide support for the initial
psychometric properties and construct validity
of the EQ, a self-report measure of decentering.
Using exploratory and confirmatory factor ana-
lysis in two consecutive samples of college
students, the first study sought to establish the
factor structure of the EQ. Given that decenter-
ing is believed to be a capacity present in healthy
individuals, a relatively healthy sample of college
students was deemed a suitable starting point for
scale construction. The second study sought to
establish the concurrent and discriminant validity
of the decentering measure derived from the EQ
by examining its relationship to theoretically
related constructs, such as depressive rumination,
experiential avoidance, and emotion regulation,
in a sample of college students who received
extensive clinician assessment of their current
and lifetime psychopathology. Finally, Study 3
sought to replicate and extend the findings of
Study 1 by examining the factor structure of the
EQ initially derived in the college student
samples within a sample of patients with
remitted major depressive disorder. Study 3 also
sought to further examine the concurrent validity
of the decentering factor in relation to measures
of depression symptoms.
Study 1: Initial Factor Structure of the EQ
Participants and Procedures. Data were obtained
from two samples (Sample 1 and 2) of students
enrolled in Introduction to Psychology courses at a
large, midwestern United States university and
involved in that university’s mass testing procedure.
All students received course credit in return for their
participation. Sample 1 (n=1,150) was composed
of 765 females (66.5%) and 385 males (33.5%),
with a mean age of 19.1 years (SD=4.1). Sample 2
(n=519) was composed of 335 females (64.5%)
and 184 males (35.5%), with a mean age of 19.3
Measure. The EQ is a 20-item self-report
inventory designed to measure a decentering or
disidentification with content of negative thinking,
which is hypothesized to be a process of change in
MBCT. Helping patients learn how to decenter, or
take a step back from their negative thoughts, is one
of the core goals of mindfulness-based clinical
interventions, but can also occur following success-
ful cognitive therapy. The EQ was originally
designed to possess two rationally derived sub-
scales: rumination (6 items) and wider perspective
(14 items). Rumination items (e.g., “I think over
and over again about what others have said to me”)
were included in the EQ to help rule out the
possibility that any increases in wider perspective
might be explained by an acquiescent response bias.
Items are rated on a 5-point Likert scale (1=never,
5=all the time).
We first attempted to confirm the presence of the
rationally derived factor structure initially pro-
posed for the EQ. Confirmatory factor analysis
(CFA) was utilized in Sample 1 with maximum
likelihood (ML) estimation using the EQS 6.1
software (Bentler & Wu, 1995). ML estimation is
the most widely used estimation method in CFA
and provides robust results even in light of small
violations of normality (Chou & Bentler, 1995;
Hoyle & Panter, 1995). However, this rationally
derived structure did not fit the data well, even
fresco et al.
after modifications were made to improve fit based
upon the LaGrange Multiplier Test: χ2=
1224.96; χ2/df=7.38; CFI=.76; RMSEA=.08;
SRMR=.09.Therefore, exploratory factor analysis
(EFA) using ML estimation and promax rotation
was undertaken in this same sample to provide an
initial evaluation of the factor structure for the 20
EQ items. EFA was also conducted using ML
estimation and promax rotation using the PRELIS
(Jöreskog & Sörbom, 1988) software package to
obtain additional fit indices for these various
models. Promax rotation was utilized for its ability
to represent correlations between latent variables
(Fabrigar et al., 1999). Although this method
identified four factors with eigenvalues greater
than 1, inspection of the scree plot revealed that
two primary factors accounted for a significant
portion of the total variance (35.87%), whereas
the other two factors contributed relatively little
(12.48%). In addition, the content of Factor 3 and
4 was uninterpretable. Factor 3 contained only 3
items that did not substantially cross-load onto
other factors (4 items in total), and Factor 4 was
not represented by any items. Floyd and Widaman
1.0 as a criterion for factor retention tends to
overestimate the number of factors and that inspec-
tion of the scree plot is a more valid method. The
additional fit indices obtained using PRELIS indi-
cated the superiority of a three- (χ2= 408.46;
χ2/df=3.07; RMSEA=.05) or four-factor solution
(χ2=270.58; χ2/df=2.33; RMSEA=.04) to a
two-factor model (χ2= 742.22; χ2/df=4.92;
suffered from the problems of low item representa-
tion and interpretability mentioned above. There-
fore, the EFAwas rerun, again using ML estimation
and promax rotation, but with the additional
constraint of only extracting two factors (see Table
1foralistingoffactorloadings).The content of the
items essentially replicated the original, rationally
derived two-factor structure, with the exception of
Item 4, which loaded more highly onto Rumina-
tion than Decentering; Item 2, which did not load
significantly on any factor (higher than .32;
Comrey & Lee, 1992); and Item 20, which was
a cross-loading item (loadings on both factors
were greater than .32; Tabachnick & Fidell,
2001). The Decentering factor evidenced good
internal consistency in Sample 1 (α=.83). How-
ever, the Rumination factor was in the lower
range of “adequate” in terms of its internal
consistency (α=.70; Hunsley & Mash, in press;
To confirm the presence of a two-factor solution,
CFA was utilized in a second sample of college
students (Sample 2) using EQS 6.1 (Bentler & Wu,
1995). ML estimation was again used; however,
poor fit resulted when testing the two-factor
solution obtained using EFA: χ2=558.06;
χ2/df=3.80; CFI=.81, RMSEA=.08; SRMR=.09.
The LaGrange Multiplier Test, an atheoretical
psychometric means of improving fit, indicated
that allowing items to cross-load was the most
effective way to improve fit (representing 6 of the
10 parameters which, if added, would improve fit
the most). However, cross-loading items would
make interpretation of the constructs measured by
these factors difficult. Therefore, a unifactorial
model was tested with all of the items from the
two-factor model obtained from the EFA con-
strained on a single factor. By doing so, the
problem of item cross-loadings might be
addressed without having to delete items from
the model. Unfortunately, this model also evi-
denced poor fit (χ2=741.12; χ2/df=5.01;
CFI=.73, RMSEA=.09; SRMR=.10), due primar-
ily to poor loadings from items originally placed
onto the Rumination factor (7 items had standar-
dized loadings <.40). Therefore, the model was
rerun using only the decentering items. The
decentering factor that emerged was identical to
the one identified in EFA with three exceptions:
two items which the initial CFA indicated did not
load significantly were omitted, and Item 20 was
added as it was the single item determined
theoretically to best represent the factor. Error
terms from scale items theoretically determined to
share substantial item content were allowed to
intercorrelate, with the assumption that these items
would share measurement error (Bollen, 1989;
Rozeboom, 1966; Rubio & Gillespie, 1995). Items
3 (“I am better able to accept myself as I am”) and
14 (“I can treat myself kindly”), 9 (“I notice that I
don’t take difficulties so personally”) and 10 (“I
can separate myself from my thoughts and
feelings”), and 16 (“I have the sense that I am
fully aware of what is going on around me and
inside me”) and 18 (“I am consciously aware of a
sense of my body as a whole”) were allowed to
intercorrelate in the present study, owing to the
similarity of their item content (these items were
also allowed to intercorrelate in the two-factor
solution). Research has shown that if correlated
measurement error is not accounted for, biased
estimators can result, and this bias can have both
general and random effects (Biddle & Marlin,
1987; Bollen, 1989; MacCallum & Tucker, 1991).
However, methodologists have also critiqued the
use of atheoretical, statistically determined corre-
lated measurement error (Cortina, 2002). The use
of theoretically guided correlated measurement
both of these concerns. The model indicated
good fit (χ2=103.79; χ2/df=2.53; CFI=.95,
RMSEA=.06; SRMR=.04; see Table 1 for factor
loadings and Table 2 for item-level descriptive
statistics for Sample 2) and adequate internal
Multi-Group CFA (MGCFA) was utilized to
determine if this model represented the data well
in both males and females. We sought to evaluate
if the loadings of the various items composing the
decentering factor were equal across gender and
possessed what is known as metric invariance
(Bollen, 1989). Testing metric invariance involves
first estimating a model where all factor loadings
are constrained to be equal across the two groups
being compared (in this case, males and females).
This model is then compared to several alternative
models where the factor loading of each item is
unconstrained, and allowed to be freely estimated.
Differences in goodness-of-fit between the con-
strained and the unconstrained models indicate a
lack of metric invariance, or that the item-factor
loadings are unequal across groups. Cheung and
Rensvold (2002) suggest that a difference in CFI
(ΔCFI) greater than .01 is indicative of an item not
loading equally between groups. None of the
values of ΔCFI exceeded this cutoff in our
Experiences questionnaire item-factor and factor-scale loadings
Item Study 1: Sample 1 Study 1: Sample 2
Factor 1 Factor 2 Standardized Unstandardized
Factor 1 (Decentering)
EQ03 I am better able to accept myself as I am.
EQ15 I can observe unpleasant feelings
without being drawn into them.
EQ09 I notice that I don’t take difficulties
EQ14 I can treat myself kindly.
EQ10 I can separate myself from my thoughts
EQ16 I have the sense that I am fully aware of
what is going on around me and inside me.
EQ06 I can slow my thinking at times of stress.
EQ08 I am not so easily carried away by my
thoughts and feelings.
EQ17 I can actually see that I am not my
EQ18 I am consciously aware of a sense of my
body as a whole.
EQ05 I am kinder to myself when things
EQ12 I can take time to respond to difficulties.
EQ20 I view things from a wider perspective.
.512 .942 (.11) .7311.076 (.13)
.6291.083 (.11) .483.673 (.12)
.529.862 (.10).667 1.047 (.14)
.560.874 (.10) .516.879 (.15)
Factor 2 (Rumination)
EQ13 I think over and over again about what
others have said to me.
EQ11 I analyze why things turn out the way
EQ19 I think about the ways in which I
am different from other people.
EQ04 I notice all sorts of little things and details
in the world around me.
EQ01 I think about what will happen in the future.
EQ07 I wonder what kind of person I really am.
Items not loading on a factor or double-loading
EQ20 I view things from a wider perspective.
EQ02 I remind myself that thoughts aren’t facts.
Note. Standardized factor loadings in bold face represent an item loading on that factor. “NA” indicates that an item was dropped and
therefore was not included for analysis.
fresco et al.
MGCFA of gender, indicating that the unifactorial,
decentering model fit both males and females
An initial attempt to replicate the rationally derived
factor structure of the EQ proposed originally by
Teasdale indicated that this model failed to fit the
data in a sample of college students. A subsequent
exploratory factor analysis revealed the presence of
two factors that essentially mimicked the original,
rationally derived structure. However, this two-
factor model for the EQ was not found to fit the
data when a confirmatory factor analysis was
undertaken on a second sample of college students.
Subsequent adjustments to the model indicated the
presence of a unifactorial decentering construct that
was found to fit the second college-student sample
well. MGCFA also indicated that this construct fit
both males and females in this sample equally well.
Although this replication was encouraging, the fact
that a convenience sample was used implies that
additional work is needed to further replicate this
factor structure in community and clinical samples
to ensure broader generalizability. In addition, the
small number of males and females calls into
question the replicability and validity of the multi-
group analyses. Another limitation of Study 1 was
the lack of concurrent and discriminant validity
measures by which to evaluate the derived decen-
Study 2: Examining the Concurrent Validity of
Decentering to Related Constructs
Study 2 sought to examine the concurrent and
discriminant validity of the decentering scale in
relation to four theoretically meaningful constructs
in a nonpatient sample where participants also
underwent clinician assessment for current and
lifetime psychopathology. The four concurrent
validity measures selected were depressive rumina-
tion, which emerges from the response styles
theory (Nolen-Hoeksema, 1998); experiential
avoidance, which emerges from Acceptance and
Commitment Therapy (ACT; Hayes, Strosahl, &
Wilson, 1999), a contemporary theoretical model
within behavior therapy; and cognitive reappraisal
and emotion suppression, which emerged from
Gross’ (1998a, 1998b,1999) model of emotion
As noted above, decentering is conceptually anti-
thetical to depressive rumination, which Nolen--
Hoeksema (1998; p. 216) defines as “focusing
passively and repetitively on one’s symptoms of
taking action to correct the problems one identi-
fies.” Following recommendations of Treynor,
depressive rumination with the five-item brooding
A central tenet of the ACT model of psycho-
pathology is that much human suffering can be
accounted for by experiential avoidance, which is
defined as the tendency to attempt to suppress or
inhibit the frequency or severity of emotions,
thoughts, and memories (Hayes et al., 1999).
ACT, as a therapy approach, strives to lessen indivi-
duals’ efforts at escaping or avoiding this painful
psychological content in favor of engaging in acti-
vities likely to promote positive reinforcement and
purposeful outcomes. Given this conceptual rele-
vance, the present study sought to examine the
association of decentering to experiential avoid-
ance, which is assessed with the Acceptance and
Action Questionnaire (Hayes et al., 2004).
Decentering also shows some conceptual rela-
tionship to the model of emotion regulation posited
by Gross and colleagues (Gross, 1998a,1998b,
1999; Gross & Muñoz, 1995). Gross and John
(2003) have developed and validated a self-report
measure called the Emotion Regulation Question-
naire (ERQ), which consists of reappraisal and
suppression subscales. Reappraisal involves cogni-
tively transforming a situation so as to alter its
emotional impact, and has shown to help indivi-
duals attenuate negative emotion experiences
(Gross, 1998a). Emotional suppression refers to
attempts to suppress overt emotional responses, a
response that has been linked to greater cardiovas-
cular reactivity to emotional stimuli (Gross &
Levenson, 1993). Relative to these constructs,
decentering would be expected to be positively
associated with cognitive reappraisal and nega-
tively associated with emotion suppression.
Experiences Questionnaire decentering items descriptive
statistics by sample
ItemMean (Standard Deviation)
Sample 1Sample 2 Sample 3
Note. EQ=Experiences Questionnaire.
The primary goal of Study 2 was to examine the
relationship of decentering to depressive rumina-
tion, experiential avoidance, reappraisal, and emo-
tion suppression. A secondary goal of the study was
to examine the association of decentering to current
or lifetime major depressive disorder. The first
hypothesis (Hypothesis 1) was that decentering
would demonstrate a significant positive correla-
tion with cognitive reappraisal, and significant
negative correlations with depressive rumination,
experiential avoidance, emotion suppression, and
concurrent levels of depression symptoms. The
second hypothesis (Hypothesis 2) was that a history
of major depressive disorder would be associated
with lower levels of decentering as compared to no
history of psychopathology.
Participants. Participants were 61 college stu-
dents (34 women) who were primarily Caucasian
(88%); the remaining participants were African
American (7%), Hispanic (3%), or Native Amer-
ican (2%). The participants had an average age of
19.81 years (SD=2.87). Participants received par-
tial course credit for participating in the study.
Measures. Experiences Questionnaire—Decenter-
This study utilized the 11-item decen-
tering factor that was derived in Study 1. Scores on
this factor can range from 11 to 55, with higher
scores indicating greater decentering. Also, despite
the complexity of the concept described by the
decentering factor, the reading level for the items
evidencedaFlesch-Kincaid gradelevel of4.3,which
compares favorably to the 6th grade level of the
items in the Minnesota Multiphasic Personality
Inventory-2 (Graham, 2005).
Acceptance and Action Questionnaire (AAQ;
Hayes et al., 2004).
The AAQ is a 16-item self-
report measure of experiential avoidance. Partici-
pants rate each statement on Likert-type scales
ranging from 1 (never true) to 7 (always true), with
higher scores reflecting an increasing degree of
experiential avoidance. Sample items include “I am
able to take action on a problem even if I am
uncertain what is the right thing to do” (reverse
scored) and “If I could magically remove all the
painful experiences I’ve had in my life, I would do
so.” The 16-item version of the AAQ was obtained
from its authors prior to their settling on the 9-item
version. Still, preliminary analyses support the
reliability and validity of the 16-item version as
well as the final 9-item version. The two versions
are highly correlated with one another (r=.89;
Hayes et al., 2004) and assess experiential avoid-
ance and control, negative evaluations of internal
experience, psychological acceptance, and the
tendency to act regardless of emotional distress. In
the current study, the AAQ demonstrated accep-
table internal consistency (α=.72).
Ruminative Responses Subscale of the Response
Styles Questionnaire (RSQ; Nolen-Hoeksema &
The RSQ, a 22-item self-report
instrument, is designed to assess an individual’s
characteristic tendency to engage in ruminative be-
havior when feeling depressed. In the current study,
a 25-item version was utilized consisting of the
original 22-item RRS items plus the 3 additional
expanded rumination items used by Treynor et al.
(2003) to derive their brooding and pondering
solution. In the current study, reliability for the five-
item brooding rumination subscale was good
Emotion Regulation Questionnaire (ERQ; Gross
& John, 2003).
The ERQ is a 10-item rationally
derived measure of two aspects of emotion regula-
tion: reappraisal and suppression. The reappraisal
subscale, consistingof 6items, assessesthe abilityto
modify or change the emotions one experiences
(e.g., “I control my emotions by changing the way I
think about the situation I’m in”). The suppression
subscale, consistingof 4items, assessesthe abilityto
avoid or prevent the expression of emotions (e.g., “I
control my emotions by not expressing them”). In
this sample, the internal consistency was good for
both the reappraisal subscale (α=.84) and the
suppression subscale (α=.82).
Beck Depression Inventory-II (BDI-II; Beck, Steer,
& Brown, 1996).
The BDI-II is a 21-item self-
report measure that assesses the severity of depres-
sive symptomatology, including the affective, cog-
nitive, behavioral, somatic, and motivational
components of depression as well as suicidal
wishes. Items are rated on a 0-to-3 scale and reflect
a 2-week time period. The BDI-II has strong
internal consistency in both student and clinical
samples (Beck et al., 1996), and excellent test-retest
reliability (Sprinkle et al., 2002). In the current
study, the BDI-II achieved strong internal consis-
Mood and Anxiety Symptom Questionnaire-Short
Form (MASQ; Watson & Clark, 1991).
MASQ is a 62-item instrument designed to assess
symptoms commonly occurring in the mood and
5 (extremely) Likert-type scale. These 62 items are
subdivided into four subscales: General Distress
Anxious Symptoms (GDA), General Distress
Depressive Symptoms (GDD), Anxious Arousal
(AA), and Anhedonic Depression (AD). The GDA
subscale is comprised of 11 items indicative of
fresco et al.
anxious mood, but provides little discrimination
12 items indicative of depressed mood, but provides
little discrimination from anxious mood (e.g., “Felt
sad”; “Felt like crying”). The AA subscale contains
17 items detailing symptoms of somatic tension and
hyperarousal (e.g., “Startled easily”; “Was trem-
specifically assessing symptoms related to depres-
and low energy (e.g., “Felt like nothing was very
enjoyable”) and 14 reverse-coded items assessing
positive emotional experiences (e.g., “Felt cheer-
ful”). In the present sample, reliability for these sub-
scales was good to excellent (GDA: α=.82; GDD:
α=.90; AA: α=.85; AD: α=.84).
Structured Clinical Interview for DSM-IV-TR,
Research Version (SCID; First, Spitzer, Gibbon, &
The SCID is a widely used
semistructured interview that allows for current
and lifetime diagnoses of Axis I disorders. For the
DSM-IV version of the SCID, Ventura et al. (1998)
reported high interrater agreement (kappas) for
current diagnosis, with an overall weighted kappa
of .82 and a range between .71 and .90 for specific
diagnoses. In the current study, the SCID mood,
anxiety, eating disorder, and substance abuse/
dependence modules were administered by four
doctoral students trained in the SCID as part of
their programmatic clinical training. Doctoral stu-
dents receive formal training in the SCID prior to
the start of their second year of graduate school and
then receive weekly SCID supervision by the first
author as they serve as intake interviewers in the
department’s training clinic.
Procedure. After providing informed consent,
participants completed a battery of self-report
measures that assessed decentering, experiential
avoidance, emotion regulation strategies, depres-
sion and anxiety symptoms, as well as measures not
relevant to the current study. Participants also
underwent a SCID conducted by a graduate
Rates of current/lifetime major depression. Find-
ings from the SCID interviews revealed that 7
participants met diagnostic criteria for current
major depressive disorder (MDD) and an addi-
tional 7 participants met criteria for past MDD, as
well as other current diagnoses (e.g., generalized
anxiety disorder, social phobia, posttraumatic
stress disorder, alcohol abuse). The remaining 47
participants had no current or lifetime history of
Association of decentering to experiential avoid-
ance, emotion regulation, and depression and
anxiety symptoms. Table 3 presents sample
means, standard deviations, and zero-order corre-
lations among study measures. Consistent with
Hypothesis 1, decentering evidenced a significant
positive correlation with cognitive reappraisal and
significant negative correlations with experiential
avoidance, brooding rumination, emotion suppres-
sion, and self-report measures of current depression
and anxiety symptoms.
Association of decentering to current/lifetime
history of depression. To evaluate Hypothesis 2, the
14 participants with current or lifetime MDD were
grouped together and compared to the 47 partici-
pants with no current or lifetime psychopathology.
Consistent with Hypothesis 2, participants with
current or lifetime MDD (M=3.09; SD=0.57)
endorsed significantly lower levels of decentering,
F(1, 59)=5.77, p=.02, d=0.68, as compared to the
Means, (standard deviations), and zero-order correlations among decentering, experiential avoidance, brooding rumination, emotion
regulation, and depression and anxiety symptoms
Note. Experiential Avoidance=Acceptance and Action Questionnaire total score; Brooding=Response Styles Questionnaire Brooding
Subscale; Reappraisal=Emotion Regulation Questionnaire (ERQ) Reappraisal subscale; Suppression=ERQ Suppression subscale; BDI-
II=Beck Depression Inventory-II; MASQ-AD=Mood and Anxiety Symptom Questionnaire Anhedonic Depression subscale; MASQ-
AA=Mood and Anxiety Symptom Questionnaire Anxious Arousal subscale; ⁎Correlation significantly different from zero (p<.05).
participants with no current or lifetime psycho-
pathology (M=3.44; SD=0.44). This difference
exceeded conventions for a medium effect size.
Consistent with expectations, the EQ decentering
factor was found to correlate positively and
significantly with cognitive reappraisal and signifi-
cantly and negatively with depressive rumination,
experiential avoidance, emotion suppression, and
symptoms of depression. In addition, individuals
with no lifetime history of psychopathology were
found to endorse significantly higher mean levels of
decentering than individuals with MDD. In sum,
these findings add confidence to the assertion that
the factor derived in Study 1 from the EQ measures
decentering. However, Studies 1 and 2 both utilized
samples of college students, and further replication
in a treatment-seeking sample is still necessary, a
shortcoming that is addressed in Study 3.
Study 3: Confirmatory Factor Analysis of the
Experiences Questionnaire in a Clinical Sample
Study 3 sought to replicate and extend previous
findings by evaluating both the factor structure and
the clinical validity of the EQ in a sample of patients
with remitted MDD. The factor solution derived
following CFA in the second student sample was
evaluated in a sample of patients with remitted
MDD who were participants in, but had not yet
commenced, a trial of MBCT. The clinical validity
of the factor solution in the patient sample was
assessed in two ways. First, levels of decentering in
the remitted-MDD patients were comparedto levels
of decentering in a sample of healthy control
participants. Hypothesis 1 posited that remitted
MDD patients would endorse lower levels of
decentering as compared to healthy control parti-
cipants. Second, the relationship of decentering to
concurrent levels of self-report and clinician-
assessed depression symptoms was evaluated.
Hypothesis 2 posited a significant and negative
relationship between decentering and the two
measures of concurrent depression symptoms.
Participants. Participants consisted of two clin-
ical subsamples recruited to evaluate the efficacy of
MBCT. Pre-MBCT data from 145 depressed parti-
cipants in remission or recovery at the time of
assessment were obtained from Teasdale et al.
(2000). The participants were recruited from
media advertisements and community health care
facilities at three sites: one in Cambridge, England,
and sampled from its surrounding rural towns and
villages; one in a rural area of north Wales, centered
on Bangor; and one in the metropolitan area of
Toronto, Ontario, Canada. These data were com-
bined with pre-MBCT data from 75 previously
depressed participants who participated in an
MBCT trial conducted by Ma and Teasdale (2004)
in Cambridge, England. In total, the remitted MDD
sample (n=220) was composed of 165 females
(75.0%) and 55 males (25.0%), with a mean age of
43.7 years (SD=9.6). In addition, Ma and Teasdale
(2004) also obtained data from 50 healthy controls
recruited from a volunteer panel and who received
TAU both prior to and during the period of data
collection. These control participants were similar
in age (M=44.5, SD=8.9) and gender (37 women,
74%) to the total clinical sample. In addition,
MGCFA was utilized to determine the metric
invariance of the decentering factor structure in
the two clinical samples and the suitability of
aggregating them. Results indicated that all items
loaded equally well onto the decentering factor in
both samples (all values of ΔCFI<.01, average
Measures. Experiences Questionnaire-Decentering
This study utilized the 11-item decentering
factor that was derived in Study 1.
Hamilton Rating Scale for Depression (HRSD;
The HRSD is a 17-item inter-
view-administered measure of the severity of symp-
toms of depression. It covers a range of affective,
behavioral, and somatic symptoms and has accep-
table psychometric properties (Rabkin & Klein,
1987). Scores can range from 0 to 52, with higher
scores indicating a larger number of symptoms of
depression. Only participants in Sample 3 com-
pleted the HRSD.
Beck Depression Inventory (BDI; Beck, Rush,
Shaw, & Emery, 1979).
measure of the severity of affective, cognitive,
behavioral, and somatic symptoms of depression,
BDI item scores can range from 0 to 3, and total
scores can range from 0 to 63, with higher scores
indicating a larger number of symptoms of depres-
sion. It has been shown to possess adequate
reliability and validity in both psychiatric and
undergraduate populations (Beck, Steer, & Garbin,
1988). Only participants in Sample 3 completed the
A 21-item self-report
Confirmatory Factor Analysis in a Patient
Sample. To both extend the generalizability of the
unifactorial model to a patient sample and improve
confidence in its reliability, CFA was used again in
Sample 3. ML estimation was again used with EQS
6.1 software. The model fit the data reasonably well
fresco et al.
(χ2=61.87; χ2/df=1.51; CFI=.97, RMSEA=
.06; SRMR=.05; see Table 1 for factor loadings
and Table 2 for item-level descriptive statistics) and
possessed good internal consistency (Sample 3:
Clinical Validity. Hypothesis 1 posited that
remitted MDD patients would endorse lower levels
of decentering as compared to healthy controls. In
support of Hypothesis 1, remitted MDD patients
(M=1.81, SD=0.54) scored significantly lower
than the healthy control participants (M=2.47,
SD=0.42), with an effect size exceeding conven-
tions for a large effect: F(1, 117)=50.32, p<.0001,
d=1.31. In support of Hypothesis 2, levels of
decentering among remitted MDD patients were
significantly and negatively correlated with con-
current self-report (r=−.46) and clinician-assessed
(r=−.31) levels of depression symptoms.
The results of Study 3 provide further support for
the unifactorial decentering solution of the EQ.
CFA in a treatment-seeking sample indicated
acceptable fit of the decentering factor. As ex-
pected, participants with remitted MDD endorsed
lower levels of decentering than healthy control
participants and decentering was significantly and
negatively correlated with self-report and clini-
cian-assessed measures of depression symptoms.
The results of the present set of studies indicate that
the EQ measures a unifactorial construct that
performs in a manner that would be consistent
with the construct of decentering in relation to
theoretically meaningful concurrent validity mea-
sures. This one-factor model, thought to represent
the construct of decentering, was extracted from
both a student as well as a composite, multi-
national, clinical sample, illustrating the general-
izability of the model. This decentering factor
possessed adequate to good internal consistency in
both undergraduate and clinical samples. The EQ
was conceived as a measure of decentering that
would generalize across gender, race, and ethnicity.
Although the current study provides preliminary
evidence for generalizability across males and
females, further research will need to be done to
illustrate that the decentering factor found here is
robust across differences in both race and ethnicity.
In addition, preliminary convergent and discrimi-
nant validity was established in the theoretically
consistent correlations found between decentering
and measures of depressive rumination, experien-
tial avoidance, and emotion regulation in an under-
graduate sample. Similarly, decentering evidenced
theoretically meaningful patterns of correlation
with concurrent measures of depression in both
undergraduate and patient samples. Taken toge-
ther, findings from this series of studies provide
initial confidence that the EQ is a reliable and
valid measure of decentering. However, future
research is needed before this claim can be made
Within the cognitive-behavioral tradition, decen-
tering has long been regarded as a possible
mechanism of change in the context of cognitive
therapy of depression (Beck et al., 1979; Ingram &
Hollon, 1986; Safran & Segal, 1990) and predates
mindfulness and acceptance approaches within
psychology (cf. Hayes et al., 1999; Segal et al.,
2002; Teasdale et al., 2002). Still, one of the
motivations for generating the initial items of the
EQ, and submitting them to psychometric scrutiny,
in the context of MBCT, a cognitive-behavioral
that is enriched with mindfulness exercises. Draw-
the capacity for decentering to increase over the
duals suffering from emotional disorders evidence
deficits in the capacity for metacognition. Accord-
ingly, we view decentering as an individual differ-
ence variable that is naturally malleable, that is
likely to be impaired in individuals suffering from
emotional disorders, and which may represent the
mark of a durable treatment response when indivi-
duals evidence gains in decentering (cf. Fresco,
Segal, Buis, & Kennedy, in press; Teasdale et al.,
2002). However, our measure of decentering was
never intended to be synonymous with mindfulness.
Bishop and colleagues (2004), following a consen-
sus conference, arrived at a two-faceted operational
definition of mindfulness. The first facet of this
definition of mindfulness is the capacity for the
sustained self-regulation of attention so that an
individual can remain engaged in the present
moment. The second facet of the consensus defini-
tion of mindfulness is the capacity to maintain an
acceptance and curiosity regarding all of one’s
thoughts, feelings, and sensations. Our definition
of decentering is certainly complementary to Bishop
et al.’s definition of mindfulness, but it is in no way
synonymous with it. Interestingly, Bishop et al. (p.
235) describe decentering and mindfulness as fall-
ing within a “general domain of constructs that
describe the ability to observe the temporal stream
of thoughts and feelings.”
Our decentering scale joins a burgeoning field of
self-report indices with relevance to mindfulness
and acceptance. Findings from the current study
provideaninitial glimpse into how the EQrelates to
one already established construct in this field.
Specifically, decentering evidenced a strong nega-
tive correlation with experiential avoidance, which
emerges from the acceptance-based treatment
approach (cf. Hayes et al., 1999). However, an
important area for future research is to evaluate the
relationship of the EQ to other measures of
mindfulness (cf. Baer, Smith, & Allen, 2004;
Brown & Ryan, 2003; Feldman, Hayes, Kumar,
& Greeson, 2003) as well as to general measures of
quality of life and well-being (cf. Frisch, 1994;
Frisch, Cornell, Villanueva, & Retzlaff, 1992).
Although the findings from the current set of
studies support the factor structure of the EQ and
present some initial concurrent validity data, the
relationship of decentering to the process of
protecting against relapse of major depression has
not yet been assessed. However, Fresco et al.
(submitted for publication) provide some initial
evidence for the role of decentering, as assessed by
the EQ, in the course of acute treatment of major
depression. Findings revealed that CBT patients
treated to remission evidenced increases in decen-
tering whereas patients treated to remission with
antidepressant medication did not. The difference
between CBT and antidepressant medication in
decentering gains corresponded to a medium effect
size (d=.58). The second major finding of this study
was that high posttreatment decentering was
associated with a more durable treatment response
in the subsequent 18 months, especially for CBT
In conclusion, findings from the current series of
studies offer initial enthusiasm for the EQ as a
measure that assesses a construct corresponding to
a protective or resilience factor. The decentering
scale of the EQ shows significant negative relation-
ships with concurrent levels of depression and
anxiety symptoms and theoretically meaningful
associations with constructs including depressive
rumination, experiential avoidance, and emotion
regulation. Greater confidence in proclaiming that
the EQ assesses decentering awaits further investi-
gations with a broader selection of measures within
the mindfulness and acceptance field. However, the
findings to this point warrant sufficient interest in
the EQ to pursue these important issues of con-
Baer, R. A., Smith, G. T., & Allen, K. B. (2004). Assessment of
mindfulness by self-report: The Kentucky Inventory of
Mindfulness Skils. Assessment, 11, 191–206.
Beck, A. T., Rush, A. J., Shaw, B. F., & Emery, G. (1979).
Cognitive therapy of depression. New York: Guildford
Beck, A. T., Steer, R. A., & Brown, G. K. (1996). Manual for the
Beck Depression Inventory, 2nd ed. San Antonio, TX: The
Beck, A. T., Steer, R. A., & Garbin, M. G. (1988). Psychometric
properties of the Beck Depression Inventory: Twenty-five
years of evaluation. Clinical Psychology Review, 8,
Bentler, P. M., & Wu, E. J. C. (1995). EQS for Windows: User’s
guide. Encino, CA: Multivariate Software, Inc.
Biddle, B. J., & Marlin, M. M. (1987). Causality, confirmation,
credulity, and structural equation modeling. Child Deve-
lopment, 58, 4–17.
Bishop, S. R., Lau, M., Shapiro, S., Carlson, L., Anderson,
N. D., Carmody, J., Segal, Z. V., Abbey, S., Speca, M.,
Velting, D., & Devins, G. (2004). Mindfulness: A
proposed operational definition. Clinical Psychology:
Science and Practice, 11, 230–241.
Bollen, K. A. (1989). Structural equations with latent variables.
Oxford: John Wiley & Sons.
Brown, K. W., & Ryan, R. M. (2003). The benefits of
being present: Mindfulness and its role in psychological
well-being. Journal of Personality and Social Psychology,
Cheung, G. W., & Rensvold, R. B. (2002). Evaluating good-
ness-of-fit indexes for testing measurement invariance.
Structural Equation Modeling, 9, 233–255.
Chou, C. P., & Bentler, P. M. (1995). Estimates and tests in
structural equation modeling. In R. H. Hoyle (Ed.),
Structural equation modeling: Concepts, issues, and
applications (pp. 37–55). Thousand Oaks, CA: Sage
Cohen, J. (1988). Statistical power analysis for the behavioral
sciences (2nd ed.). Hillsdale, NJ: Erlbaum.
Comrey, A. L., & Lee, H. B. (1992). A first course in factor
analysis. Hillsdale, NJ: Lawrence Erlbaum.
Cortina, J. M. (2002). Big things have small beginnings: An
assortment of “minor” methodological misunderstandings.
Journal of Management, 28, 339–362.
Evans, J., Williams, J. M. G., O’Loughlin, S., & Howells, K.
(1992). Autobiographical memory and problem solving
strategies of parasuicidal patients. Psychological Medicine,
Fabrigar, L. R., Wegener, D. T., MacCallum, R. C., & Strahan,
E. J. (1999). Evaluating the use of exploratory factor
analysis in psychological research. Psychological Methods,
Feldman, G. C., Hayes, A. M., Kumar, S. M., & Greeson, J. M.
(2003, November). Clarifying the construct of mindfulness:
Relations with emotional avoidance, over-engagement, and
change with mindfulness training. Paper presented at the
annual meeting of the Association for Advancement of
Behavior Therapy, Boston, MA.
First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. W.
(2002). Structured Clinical Interview for DSM-IV-TR Axis I
Disorders, Research Version, Patient Edition With Psychotic
Screen. New York: Biometrics Research, New York State
Floyd, F. J., & Widaman, K. F. (1995). Factor analysis in the
development and refinement of clinical assessment instru-
ments. Psychological Assessment, 7, 286–299.
Fresco, D. M., Segal, Z. V., Buis, T., & Kennedy, S. (in press).
Relationship of post treatment decentering and cognitive
reactivity to relapse in major depressive disorder. Journal of
Consulting and Clinical Psychology.
fresco et al.
Frisch, M. B. (1994). Manual and treatment guide for the
Quality of Life Inventory. Minneapolis, MN: National
Computer Systems, Inc.
Frisch, M. B., Cornell, J., Villanueva, M., & Retzlaff, P. J.
(1992). Clinical validation of the Quality of Life Inventory:
A measure of life satisfaction for use in treatment planning
and outcome assessment. Psychological Assessment, 4,
Graham, J. R. (2005). MMPI-2: Assessing personality and
psychopathology. New York: Oxford University Press.
Gross, J. J. (1998a). Antecedent- and response-focused emotion
regulation: Divergent consequences for experience, expres-
sion, and physiology. Journal of Personality and Social
Psychology, 74, 224–237.
Gross, J. J. (1998b). The emerging field of emotion regulation:
An integrative review. Review of General Psychology, 2,
Gross, J. J. (1999). Emotion regulation: Past, present, future.
Cognition and Emotion, 13, 551–573.
Gross, J. J., & John, O. P. (2003). Individual differences in two
emotion regulation processes: Implications affect, relation-
ships, and well-being. Journal of Personality and Social
Psychology, 85, 348–362.
Gross, J. J., & Levenson, R. W. (1993). Emotional suppression:
Physiology, self-report, and expressive behavior. Journal of
Personality and Social Psychology, 64, 970–986.
Gross, J. J., & Muñoz, R. F. (1995). Emotion regulation and
mental health. Clinical Psychology: Science and Practice, 2,
Hamilton, M. (1960). A rating scale for depression. Journal of
Neurology, Neurosurgery, and Psychiatry, 23, 56–62.
Hayes, S. C., Strosahl, K. D., & Wilson, K. G. (1999).
Acceptance and Commitment Therapy: An experiential
approach to behavior change. New York: The Guilford
Hayes, S. C., Strosahl, K. D., Wilson, K. G., Bissett, R. T.,
Pistorello, J., Toarmino, D., Polusny, M. A., Dykstra, T. A.,
Batten, S. V., Bergan, J., Stewart, S. H., Zvolensky, M. J.,
Eifert, G. H., Bond, F. W., Forsyth, J. P., Karekla, M., &
McCurry, S. M. (2004). Measuring experiential avoidance:
A preliminary test of a working model. The Psychological
Record, 54, 553–578.
Hoyle, R. H., & Panter, A. T. (1995). Writing about structural
equation models. In R. H. Hoyle (Ed.), Structural equation
modeling: Concepts, issues, and applications (pp. 158–176).
Thousand Oaks, CA: Sage Publications.
Hunsley, J., & Mash, E. J. (in press). Guidelines for rating
psychometric properties of assessment instruments included
in summary tables. In: J. Hunsley and E. J. Mash (Eds.), A
guide to assessments that work. New York: Oxford
Ingram, R. E., & Hollon, S. D. (1986). Cognitive therapy for
depression from an information processing perspective. In
R. E. Ingram (Ed.), Information processing approaches to
clinical psychology (pp. 259–281). San Diego: Academic
Inhelder, B., & Piaget, J. (1958). The growth of logical thinking
from childhood to adolescence. New York: Basic Books.
Jöreskog, K. G., & Sörbom, D. (1988). PRELIS: A program for
multivariate data screening and data summarization. A
preprocessor for LISREL. Mooreseville, IN: Scientific
Lyubormirsky, S., & Nolen-Hoeksema, S. (1995). Effects of
self-focused rumination on negative thinking and interper-
sonal problem solving. Journal of Personality and Social
Psychology, 69, 176–190.
Lyubormirsky, S., Tucker, K. L., Caldwell, N. D., & Berg, K.
(1999). Why ruminators are poor problem solvers: Clues
from the phenomenology of dysphoric rumination. Journal
of Personality and Social Psychology, 77, 1041–1060.
Ma, S. H., & Teasdale, J. D. (2004). Mindfulness-based
cognitive therapy for depression: Replication and explora-
tion of differential relapse prevention effects. Journal of
Consulting and Clinical Psychology, 72, 31–40.
MacCallum, R. C., & Tucker, L. R. (1991). Representing
sources of error in the common-factor model: Implications
for theory and practice. Psychological Bulletin, 109,
Moore, R. G., Hayhurst, H., & Teasdale, J. D. (1996). Measure
of Awareness and Coping in Autobiographical Memory:
Instruction for administering and coding. Unpublished
manuscript, University of Cambridge: Cambridge, MA.
Morrow, J., & Nolen-Hoeksema, S. (1990). Effects of
responses to depression on the remediation of depressive
affect. Journal of Personality and Social Psychology, 58,
Nolen-Hoeksema, S. (1998). The other end of the continuum:
The costs of rumination. Psychological Inquiry, 9, 216–219.
Nolen-Hoeksema, S., & Morrow, J. (1991). Effects of rumina-
tion and distraction on naturally occurring depressed mood.
Cognition and Emotion, 7, 561–570.
Nunnally, J. C. (1967). Psychometric theory. New York:
Rabkin, J. G., & Klein, D. F. (1987). The clinical measurement
of depressive disorders. In A. Marsella, R. Hirschfeld, & M.
Katz (Eds.), The measurement of depression (pp. 30–83).
New York: The Guilford Press.
Rozeboom, W. (1966). Foundations of the theory of prediction.
Homewood, IL: The Porsey Press.
Rubio, D. M., & Gillespie, D. F. (1995). Problems with error in
structural equation models. Structural Equation Modeling,
Safran, J. D., & Segal, Z. V. (1990). Interpersonal process in
cognitive therapy. New York: Basic Books.
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: The Guilford
Sidley, G. L., Whitaker, K., Calam, R. M., & Wells, A. (1997).
The relationship between problem solving and autobiogra-
phical memory in parasuicidal patients. Behavioural and
Cognitive Psychotherapy, 25, 195–202.
Sprinkle, S. D., Lurie, D., Insko, S. L., Atkinson, G., Jones,
G. L., Logan, A. R., & Bissada, N. N. (2002). Criterion
validity, severity cut scores, and test-retest reliability of the
Beck Depression Inventory-II in a university counseling
center sample. Journal of Counseling Psychology, 49,
Tabachnick, B. G., & Fidell, L. S. (2001). Using multivariate
statistics. Boston: Allyn & Bacon.
Teasdale, J. D., Moore, R. G., Hayhurst, H., Pope, M.,
Williams, S., & Segal, Z. V. (2002). Metacognitive aware-
ness and prevention of relapse in depression: Empirical
evidence. Journal of Consulting and Clinical Psychology, 70,
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 Consulting and Clinical
Psychology, 68, 615–623.
Treynor, W. F. C., Gonzalez, R., & Nolen-Hoeksema, S. (2003).
Rumination reconsidered: A psychometric analysis. Cogni-
tive Therapy and Research, 27, 247–259.
Ventura, J., Liberman, R. P., Green, M. F., Shaner, A., & Mintz,
J. (1998). Training and quality assurance with Structured
Clinical Interview for DSM-IV (SCID-I/P). Psychiatry
Research, 79, 163–173.
Watkins, E., Teasdale, J. D., & Williams, J. M. G. (2000).
Decentering and distraction reduce overgeneral autobiogra-
phical memory in depression. Psychological Medicine, 30,
Watson, D., & Clark, L. A. (1991). The Mood and Anxiety
Symptom Questionnaire (MASQ). Unpublished manuscript.
University of Iowa, Iowa City.
RECEIVED: November 4, 2005
ACCEPTED: August 31, 2006
Available online 24 April 2007
fresco et al.