Shame in patients with narcissistic personality disorder

Abstract

Shame has been described as a central emotion in narcissistic personality disorder (NPD). However, there is a dearth of empirical data on shame in NPD. Patients with NPD (N=28), non-clinical controls (N=34) and individuals with borderline personality disorder (BPD, N=31) completed self-report measures of state shame, shame-proneness, and guilt-proneness. Furthermore, the Implicit Association Test (IAT) was included as a measure of implicit shame, assessing implicit shame-self associations relative to anxiety-self associations. Participants with NPD reported higher levels of explicit shame than non-clinical controls, but lower levels than patients with BPD. Levels of guilt-proneness did not differ among the three study groups. The implicit shame-self associations (relative to anxiety-self associations) were significantly stronger among patients with NPD compared to nonclinical controls and BPD patients. Our findings indicate that shame is a prominent feature of NPD. Implications for diagnosis and treatment are discussed.

Full text

Shame in patients with narcissistic personality disorder
Kathrin Ritter
a,b,
n
, Aline Vater
a,c,d
, Nicolas Rüsch
e
, Michela Schröder-Abé
d
, Astrid Schütz
f
,
Thomas Fydrich
g
, Claas-Hinrich Lammers
h
, Stefan Roepke
c,
n
a
Department of Psychiatry, Charité –Universitätsmedizin Berlin, Campus Benjamin Franklin, Berlin, Germany
b
Department of Educational Science and Psychology, Freie Universität Berlin, Berlin, Germany
c
Cluster of Excellence “Languages of Emotion”, Freie Universität Berlin, Berlin, Germany
d
Institute of Psychology, Technische Universität Darmstadt, Germany
e
Department of Psychiatry II, Ulm University, Germany
f
Department of Psychology, Otto-Friedrich-Universität Bamberg, Germany
g
Institute of Psychology, Humboldt University Berlin, Berlin, Germany
h
Asklepios Clinic North –Ochsenzoll, Hamburg, Germany
article info
Article history:
Received 27 October 2012
Received in revised form
18 November 2013
Accepted 20 November 2013
Keywords:
Narcissistic personality disorder
Shame
Shame-proneness
Implicit association test
Borderline personality disorder
Guilt
abstract
Shame has been described as a central emotion in narcissistic personality disorder (NPD). However, there
is a dearth of empirical data on shame in NPD. Patients with NPD (N¼28), non-clinical controls (N¼34)
and individuals with borderline personality disorder (BPD, N¼31) completed self-report measures of
state shame, shame-proneness, and guilt-proneness. Furthermore, the Implicit Association Test (IAT) was
included as a measure of implicit shame, assessing implicit shame-self associations relative to anxiety-
self associations. Participants with NPD reported higher levels of explicit shame than non-clinical
controls, but lower levels than patients with BPD. Levels of guilt-proneness did not differ among the
three study groups. The implicit shame-self associations (relative to anxiety-self associations) were
significantly stronger among patients with NPD compared to nonclinical controls and BPD patients. Our
findings indicate that shame is a prominent feature of NPD. Implications for diagnosis and treatment are
discussed.
&2013 Elsevier Ireland Ltd. All rights reserved.
1. Introduction
Whilst the DSM-IV-TR (APA, 2000)defines narcissistic person-
ality disorder (NPD) foremost by a sense of grandiosity, clinical
literature depicts a paradoxical combination of grandiosity and
vulnerability (Dickinson and Pincus, 2003;Pincus and Lukowitsky,
2010;Ronningstam, 2010;Miller et al., 2010). A prominent clinical
feature of narcissistic vulnerability is the patient's propensity to
suffer from feelings of intense shame (Dickinson and Pincus, 2003)
which has only been recognized as an associated feature of NPD in
the DSM-IV-TR. However, empirical data on shame in NPD are very
limited.
In the current manuscript, the term NPD refers to clinical cases
as defined by the most up-to-date edition of the DSM available at
the time the research was conducted. ‘Pathological narcissism’
refers to clinical descriptions or constructs that may not overlap
completely with the DSM definition and often extend beyond this
definition, e.g. by acknowledging grandiose and vulnerable facets
or proposing a regulatory etiological model (e.g., Kohut, 1971;
Kernberg, 1975,2009;Horowitz, 2009;Ronningstam, 2010).
Throughout the present manuscript, the term ‘narcissism’refers
to cases from non-clinical samples (mainly assessed with the
Narcissistic Personality Inventory, (NPI); Raskin and Terry, 1988)
and definitions from social psychology (e.g., Morf and Rhodewalt,
2001;Tracy and Robins, 2004).
Shame encompasses an emotion resulting from a negative evalua-
tion of the stable, global self, elicited by a perceived failure (Lewis,
1971 ;Tangney and Dearing, 2002). Explicit shame is defined as a
deliberative, reflected emotional response towards negative evalua-
tions of the self and is assessed with direct self-report measures
(e.g., Lewis, 1971). Implicit shame is an automatic, overlearned,
presumably non-conscious emotional response and is assessed with
indirect measures (Greenwald and Banaji, 1995;Fazio and Towles-
Schwen, 1999;Pelham and Hetts, 1999;Rüsch et al., 2007b). Further-
more, shame is often associated with characteristic bodily postures
(e.g., posture that make the body appear smaller), head movements
(e.g., head tilting down or to the side), covering the face with the hand
and downcast eye-gaze (Keltner and Buswell, 1996).
The initial introduction of NPD in the DSM III was largely
influenced by psychoanalytic theories that describe shame as a core
emotioninnarcissisticpsychopathology(e.g.,Morrison, 1983). For
instance, Kohut (1971,1977) views shame as a prominent clinical
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/psychres
Psychiatry Research
0165-1781/$ - see front matter &2013 Elsevier Ireland Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.psychres.2013.11.019
n
Correspondingauthors at: Departmentof Psychiatry, Charité –Universitätsmedizin
Berlin, Campus Benjamin Franklin, Eschenallee 3, 14050 Berlin, Germany.
Tel. : þ49 30 8445 8796; fax: þ49 30 8445 8757.
E-mail addresses: kathrin.ritter@fu-berlin.de (K. Ritter),
stefan.roepke@charite.de (S. Roepke).
Please cite this article as: Ritter, K., et al., Shame in patients with narcissistic personality disorder. Psychiatry Research (2013), http://dx.
doi.org/10.1016/j.psychres.2013.11.019i
Psychiatry Research ∎(∎∎∎∎)∎∎∎–∎∎∎

feature in pathological narcissism. According to his view, children
carry egocentric narcissistic needs that are tempered through
empathic, realistic mirroring by their parents. Kohut hypothesized
that repeated negative parental evaluations in childhood leads to
increased shame reactivity in narcissistic patients. Moreover, due to a
lack of empathic parental responses, narcissistic patients never move
beyond earlier narcissistic developmental states that are characterized
by narcissistic needs (i.e., need to receive excessive attention). Accord-
ing to Kohut's theory, individuals with pathological narcissism avoid
frequent experiences of shame by reacting with rage or withdrawal. In
line with this theory, Kernberg (1975) hypothesized that narcissistic
patients suffer from negative interactions with primary nurturing
figures. In contrast to Kohut, Kernberg does not consider pathological
narcissism as a normal developmental stage. Instead, he proposes that
negative parental interaction fosters narcissistic features that are
characterized by unconscious negative self-representations that are
strongly connected to the experience of explicit shame. In later life,
acquired grandiose self-representations may conflict with implicit
feelings of inferiority that are strongly connected to affective expe-
riences of shame. Consequently, narcissistic patients use defense
mechanisms that limit feelings of explicit shame in response to
failures.
Further, theories from social-psychology also focus on self-
regulatory processes as a core element of narcissism (e.g., Morf and
Rhodewalt, 2001). Shame has been hypothesized as the central
emotional component in this process (e.g., Tracy and Robins, 2004).
According to theory, shame, as well as guilt, is elicited when an
individual attributes the cause of a negative event to internal factors
(e.g., Lewis, 1971;Tracy and Robins, 2004). Thus, shame and guilt are
elicited by a common set of cognitive processes. However, shame
involves negative feelings about the dispositional (internal), stable and
global self, whereas guilt involves specific, internal attribution patterns
in response to failures (Tangney and Dearing, 2002;Tracy and Robins,
2004;Hasson-Ohayon et al., 2012). Thus, according to theory, indivi-
duals who tend to frequently engage in internal, global attributions
when experiencing negative events should be more shame-prone. In
contrast, individuals who tend to make more specific, internal attribu-
tions when experiencing a negative event are said to be more guilt-
prone. Tracy and Robins (2004) proposed that the experience of
shame (but not guilt) is the central feature of narcissistic individuals.
The authors hypothesized that increased shame-proneness in narcis-
sistic individuals is related to self-esteem discrepancies, i.e., verbally
expressed grandiose self-views that contradict unconscious feelings of
insecurity. In their view, narcissistic individuals are more self-focused
and use different regulation strategies to prevent unconscious feelings
of low self-esteem from becoming explicit, and thereby, experience
explicit shame (e.g., appraise negative events as irrelevant to identity
goals or attribute failure externally and become angry or aggressive)
(Tracy and Robins, 2004).
Numerous empirical studies demonstrated that shame is in
general more maladaptive than guilt (Tangney et al., 1992;Tracy
and Robins, 2004). With respect to psychopathology, several studies
provide evidence that shame and psychiatric impairment are
strongly associated (e.g., depression, Andrews, 1995;posttraumatic
stress disorder, Andrews et al., 2000;socialphobia,Browning, 2005;
borderline personality disorder, Rüsch et al., 2007b; reaction after
negative live events, Uji et al., 2012; caregivers' distress, Weisman de
Mamani, 2010).
Current clinical conceptualizations of pathological narcissism also
propose a regulatory etiological model (e.g., Horowitz, 2009;Kernberg,
2009;Ronningstam, 2010). Grandiose and vulnerable facets in patho-
logical narcissism can be understood as consequences of attempts to
regulate self and self-esteem (e.g., Ronningstam, 2010). According to
Ronningstam, individuals with pathological narcissism can fluctuate
between grandiosity and vulnerability depending on external or
internal factors. Intense feelings of explicit shame belong to the
vulnerable features of pathological narcissism and occur, for example,
as response to negatively perceived events. As individuals with
pathological narcissism try to avoid these intense feelings of shame,
they engage in various intrapersonal and interpersonal strategies in
order to prevent explicit shame (e.g., devaluation of others, responding
with anger, and self-enhancement). Further, Ronningstam (2010)
emphasizes that perfectionism is a significant feature of self-
enhancement that is closely related to shame in pathological narcis-
sism. When perfectionism is not sufficient enough to bridge the gap
between real abilities and ideal imaginations about the self, feelings of
explicit shame are especially likely to be elicited.
In summary, shame is a central feature of non-clinical and
pathological narcissism in several theoretical models that might be
relevant for the future definition of NPD. Until the present, there
have only been a few studies on shame and narcissism, and these
have relied on non-clinical or mixed clinical populations. For
instance, explicit shame and narcissism (assessed with the NPI)
are negatively correlated in non-clinical individuals (Gramzow and
Tangney, 1992;Watson et al., 1996;Pincus et al., 2009). However, a
recent study suggests that the NPI measures a grandiose variant of
(normal or subclinical) narcissism that strongly overlaps with
(high explicit) self-esteem (Vater et al., 2013b). Thus, the NPI is
likely not appropriate for assessing pathological narcissism in
clinical research on NPD. Another study used a more valid measure
to assess pathological narcissism, the Pathological Narcissism
Inventory (PNI, Pincus et al., 2009). The authors found a moder-
ately positive correlation between explicit shame and pathological
narcissism in a mixed clinical sample (PNI, Pincus et al., 2009).
These data emphasize the importance of differentiating non-
clinical and pathological narcissism, especially when assessing
vulnerable facets of the disorder.
The overall aim of this study was to provide evidence of altered
implicit and explicit shame in patients with NPD compared to
controls. To our knowledge, this is the only study that assessed
shame in a clinical sample of patients with NPD.
The first aim of this study was to assess explicit shame-
proneness and state shame in patients with NPD compared to
non-clinical controls. Explicit shame-proneness is a conscious,
self-reported tendency to react with shame towards external
events. Building upon theory and previous empirical findings of
shame in pathological narcissism (provided above), we hypothe-
sized that patients with NPD score higher on explicit state shame
and explicit shame-proneness compared to non-clinical controls.
Second, existing studies on shame and narcissism exclusively
assessed shame with self-report measures. Building upon clinical
theories of NPD that propose high levels of not necessarily
conscious shameful reactions in patients with NPD (see above),
we hypothesized that patients with NPD show higher levels of
implicit shame than non-clinical controls.
Third, shame and guilt are the two possible emotional
responses in reaction to perceived failures. Several theories indi-
cate that narcissistic individuals are more shame-prone than guilt-
prone (e.g., Tracy and Robins, 2004;Martens, 2005). Thus, we
hypothesize that patients with NPD do not differ significantly in
guilt-proneness from non-clinical controls. By doing so, we aim to
provide initial evidence that shame (but not guilt) is a central self-
conscious emotion of NPD.
Fourth, and in order to investigate specificity, we included a
clinical comparison group of inpatients with borderline person-
ality disorder (BPD). We decided to include this clinical group as
shame has previously been described as a prominent clinical
feature in BPD (Crowe, 2004;Brown et al., 2009). Moreover, prior
empirical data indicate that BPD patients had higher explicit levels
of shame-proneness, state shame, and stronger implicit shame-self
(relative to anxiety-self) associations in comparison to non-clinical
controls (Rüsch et al., 2007b). Furthermore, we used BPD as a
K. Ritter et al. / Psychiatry Research ∎(∎∎∎∎)∎∎∎–∎∎∎2
Please cite this article as: Ritter, K., et al., Shame in patients with narcissistic personality disorder. Psychiatry Research (2013), http://dx.
doi.org/10.1016/j.psychres.2013.11.019i

comparison group as both are personality disorders, show a high
comorbidity rate (Westen et al., 2006) and overlap in symptoms
such as affect dysregulation (Blais et al., 1997). Thus, by comparing
NPD to this “near neighbor”disorder, we tested whether heigh-
tened shame is specific to NPD patients or a characteristic of
psychopathology in patients with personality disorder in general.
2. Method
2.1. Participants and procedure
We recruited 53 patients diagnosed with NPD (according to DSM-IV, APA,
2000), 31 patients diagnosed with BPD (according to DSM-IV, APA, 2000), and 34
non-clinical controls. All patients were enrolled in a broad, multicenter clinical
study on NPD at the Department of Psychiatry, Charité Berlin, and cooperating
hospitals and outpatient settings in Germany. Axis-II diagnosis was assessed using
the interview section of the German Structured Clinical Interview for DSM-IV-
Personality Disorders (SCID-II, First et al., 1997;Fydrich et al., 1997). Two NPD
patients were outpatients, the remaining NPD and all BPD patients were inpatients.
All patients fulfilling criteria for both diagnoses NPD and BPD were excluded
(N¼25). All interviews were conducted by a trained psychiatrist and three trained
psychologists. Interrater reliabilities of SCID-II Personality Disorder (PD) diagnoses
were acceptable with κ¼0.80 for NPD and κ¼0.82 for BPD. Internal consistencies
for SCID-II PD diagnoses (sum of criteria) were acceptable with Cronbach's α¼0.86
for NPD and Cronbach's α¼0.88 for BPD in the current study.
Axis-I comorbidity for the NPD sample was assessed with the German
Structured Clinical Interview for DSM-IV Axis-I Psychiatric Disorders (SCID-I, First
et al., 1996;Wittchen et al., 1997). The German Mini International Neuropsychiatric
Interview (M.I.N.I., Sheehan et al.,1998;Ackenheil et al.,1999) was used for the BPD
sample. We only included comparable sections of the M.I.N.I. and SCID-I because of
different diagnostic sections in both interviews. Exclusion criteria for all patients
included a history of psychotic disorder, current mania or hypomania, current
substance-induced disorder, mental retardation (IQo80) assessed with subtest 4 of
a German test of cognitive performance (Leistungsprüfsystem, LPS, Horn, 1983), or
being a non-native speaker. Fifty-seven percent (N¼16) of NPD and 61% (N¼19) of
BPD patients were treated with psychotropic medication (antipsychotics, antide-
pressants, mood stabilizers). No patients were treated with benzodiazepines.
Thirty-four non-clinical comparison subjects were recruited via media adver-
tisements in Berlin, Germany. Non-clinical controls were only included if they had
no current or lifetime Axis-I disorder (SCID-I screening, First et al., 1996;Wittchen
et al., 1997), fulfilled no more than three criteria for any PD section in the SCID-II
questionnaire (First et al., 1997;Fydrich et al., 1997), exhibited general intellectual
functioning within the normal range (IQ480), and were native speakers of
German. For sociodemographic and clinical data see Table 1. All procedures were
approved by the Human Subjects and Ethics Committee of Charité Berlin. Written
informed consent was obtained from each participant.
2.2. Measures
2.2.1. Implicit shame
Implicit shame was assessed using the IAT (Greenwald et al., 1998;Rüsch et al.,
2007b). As the IAT is a measure that uses reaction times to determine relative
strength of associations (Greenwald et al., 1998;Greenwald and Farnham, 2000)
between the self-concept and the attribute category, an equivalent comparison
category for shame was required. As in a previous study that used this approach
(Rüsch et al., 2007b), we selected anxiety as a reference for shame since both are
negative emotions. Thus, we conservatively assessed implicit shame-self associa-
tions relative to associating oneself with another negative emotion (anxiety). We
did not select a positive or neutral attribute category for comparison with shame as
that would have led to interpretation difficulties; that is, it would not be clear
whether we were assessing an effect specifically related to shame or an effect
related to negative effect in general. Thus, we controlled for the possibility that
subjects may associate themselves with negative emotions or attributes in general,
rather than with shame in particular. Target categories were me vs. not me, attribute
categories were shame vs. anxiety. The reliability (split-half, Spearman–Brown) of
the IAT in the present study was good r¼0.73 (Po0.001).
During the computerized test, target category labels (me vs. not me) represent-
ing the self were displayed in the upper right and left corners of the screen and
assigned to a left or right response key. The subjects were asked to assign stimuli
(e.g. own name vs. others name) presented in the center of the screen to one of two
response keys. Participants were instructed to make their judgments as quickly, but
as accurately, as possible. If an incorrect key was pressed, a red X appeared in the
center of the screen, and the next item did not appear until one of the correct keys
had been pressed. All participants completed five blocks of category judgments
(Table 2). In the first block of trials, classification of the target categories (me vs. not
me) took place, and in the second block, classification of the attribute categories
(shame vs. anxiety). The subjects were asked again to assign stimuli (e.g. ashamed
vs. fearful) presented in the center of the screen to one of two response keys. In
block 3, both classification tasks targets and attributes were combined. In block 4,
the key assignments for the target category were reversed as compared to block 1
(Table 2) and the target and attribute discriminators were inversely recombined in
block 5 (i.e., if shame and me had shared a response key in block 3, shame and not
me shared a response key in block 5, and vice versa). Practice blocks (1, 2, and 4)
consisted of 20 trials each, critical blocks (3 and 5) consisted of 20 practice trials
and 40 test trials.
Sample items were ashamed, embarrassed, shameful, self-conscious (shame),
anxious, fearful, nervous, uncertain, afraid (anxiety), first name, family name, year
of birth (me), others' first names, others' family names, or others' years of birth (not
me). Before the IAT started, the participants were asked to enter their first name,
family name, and year of birth and choose one out of four first names, four family
names, and four birthdays that did not relate to or concern them. Female
participants were shown female names (Brigitte, Susanne, Claudia, Johanna), and
male participants were shown male names (Wolfgang, Matthias, Thorsten, Johan).
For each trial, the computer recorded reaction time in milliseconds from the
appearance of the stimuli to the correct response. Based on the assumption that
quicker processing reflects stronger associations, these reaction times were used to
determine the relative strength of implicit associations between self- and other
concepts on the one hand, and shame and anxiety attributes on the other by
computing the D-algorithm (Greenwald et al., 2003). In this algorithm, all trials
(including practice phases) of the combined blocks (block 3 and 5) are included.
The mean reaction time of the shame-self combination was subtracted from the
mean reaction of the anxiety-self combination, and this difference was divided by a
personalized standard deviation of the combination phases, which ensured that the
measurement was not influenced by differences in response speed between
participants. This approach is thus optimal for comparisons between clinical and
non-clinical groups. Positive scores represent a stronger association between self
and shame (relative to anxiety). Negative values indicate a stronger association
between self and anxiety (relative to shame).
No participant had to be excluded due to latencies (between 30 0 ms and
3000 ms) or error rate (20% or more). Error rates were low in all three groups
(o4%) and did not differ significantly between groups.
2.2.2. Explicit state shame
We used the German version of the Experiential Shame Scale (ESS, Turner,
1998;Rüsch et al., 2007a) which was designed to evaluate explicit state shame. It is
composed of 11 semantic differential items assessing physical, emotional, and
social markers of shame experiences “in this moment”on a scale from 1 to 7.
Sample items are: “Physically, I feel: 1, pale,to7,flushed”,“Emotionally, I feel: 1,
content,to7,distressed”,“Socially, I feel like: 1, hiding,to7,being sociable
[reversed]”. In the current study, the ESS demonstrated acceptable internal
consistency (Cronbach's α¼0.75). Construct validity of the ESS has been shown
(Turner and Waugh, 2001).
2.2.3. Explicit shame-proneness and guilt-proneness
The German version of the Test of Self-Conscious Affects version 3 (TOSCA-3,
Tangney et al., 2000;Rüsch et al., 2007a) was used as a self-rating measure to
determine degrees of explicit shame-proneness and explicit guilt-proneness. It
consists of a series of 16 brief positive and negative scenarios each representing a
different affective tendency (e.g. guilt, shame) and associated potential responses.
Responses are rated from 1¼not likely to 5 ¼very likely. Sample scenario: “You are
driving down the road and you hit a small animal. (A) You would think the animal
shouldn't have been on the road. (B) You would think: ‘I’m terrible.’(C) You would
feel: ‘Well, it's an accident.’(D) You'd feel bad you hadn't been more alert driving
down the road.”Cronbach's alpha was α¼0.80 for shame-proneness and α¼0.79
for guilt-proneness. Construct validity has been demonstrated (Tangney et al.,
2000;Rüsch et al., 2007a).
2.2.4. Explicit state and trait anxiety
In the study we used a relative Shame-Anxiety-IAT. As anxiety was our
reference category, we provide additional evidence on explicit anxiety. By doing
so, we aimed at examining whether anxiety is a prevalent emotion in NPD.
Moreover, information on explicit anxiety may assist to rule out the possibility
that the IAT effect is solely explained by lower anxiety. Therefore, we used the State
Trait Anxiety Inventory (STAI, Spielberger et al., 1970) for measuring state anxiety
(20 items) and trait anxiety (20 items). Items are rated on a scale from 1¼not at all
to 4¼very much. Internal consistency was excellent for state anxiety (Cronbach's
α¼0.94) and trait anxiety (Cronbachand social markers of shame exps α¼0.94).
The experimental procedure was as follows: after applying inclusion and
exclusion criteria, axis I and II comordidity were assessed. Psychometric scales,
including state measures, were applied directly before the IAT. After the IAT,
patients were debriefed.
K. Ritter et al. / Psychiatry Research ∎(∎∎∎∎)∎∎∎–∎∎∎ 3
Please cite this article as: Ritter, K., et al., Shame in patients with narcissistic personality disorder. Psychiatry Research (2013), http://dx.
doi.org/10.1016/j.psychres.2013.11.019i

2.3. Statistical analyses
All statistical analyses were conducted with IBM SPSS Statistics 20 (SPSS Inc., 2011).
Pearson product moment correlations were computed for correlations between the
measures. We used Student's ttests, Welch's ttests, Analyses of Variance (ANOVAs), and
Pearson's χ
2
tests for group comparisons of socio-demographic and clinical variables.
As age and gender differed significantly between study groups, we assessed main
and interaction effects by including both variables as covariates in all analyses. For group
comparisons of self-ratings, we used single factor analyses of covariance (ANCOVAs) and
post hoc pairwise comparisons, which were based on estimated marginal means with
covariates. All analyses were two-tailed and the alpha level was set at Po0.05. We used
partial eta-squared (η
2
p
) as the effect size for the ANCOVAs (Pierce et al., 2004)and
Bonferroni adjustment for multiple comparisons.
3. Results
3.1. Correlations
Table 3 presents the intercorrelations of explicit and implicit
shame, explicit guilt and anxiety variables. Consistent with previous
findings (Hofmann et al., 2005;Krizan and Suls, 2008) shame-prone
implicit self-concept and explicit variables were uncorrelated.
3.2. Group differences
For demographic and clinical data, see Table 1. As there were
significant differences in gender and age between the three study
groups, we used both variables as covariates in the following
ANCOVAs.
All ANCOVAs (with the exception of guilt-proneness) revealed
significant main effects of group and are shown in Table 4. NPD
patients had significantly higher scores in explicit state shame,
explicit shame-proneness, explicit state and trait anxiety than
non-clinical controls and significantly lower scores than BPD
patients. The three study groups did not differ significantly in
guilt-proneness. In all ANCOVAs, the covariates of gender and age
revealed neither significant main nor interaction effects.
An ANCOVA with the IAT D-score (representing shame-self
associations relative to anxiety-self associations) as the dependent
variable and gender and age as covariates revealed a significant
Table 1
Sociodemographic and clinical variables of patients with narcissistic personality disorder, borderline personality disorder, and non-clinical comparison subjects.
1. NPD (N¼28) 2. BPD (N¼31) 3. NCC (N¼34) Statistical test
M(S.D.) M(S.D.) M(S.D.) Statistics Pvalue
Age (years) 37.46 (9.95) 28.77 (8.43) 31.61 (14.00) F¼4.606, d.f.¼2 0.012
Fluid intelligence
a
113.82 (12.09) 116.68 (12.31) 121.35 (11.76) F¼2.673, d.f.¼2 0.075
Number of comorbid diagnoses 2.71 (1.80) 3.14 (1.94) n.a. t¼0.857, d.f.¼54 0.395
N(%) N(%) N(%)
Women 9 (32.1) 25 (80.7) 16 (47.1) χ
2
¼14.892, d.f.¼2 0.001
Any affective disorder 16 (57.1) 23 (74.2) n.a. χ
2
¼1.909, d.f.¼10.167
MDE current 13 (46.4) 9 (29.0) n.a. χ
2
¼1.198, d.f.¼1 0.274
MDE lifetime 15 (53.6) 16 (51.6) n.a. χ
2
¼0.015, d.f.¼1 0.903
Dysthymia 4 (14.3) 7 (22.6) n.a. Fisher: χ
2
¼1.018, d.f. ¼1 0.313
Any substance use disorder 11 (39.3) 16 (51.6) n.a. χ
2
¼0.901, d.f.¼1 0.343
Any anxiety disorder 8 (28.6) 15 (48.4) n.a. χ
2
¼2.429, d.f.¼10.119
PTSD 2 (7.1) 10 (32.3) n.a. χ
2
¼6.788, d.f.¼1 0.020
Any eating disorder 4 (14.3) 11 (35.5) n.a. Fisher: χ
2
¼3.487, d.f.¼1 0.078
Any cluster A PD 7(25.0) 3 (9.7) n.a. Fisher: χ
2
¼1.791, d.f. ¼1 0.293
Any other cluster B PD
b
6 (21.4) 2 (6.5) n.a. Fisher: χ
2
¼2.538, d.f.¼10.142
Any cluster C PD 7 (25.0) 8 (25.8) n.a. χ
2
¼0.153, d.f.¼1 0.696
Without psychotropic medication 12 (42.9) 12 (38.7) n.a. Fisher: χ
2
¼0.540, d.f.¼1 0.584
Note. NPD¼narcissistic personality disorder, BPD ¼borderline personality disorder, NCC¼non-clinical controls, MDE ¼major depression episode, PTSD¼posttraumatic stress
disorder, PD¼personality disorder, IAT¼implicit association test, n.a. ¼not applicable.
a
LPS¼Leistungsprüfsystem.
b
Without BPD and NPD.
Table 2
Design of the Implicit Association Test for the assessment of implicit shame-self associations relative to implicit anxiety-self associations.
Adapted from Greenwald et al., 1998.
Task description Block 1 Block 2 Block 3 Block 4 Block 5
Target discrimination Attribute discrimination Initial combined task Reversed target discrimination Reversed combined task
Task instruction ●Me Me ●
●Me ●Shame ●Shame Me ●●Shame
Not me ●Anxiety ●Not me ●●Not me ●Not me
Anxiety ●Anxiety ●
Sample stimuli ●Julia Meyer●
●Julia ●Ashamed ●Shame Julia ●●Embarrassed
Brigitte ●Anxious ●1950●●Brigitte ●1950
Anxiety ●Fear●
Number of trials 20 20 20þ40 20 20þ40
Note. Assignment to the left or right response key is indicated by black circles. The sample items are examples of idiographic stimuli for a subject called Julia Meyer, born in
1979, who is not familiar with a person called Brigitte Franke, born in 1950. (For detailed description of the IAT see Section 2.)
K. Ritter et al. / Psychiatry Research ∎(∎∎∎∎)∎∎∎–∎∎∎4
Please cite this article as: Ritter, K., et al., Shame in patients with narcissistic personality disorder. Psychiatry Research (2013), http://dx.
doi.org/10.1016/j.psychres.2013.11.019i

main effect of group. No significant main effect of age was
observed. Gender had a significant main effect on the IAT score,
F(1,87)¼9.19, P¼0.003, η
2
p
¼0.096 with females showing stronger
shame-self associations (relative to anxiety-self associations) than
males. Post hoc analyses revealed significantly higher D-scores for
patients with NPD in comparison to non-clinical controls and
significantly higher D-scores for patients with NPD in comparison
to BPD patients. This indicated stronger shame-self associations
(relative to anxiety-self associations) for NPD patients than for
controls and for BPD patients. The data showed no significant
difference in D-scores between BPD patients and non-clinical
controls (Table 4,Fig. 1).
Medication status as an additional co-variate in all ANCOVAs
(state shame, explicit shame- and guilt-proneness, implicit shame-
proneness, state and trait anxiety) revealed neither a significant
main effect nor a significant moderator effect of medication status.
All main effects of group did not change when medication status
was included as a co-variate.
4. Discussion
The present study aimed to investigate explicit and implicit
shame in patients with NPD and, to the best of our knowledge,
presents the first empirical data on shame in a clinical sample of
patients with NPD. According to our data, explicit measures of
state shame and shame-proneness were significantly higher in
NPD patients compared to non-clinical controls, but significantly
lower compared to BPD patients. However, NPD patients carried
the highest levels of implicit shame-self associations (relative to
anxiety-self associations) compared to both control groups. Expli-
cit guilt-proneness did not significantly differ between NPD
patients, BPD patients and non-clinical controls. In the following,
we discuss these findings with reference to current conceptualiza-
tion of NPD.
Our data provide the first empirical evidence that patients with
NPD, compared to non-clinical controls, show higher explicit state
shame and shame-proneness, as hypothesized by clinical theories
(e.g., Martens, 2005;Ronningstam, 2010,Pincus and Lukowitsky,
2010). Our results extend previous findings in a mixed clinical sample,
identifying explicit shame as a correlate of pathological narcissism
(Pincus et al., 2009). However, increased explicit shame has been
reported in a variety of psychopathologies (e.g., Andrews et al., 2000;
Browning, 2005). In accordance, we found that high explicit shame
was not specific to NPD, as BPD patients reported even more explicit
shame. Moreover, guilt has been described as a more adaptive reaction
to failure (Tracy and Robins, 2004) and has been associated more with
Table 3
Pearson product moment correlations between explicit and implicit measures in all participants (N¼93).
Implicit shame
a
State shame Shame-proneness Guilt-proneness State anxiety
State shame 0.029
Shame-proneness 0.010 0.438**
Guilt-proneness 0.017 0.064 0.431**
State anxiety 0.011 0.603** 0.540** 0.011
Trait anxiety 0.103 0.611** 0.697** 0.117 0.868**
a
Implicit shame-self associations relative to anxiety-self associations.
nn
Po0.01 (two tailed).
Table 4
Group comparisons for shame-prone self-concept (NPD vs. BPD vs. non-clinical controls; without patients with both diagnoses NPD/BPD) –IAT and self-rating
questionnaires.
Measure 1: NPD (N¼28) 2: BPD (N¼31) 3: NCC (N¼34) Analysis of covariance
b
main group effect Post hoc test (P)
c
M(S.D.) M(S.D.) M(S.D.) F(2, 87) Pη
2
p
1 vs. 2 1 vs. 3 2 vs. 3
Implicit shame (IAT)
a
0.23 (0.46) 0.11 (0.33) 0.01 (0.29) 5.83 0.004 0.118 0.027 0.005 1.00
Explicit state shame (ESS) 3.83 (0.70) 4.58 (0.86) 3.03 (0.63) 35.63 o0.001 0.450 o0.0 01 0.001 o0.001
Explicit shame-proneness (TOSCA-3) 40.96 (8.72) 49.02 (9.05) 34.18 (7.67) 19.79 o0.001 0.313 0.018 0.008 o0.001
Explicit guilt-proneness (TOSCA-3) 57.50 (7.15) 58.08 (11.46) 55.12 (8.82) 0.51 0.600 0.012 1.00 1.00 1.00
Explicit state anxiety (STAI) 2.17 (0.37) 3.06 (0.56) 1.74 (0.68) 55.04 o0.001 0.559 o0.001 0.004 o0.0 01
Explicit trait anxiety (STAI) 2.43 (0.36) 3.06 (0.50) 1.71 (0.44) 67.19 o0.001 0.607 o0.001 o0.001 o0.001
Note. NPD¼narcissistic personality disorder, BPD ¼borderline personality disorder, NCC¼non-clinical controls, ESS ¼Experiential Shame Scale, TOSCA-3 ¼Test of Self-
Conscious Affects, STAI¼State Trait Anxiety Inventory, IAT¼Implicit Association Test.
a
Positive D-scores¼stronger shame-self associations relative to anxiety-self associations.
b
Covariate in all analyses¼gender and age.
c
Bonferroni for multiple testing.
Fig. 1. Mean D-scores (IAT-effects) and 95% confidence intervals for the three study
samples. NPD¼narcissistic personality disorder, BPD¼borderline personality disorder,
NCC¼non-clinical controls; positive D-scores¼more shame-prone implicit self-concept
(compared to an anxiety prone self-concept), CI ¼confidence interval.
K. Ritter et al. / Psychiatry Research ∎(∎∎∎∎)∎∎∎–∎∎∎ 5
Please cite this article as: Ritter, K., et al., Shame in patients with narcissistic personality disorder. Psychiatry Research (2013), http://dx.
doi.org/10.1016/j.psychres.2013.11.019i

reparative behavior than with psychopathology (Tangney, 1995). In
accordance with prior assumptions, NPD patients (and also BPD
patients) did not significantly differ in guilt-proneness compared to
non-clinical controls. This finding supports the notion that explicit
shame, but not guilt, is a common feature in patients with personality
disorders. These findings replicate a previous study (Rüsch et al.,
2007b) of higher explicit shame in BPD compared to non-clinical
controls and patients with social phobia.
According to clinical theories, narcissistic individuals use dif-
ferent self-regulation strategies to avoid the experience of explicit
shame. Assessing implicit shame could therefore constitute a
fruitful complement to explicit shame measures in NPD research.
Moreover, and in line with other studies (Hofmann et al., 2005;
Rüsch et al., 2007b), we found no significant correlation between
direct (self-reports) and indirect measurements (IAT). Based on
dual process theories, non-correlation data support the assump-
tion that there are separate subsystems of information processing
for shame (Gawronski and Bodenhausen, 2006). For these reasons,
indirect measures of shame may provide useful additional infor-
mation that can complement information obtained from direct
measures such as self-report questionnaires.
Our results indicate that NPD patients implicitly associated them-
selves more strongly with shame than with anxiety (indicated by a
positive IAT score), whereas the implicit association of non-clinical
controls with anxiety was as strong as with shame (indicated by an
IAT score close to zero). Interestingly, NPD patients also had signifi-
cantly stronger shame-self associations (relative to anxiety-self asso-
ciations) than BPD patients, indicating specificity of this finding for
NPD. Two interpretations are plausible: First, shame might act as a
more specific feature of NPD on an implicit level. Second, implicit
shameisprevalentbutlessspecific to NPD, i.e., lower IAT score in BPD
compared to NPD might be related to strong anxiety-self associations
in BPD, masking shame-self associations in this patient group (for
further discussion see limitation section).
Findings of high implicit and explicit shame in NPD can be
discussed within the theoretical framework of pathological narcissism
(Pincus and Lukowitsky, 2010;Miller et al., 2010;Ronningstam, 2010):
Grandiose and vulnerable facets of pathological narcissism are
hypothesized to derive from self-esteem dysregulation, the attempt
to maintain high self-representations, i.e., prevent low implicit self-
esteem from becoming explicit (Morf and Rhodewalt, 2001;
Ronningstam, 2010). In consequence, grandiose and vulnerable facets
are hypothesized to fluctuate and help seeking clinical cases of NPD
might present with more vulnerable facets of the disorder. Recent
empirical data from inpatients with NPD support this notion by
showing that these patients score lower on explicit self-esteem than
non-clinical controls (Vater et al., 2013a).
More explicit state shame and shame-proneness in NPD patients
is a further argument that clinical cases present more vulnerable
features compared to non-clinical cases. Theories from social psy-
chology hypothesize that narcissistic individuals are more self-
focused compared to controls, and thus, more prone to self-focused
emotions (Tracy and Robins, 2004). Recent empirical data of impaired
emotional empathy (the emotional response to another person's
emotional state) in NPD compared to non-clinical controls are in line
with this assumption (Ritter et al., 2011;Schulze et al., 2013). With
regard to shame, one assumed core mechanism of vulnerability in
NPD is the dysfunctional processing of perceived failures (e.g., Tracy
and Robins, 2004). In narcissism, failures are hypothesized to be
implicitly more attributable to global failures of the self, and there-
fore, elicit shame (Tracy and Robins, 2004). Our data indicating
increased explicit state shame and shame-proneness in NPD com-
pared to non-clinical controls supports this hypothesis. Shame can be
an extremely painful emotion due to devaluation of the global self
(Lewis, 1971). Patients with pathological narcissism are hypothesized
to engage in different regulation strategies (e.g., Schoenleber and
Berenbaum, 2012). Thus, NPD patients may avoid consciously experi-
encing shame and engage in external attributions (e.g., blaming the
offender for the insult) (Lewis, 1971;Kohut, 1972;Scheff et al., 1989).
This external attribution might result in feelings of anger and hostility
(emotions that may be easier to tolerate and easier to express). In line
with clinical theories, those dysfunctional behaviors in patients with
NPD are hypothesized to act as a defense against excessive explicit
shame in specificsituations(Pincus and Lukowitsky, 2010;Miller
et al., 2010;Ronningstam, 2010;Schoenleber and Berenbaum, 2012).
Moreover, implicit shame in patients with NPD could also foster
dysfunctional behavior strategies that serve to prevent experiences of
shame. For instance, self-aggrandizement across situations or enga-
ging in pride-inducing activities (e.g., Uji et al., 2012;Schoenleber and
Berenbaum, 2012) could function as reparative mechanisms for
shameful memories associated with the narcissistic self. As a con-
sequence, cognitive self-enhancement strategies could account for
the relatively lower level of explicit shame in the NPD group
compared to BPD patients (Campbell et al., 2000;Bosson et al.,
2003). Another shame-regulation strategy is hypothesized to be
perfectionism (Ronningstam, 2010). Individuals with pathological
narcissism are hypothesized to attain excessively high standards
in order to prevent upcoming failures that could elicit shame
(Ronningstam, 2010;Schoenleber and Berenbaum, 2012).
Our finding may also be discussed in light of shame inducing
interactions with primary nurturing figures. As shame-proneness
is assumed to result from long-lasting and intense levels of shame
during an individual's development (e.g., Kohut, 1971;Claesson
and Sohlberg, 2002), one could argue that NPD patients might
have been affected by previous shameful experiences during
interactions with significant others (i.e., parents). However, there
is only sparse evidence that negative implicit evaluations develop
in reaction to such early interactions (DeHart et al., 2006). More-
over, these studies refer to narcissism in non-clinical individuals.
Future research may therefore investigate the developmental
factors associated with explicit and implicit shame in patients
with NPD.
Future studies should also take the neurocognitive deficits of NPD
patients into account. With regard to perception of failure one has to
acknowledge that recent data indicate that patients with NPD show
deviant processing of social information (e.g., emotional faces,
Marissen et al., 2012) that might contribute to misperception of
external events and falsely labeling them as negative.
Our study has some limitations. Firstly, in this study we used an
IAT that measured shame-self associations relative to anxiety-self
associations. Due to the relative nature of this IAT, we cannot rule out
the possibility that lower D-scores result either from lower implicit
shame and lower implicit anxiety or higher implicit shame and higher
implicit anxiety. Increased state and trait anxiety in NPD compared to
non-clinical controls shows that anxiety is a prevalent emotion in
NPD and thus suitable as a comparison-category relative to shame for
the IAT. Nevertheless, NPD patients had lower scores on explicit state
and trait anxiety compared to BPD patients. In consequence, lower D-
scores in BPD compared to NPD might result from a strong shame-self
and anxiety-self association. Our study design does not allow answer-
ing this question and further studies are needed. Further arguments
guided the selection of anxiety as reference category. First, by
selecting anxiety we controlled for the possibility that participants
associated themselves with negative emotions in general rather than
with shame in particular. Secondly, there are alternative negatively
valenced comparison emotions such as anger, sadness (basic emo-
tions, Ekman, 1992), but we aimed at focusing on a negative,
prevalent, and rather disorder-unspecific emotion. As anger has been
specifically associated with NPD (Kohut, 1971;Martens, 2005;Miller
et al., 2010;Schoenleber and Berenbaum, 2012) and sadness sub-
sumes shame in linguistic hierarchical classifications (Shaver et al.,
1987), we decided to use anxiety, a negative basic emotion (Ekman,
K. Ritter et al. / Psychiatry Research ∎(∎∎∎∎)∎∎∎–∎∎∎6
Please cite this article as: Ritter, K., et al., Shame in patients with narcissistic personality disorder. Psychiatry Research (2013), http://dx.
doi.org/10.1016/j.psychres.2013.11.019i

1992). Future research should investigate whether patients with NPD
have an increased level of negative emotions across a broader range
and should include narcissism-specificemotionssuchasanger,
sadness, or envy. Thirdly, the IAT used in this study was already used
as a valid measure of implicit shame-self associations relative to
anxiety-self associations in patients with BPD, patients with social
phobia and non-clinical controls (Rüsch et al., 2007b). Thus, we aimed
at building upon existing findings from patients with BPD, social
phobia and non-clinical controls. Finally, there are other implicit
measures that could have served as indicators of implicit shame
(e.g. Go/No-Go Association Task, Nosek and Banaji, 2001;Extrinsic
Affective Simon Task, De Houwer, 2003;Single-CategoryIAT,
Karpinski and Steinman, 2006). However, it has to be acknowledged
that these alternative measures for assessing implicit attitudes
performed worse with regard to validity and reliabilty (Bluemke
and Friese, 2008;Rudolph et al., 2008;Stieger et al., 2011). At present,
the two-category IAT used in this study is still the gold standard when
assessing implicit attitudes (Stieger et al., 2011).
With regard to future studies, the assessment of behavioral and
emotional consequences of shame in NPD could perhaps provide
further information about the shame-related psychopathology.
Previous research has shown that film clips or stress-induction
interviews (Dimsdale et al., 1988) can reliably activate emotional
processing that could invoke emotions and shame respectively
(Lobbestael et al., 2009).
Moreover, one could speculate that BPD patients would show a
high shame-prone implicit self-concept, as experiences of invalida-
tion in early childhood and adolescence are prominent in BPD
(Linehan, 1993). Previous findings provided evidence that females
with BPD or borderline personality features are prone to shame in
their implicit self-concept (Rüsch et al., 2007b;Hawes et al., 2013).
However, we could not replicate these previous findings indicating
that BPD patients have a more shame-prone implicit self-concept
than non-clinical controls, as the differences (see Fig. 1)didnotreach
statistical significance (Rüsch et al., 2007b). This was possibly due to a
smaller BPD sample size in the present study compared to an earlier
study (Rüsch et al., 2007b). Further, the present study included both
femaleandmaleBPDpatientsandwomenexhibitedastronger
shame-prone (relative to anxiety-prone) implicit self-concept com-
pared to men (Hawes et al., 2013). Finally, patients with comorbid
NPD were not excluded in former studies (Rüsch et al., 2007b).
Additionally, and in line with prior data (e.g., Westen et al., 2006),
we had a high rate of comorbid cases (NPD and BPD diagnosis) which
were excluded from study participation. An alternative future appro-
ach would be the inclusion of these comorbid cases and analysis of
specific dimensions of psychopathology.
Furthermore, we did not distinguish between different aspects
such as shame about one´sownbodyandshamethatisexperienced
as a result of negative social evaluations (Andrews et al., 2002;
Kämmerer, 2010).Onecouldspeculatethatbody-relatedshameis
less important to NPD than shame experienced in social situations.
Finally, the number of female NPD patients and male BPD
patients was too small to analyze group by gender interactions
using a full-factorial gender by group analysis. Further, due to the
small sample size, we could not control for axis-I or axis-II
comorbidity, although we found that BPD was significantly asso-
ciated with posttraumatic stress disorder.
The study has several implications for the diagnosis and treatment
of NPD. Here we provide the first empirical evidence that shame is a
characteristic feature of NPD which might have diagnostic relevance.
The current DSM-IV-TR (APA, 2000) primarily focuses on grandiose
aspects of the disorder. Nevertheless, the former DSM-III (APA, 1980)
had already incorporated shame-proneness as part of an additional
NPD criterion, “reaction to criticism”, describing a shameful reaction
in response to narcissistic injury. Due to the fact that “the criterion
frequently failed to identify patients who were given a primary
diagnosis of NPD by their clinicians”(Millon, 1998) and the fact that
the criterion had similar (or even higher) sensitivity, specificity,
positive predictive power, and consistence (phi coefficient) for para-
noid personality disorder and BPD, the criterion had been excluded
from the DSM-IV (Gunderson et al., 1991). Our data suggest, however,
that high implicit shame might be as characteristic of NPD as explicit
shame, which was assessed in prior DSM III validation studies
(Gunderson et al., 1991;Millon, 1998).
After confirmation in future studies, the high implicit and explicit
shame may assist in differentiating NPD from psychopathy. Although
both tend to externalize blame, psychopaths or individuals with
psychopathic traits have a decreased ability to experience and
internalize shame (Cleckley, 1964;Morrison and Gilbert, 2001;
Hare, 2003).
Shame has been acknowledged as a central emotion in pathologi-
cal narcissism and NPD across different psychotherapy approaches.
Various theoretical orientations have developed psychotherapeutic
interventions designed to treat pathological shame and shame-related
cognitions, schemas, and affective or behavioral responses (e.g., self-
psychology approach, Kohut, 1971; transference focused psychother-
apy, Kernberg, 1975;schematherapy,Young et al., 2003;cognitive-
behavioral therapy, Beck et al., 2006). Nevertheless, the impact of
these interventions on NPD in general and shame in particular has not
yet been empirically evaluated and remains an important future task.
Further, the current literature provides general guidelines for mana-
ging shame in NPD patients. NPD patients are clinically described as
being sensitive to shaming words (Ogrodniczuk and Kealy, 2013)and
our finding of increased shame-proneness supports this notion. As a
consequence, therapists should be aware of the patient’s susceptibility
to shame and try to avoid shaming the patient. For example,
shame might be elicited by clarifying comments or interpretations
(Ogrodniczuk and Kealy, 2013). Furthermore, it might be helpful to
provide patients with a causal model of shame (Lecours et al., 2013).
One overarching goal might be to build affect tolerance of shame, e.g.,
by dose-by-dose desensitization (e.g., Lecours et al., 2013).
Finally, suicidality has been associated with shame in NPD and
narcissistic patients are more likely to commit suicide compared to
psychiatric patients without the disorder, even in the absence of
comorbid depression (Apter et al., 1993;Stone, 1989;Ronningstam
and Maltsberger, 1998;Pincus et al., 2009). Hence, the possibility
of a shame-related suicidal crisis should be taken into account
when treating patients with NPD.
In summary, our data indicate that explicit and implicit shame
might be a relevant feature of NPD for diagnosis and treatment.
Acknowledgments
This research was supported by a doctoral fellowship form Charité
–Universitätsmedizin Berlin (to Mrs. Ritter), and the foundation
“Sonnenfeld-Stiftung,”Berlin (to Mrs. Ritter). We are grateful to the
cooperating Departments of Psychiatry of the following German
hospitals: Theodor-Wenzel-Werk, Berlin, Asklepios Clinik North, Ham-
burg, and the Institute for Behavioral Therapy (IVB GmbH), Berlin,
Germany, for their assistance with patient recruitment.
References
Ackenheil, M., Stotz, G., Dietz-Bauer, R., Vossen, A., 1999. Mini International
Neuropsychiatric Interview. German version 5.0.0. Psychiatrische Universitätsk-
linik München, München.
American Psychiatric Association, 1980. Diagnostic and Statistical Manual of Mental
Disorders (DSM-III), 3rd edn. American Psychiatric Press, Washington DC.
American Psychiatric Association, 2000. Diagnostic and Statistical Manual of
Mental Disorders - DSM-IV-TR, 4th edn, text revision American Psychiatric
Association, Washington DC.
Andrews, B., 1995. Bodily shame as a mediator between abusive experiences and
depression. Journal of Abnormal Psychology 104, 277–285.
K. Ritter et al. / Psychiatry Research ∎(∎∎∎∎)∎∎∎–∎∎∎ 7
Please cite this article as: Ritter, K., et al., Shame in patients with narcissistic personality disorder. Psychiatry Research (2013), http://dx.
doi.org/10.1016/j.psychres.2013.11.019i

Andrews, B., Brewin, C.R., Rose, S., Krik, M., 2000. Predicting PTSD symptoms in
victims of violent crime: the role of shame, anger, and childhood abuse. Journal
of Abnormal Psychology 109, 69–73.
Andrews, B., Qian, M., Valentine, J.D., 2002. Predicting depressive symptoms with a
new measure of shame: the experience of shame scale. British Journal of
Clinical Psychology 41, 29–42.
Apter, A., Bleich, A., King, R.A., Kron, S., Fluch, A., Kotler, M., Cohen, D.J., 1993. Death
without warning? A clinical postmortem study of suicide in 43 Israeli
adolescent males. Archives of General Psychiatry 50, 138–142.
Beck, A.T., Freeman, A., Davis, D.D., 2006. Cognitive Therapy of Personality
Disorders, 2nd edn. Guilford Press, New York.
Blais, M.A., Hilsenroth, M.J., Castlebury, F.D., 1997. Psychometric characteristics of
the cluster B personality disorder under DSM-III-R and DSM-IV. Journal of
Personality Disorders 11, 270–278.
Bluemke, M., Friese, M., 2008. Reliability and validity of the Single-Target IAT (ST-
IAT): assessing automatic affect towards multiple attitude objects. European
Journal of Social Psychology 38, 977–997.
Bosson, J.K., Brown, R.P., Zeigler-Hill, V., Swann Jr., W.B., 2003. Self-enhancement
tendencies among people with high explicit self-esteem: the moderating role
of implicit self-esteem. Self and Identity 2, 169–187.
Brown, M.Z., Linehan, M.M., Comtois, K.A., Murray, A., Chapman, A.L., 20 09. Shame
as a prospective predictor of self-inflicted injury in borderline personality
disorder: a multi-modal analysis. Behaviour Research and Therapy 47, 815–822.
Browning, H.P., 2005. Social phobia and the experience of shame: childhood origins
and pastoral implications. Pastoral Psychology 53, 535–540.
Campbell, W.K., Reeder, G., Sedikides, C., Elliot, A.J., 2000. Narcissism and compara-
tive self-enhancement strategies. Journal of Research in Personality 34,
329–347.
Claesson, K., Sohlberg, S., 2002. Internalized shame and early interactions char-
acterized by indifference, abandonment, and rejection: replicated findings.
Clinical Psychology and Psychotherapy 9, 227–284.
Cleckley, H., 1964. The Mask of Sanity, 4th ed. Mosby, United States.
Crowe, M., 2004. Never good enough –Part 1: Shame or borderline personality
disorder? Journal of Psychiatric and Mental Health Nursing 11, 327–334.
DeHart, T., Pelham, B.W., Tennen, H., 2006. What lies beneath: parenting style and
implicit self-esteem. Journal of Experimental and Socical Psychology 42, 1–17 .
De Houwer, J., 2003. The extrinsic affective Simon task. Experimental Psychology
50, 77–85.
Dickinson, K.A., Pincus, A.L., 2003. Interpersonal analysis of grandiose and vulner-
able narcissism. Journal of Personality Disorders 17, 188–207.
Dimsdale, J.E., Stern, M.J., Dillon, E., 1988. The stress interview as a tool for
examining physiological reactivity. Psychosomatic Medicine 50, 4–71.
Ekman, P., 1992. An argument for basic emotions. Cognition and Emotion 6,
169 –200.
Fazio, R.H., Towles-Schwen, T., 1999. The MODE model of attitude-behavior
processes. In: Chaiken, S., Trope, Y. (Eds.), Dual Process Theories in Social
Psychology. Guilford Press, New York, pp. 97–116.
First, M.B., Spitzer, R.L., Gibbon, M., Williams, J.B.W., 1996. Structured Clinical
Interview for DSM-IV Axis I Disorders (SCID-I). Biometrics Research Depart-
ment. New York State Psychiatric Institute, New York.
First, M.B., Spitzer, R.L., Gibbon, M., Williams, J.B.W., Benjamin, L., 1997. Structured
Clinical Interview for DSM-IV Axis II Personality Disorders (SCID-II). Biometrics
Research Department. New York State Psychiatric Institute, New York.
Fydrich, T., Renneberg, B., Schmitz, B., Wittchen, H.-U., 1997. SKID-II –Strukturiertes
Klinisches Interview für DSM-IV Achse II: Persönlichkeitsstörungen [SCID II,
Structured Clinical Interview for DSM-IV Axis II: Personality Disorders].
Hogrefe, Göttingen.
Gawronski, B., Bodenhausen, G.V., 2006. Associative and propositional processes in
evaluation: an integrative review of implicit and explicit attitude change.
Psychological Bullertin 132, 692–731.
Gramzow, R., Tangney, J.P., 1992. Proneness to shame and the narcissistic person-
ality. Personality and Social Psychology Bulletin 18, 369–376.
Greenwald, A.G., Banaji, M.R., 1995. Implicit social cognition: attitudes, self-esteem,
and stereotypes. Psychological Review 102, 4–27.
Greenwald, A.G., Farnham, S.D., 2000. Using the implicit association test to measure
self-esteem and self-concept. Journal of Personality and Social Psychology 79,
1022–1038.
Greenwald, A.G., McGhee, D.E., Schwartz, J.K.L., 1998. Measuring individual differ-
ences in implicit cognition: the Implicit Association Test. Journal of Personality
and Social Psychology 74, 1464–1480.
Greenwald, A.G., Nosek, B.A., Banaji, M.R., 2003. Understanding and using the
implicit association test: I. An improved scoring algorithm. Journal of Person-
ality and Social Psychology 85, 197–216.
Gunderson, J., Ronningstam, E., Smith, L.E., 1991. Narcissisitc personality disorder: a
review of data on DSM-III-R descriptions. Journal of Personality Disorders 5,
167 –177 .
Hasson-Ohayon, I., Ehrlich-Ben, O.r.S., Vahab, K., Amiaz, R., Weiser, M., Roe, D., 2012.
Insight into mental illness and self-stigma: the mediating role of shame
proneness. Psychiatry Research 200, 802–806.
Hare, R.D., 2003. The Hare Psychopathy Checklist –Revised Manual, 2nd edition
Multi-Health Systems, Toronto.
Hawes, D.J., Helyer, R., Herlianto, E.C., Willing, J., 2013. Borderline personality
features and implicit shame-prone self-concept in middle childhood and early
adolescence. Journal of Clinical Child and Adolescent Psychology 42, 302–308.
Hofmann, W., Gawronski, B., Gschwendner, T., Le, H., Schmitt, M., 2005. A meta-
analysis of the correlation between the Implicit Association Test and explicit
self-report measures. Personality and Social Psychology Bulletin 31, 1369–1385.
Horn, W., 1983. L-P-S Leistungsprüfsystem, Göttingen. Hogrefe.
Horowitz, M., 2009. Clinical phenomenology of narcissistic pathology. Psychatric
Annals 39, 124–128.
Kämmerer, A., 2010. Zur Intensität des Erlebens von Schamgefühlen bei psy-
chischen Störungen [The intensity of shame in mental disorders]. Psychother-
apy and Psychosomatic 60, 262–270.
Karpinski, A., Steinman, R.B., 2006. The Single Category Implicit Association Test as
a measure of implicit social cognition. Journal of Personality and Social
Psychology 91, 16–32.
Keltner, D., Buswell, B.N., 1996. Evidence for the distinctness of embarrassment,
shame, and guilt: a study of recalled antecedents and facial expressions of
emotion. Cognition and Emotion 10, 155–171.
Kernberg, O., 1975. Borderline Conditions and Pathological Narcissism. Aronson,
New York.
Kernberg, O., 2009. Narcissistic personality disorders: Part I. Psychiatric Annals 39,
105–166.
Krizan, Z., Suls, J., 2008. Are implicit and explicit measures of self-esteem related? A
meta-analysis for the Name-Letter test. Personality and Individual Differences
44, 521–531.
Kohut, H., 1971. The Analysis of the Self: A Systematic Approach to the Psycho-
analytic Treatment of Narcissistic Personality Disorders. University of Chicago
Press, Chicago.
Kohut, H., 1972. Thoughts on narcissism and narcissistic rage. Psychoanalytic Study
of the Child 27, 360–400.
Kohut, H., 1977. The Restoration of the Self. International Universities Press,
New York.
Lecours, S., Briand-Malenfant, R., Descheneaux, E., 2013. Affect regulation and
mentalization in narcissistic personality disorder. In: Ogrodniczuk, J.S. (Ed.),
Understanding and Treating Pathological Narcissism. United Book Press, Balti-
more, pp. 129–146.
Lewis, H.B., 1971. Shame and Guilt in Neurosis. International Universities Press,
New York.
Linehan, M.M., 1993. Skills Training Manual for Treatment of Borderline Personality
Disorder. Guilford Press, New York.
Lobbestael, J., Arntz, A., Cima, M., Chakhssi, F., 2009. Effects of induced anger in
patients with antisocial personality disorder. Psychological Medicine 39,
557–568.
Marissen, M.A., Deen, M.L., Franken, I.H., 2012. Disturbed emotion recognition in
patients with narcissitic personality disorder. Psychiatry Research 30, 269–273.
Martens, W.H.J., 2005. Shame and narcissism: therapeutic relevance of conflicting
dimensions of excessive self esteem, pride, and pathological vulnerable self.
Annals of the American Psychotherapy Association 8, 10–17.
Miller, J.D., Widiger, T.A., Campbell, W.K., 2010. Narcissistic personality disorder and
the DSM-V. Journal of Abnormal Psychology 119, 640–649.
Millon, T., 1998. DSM narcissisitic personality disorder: historical reflections and
future directions. In: Ronningstam, E.F. (Ed.), Disorders of Narcissism: Diag-
nostic, Clinical, and Empirical Implications. American Psychiatric Press, New
York, pp. 75–101.
Morf, C.C., Rhodewalt, F., 2001. Unraveling the paradoxes of narcissism: a dynamic
self-regulatory processing model. Psychological Inquiry 12, 177–196.
Morrison, A.P., 1983. Shame, the ideal self, and narcissism. Contemporary Psycho-
analysis 19, 295–318.
Morrison, D., Gilbert, P., 2001. Social rank, shame and anger in primary and
secondary psychopaths. Journal of Forensic Psychiatry 12, 330–356.
Nosek, B.A., Banaji, M.R., 2001. The go/no-go association task. Social Cognition 19,
625–666.
Ogrodniczuk, J.S., Kealy, D., 2013. Interpersonal problems of narcissistic patients. In:
Ogrodniczuk, J.S. (Ed.), Understanding and Treating Pathological Narcissism.
United Book Press, Baltimore, pp. 113–128.
Pelham, B.W., Hetts, J.J., 1999. Implicit and explicit personal and social identity:
toward a more complete understanding of the social self. In: Tyler, R., Kramer,
R., John, O. (Eds.), The Psychology of the Social Self. Erlbaum, New York.
Pierce, C.A., Block, R.A., Aguinis, H., 2004. Cautionary note on reporting eta-squared
values from multifactor ANOVA designs. Educational and Psychological Mea-
surement 64, 916–924.
Pincus, A.L., Ansell, E.B., Pimentel, C.A., Cain, N.M., Wright, A.G.C., Levy, K.N., 2009.
Initial construction and validation of the Pathological Narcissism Inventory.
Psychological Assessment 21, 365–379.
Pincus, A.L., Lukowitsky, M.R., 2010. Pathological narcissism and narcissistic
personality disorder. Annual Review of Clinical Psychology 6, 421–446.
Raskin, R.N., Terry, H., 1988. A principal-components analysis of the Narcissistic
Personality Inventory and further evidence of its construct validity. Journal of
Personality and Social Psychology 54, 890–902.
Ritter, K., Dziobek, I., Preißler, S., Rüter, A., Vater, A., Fydrich, T., Lammers, C.-H.,
Heekeren, H.R., Roepke, S., 2011. Lack of empathy in patients with narcissistic
personality disorder. Psychiatry Research 187, 241–247.
Ronningstam, E.F., 2010. Narcissistic personality disorder: a current review. Current
Psychiatry Reports 12, 68–75.
Ronningstam, E.F., Maltsberger, J.T., 1998. Pathological narcissism and sudden
suicide-related collapse. Suicide and Life-Threatening Behavior 28, 261–271.
Rüsch, N., Corrigan, P.W., Bohus, M., Jacob, G.A., Brueck, R., Lieb, K., 2007a.
Measuring shame and guilt by self-report questionnaires: a validation study.
Psychiatry Research 150, 313–327.
K. Ritter et al. / Psychiatry Research ∎(∎∎∎∎)∎∎∎–∎∎∎8
Please cite this article as: Ritter, K., et al., Shame in patients with narcissistic personality disorder. Psychiatry Research (2013), http://dx.
doi.org/10.1016/j.psychres.2013.11.019i

Rüsch, N., Lieb, K., Göttler, I., Hermann, C., Schramm, E., Richter, H., Jacob, G.A.,
Corrigan, P.W., Bohus, M., 2007b. Shame and implicit self-concept in women
with borderline personality disorder. American Journal of Psychiatry 164,
500–508.
Rudolph, A., Schröder-Abé, M., Schütz, A., Gregg, A.P., Sedikides, C., 2008. Through a
glass, less darkly? Reassessing convergent and discriminant validity in mea-
sures of implicit self-esteem. European Journal of Psychological Assessment 24,
273–281.
Schoenleber, M., Berenbaum, H., 2012. Shame regulation in personality pathology.
Journal of Abnormal Psychology 121, 433–446.
Schulze, L., Dziobek, I., Vater, A., Heekeren, H.R., Bajbouj, M., Renneberg, B., Heuser,
I., Roepke, S., 2013. Gray matter abnormalities in patients with narcissistic
personality disorder. Journal of Psychiatric Research 47, 1363–1369.
Shaver, P., Schwartz, J., Kirson, D., O'Connor, C., 1987. Emotion knowledge: further
exploration of a prototype approach. Journal of Personality and Social Psychol-
ogy 52, 1061–1086.
Scheff, T.J., Retzinger, S.M., Ryan, M.T., 1989. Crime, violence, and self-esteem:
review and proposals. In: Mecca, A.M., Smelser, N.J., Vasconcellos, J. (Eds.), The
Social Importance of Self-Esteem. University of California Press, Berkeley,
pp. 165–199.
Sheehan, D.V., Lecrubier, Y., Sheehan, K.H., Amorim, P., Janavs, J., Weiller, E.,
Hergueta, T., Baker, R., Dunbar, G.C., 1998. The Mini-International Neuropsy-
chiatric 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 (20), 22–33.
Spielberger, C.D., Gorsuch, R.L., Lushene, R.E., 1970. Manual for the State-Trait-
Anxiety-Inventory. Consulting Psychologists Press, Palo Alto, CA.
SPSS Inc., 2011. IBM SPSS 20.
Stieger, S., Göritz, A.S., Hergovich, A., Voracek, M., 2011. Intentional faking of the
Single Categorie Implicit Association Test and the Implicit Association Test.
Psychological Report 109, 219–230.
Stone, M.H., 1989. Long-term follow-up of narcissistic/borderline patients. Psychia-
tric Clinics of North America 12, 621–641.
Tangney, J.P., 1995. Shame and guilt in interpersonal relationships. In: Tangney, J.P.,
Fischer, K.W. (Eds.), Self-Conscious Emotions: Shame, Guilt, Embarrassment,
and Pride. Guilford Press, New York, pp. 114–13 9.
Tangney, J.P., Dearing, R., 2002. Shame and Guilt. Guilford Publications, New York.
Tangney, J.P., Dearing, R.L., Wagner, P.E., Gramzow, R., 2000. The Test of Self-
Conscious Affect-3 (TOSCA-3). George Mason University.
Tangney, J.P., Wagner, P., Fletcher, C., Gramzow, R., 1992. Shamed into anger? The
relation of shame and guilt to anger and self-reported aggression. Journal of
Personality and Social Psychology 62, 669–675.
Tracy, J.L., Robins, R.W., 2004. Putting the self into self-conscious emotions: a
theoretical model. Psychology Inquiry 15, 103–125.
Turner, J.E., 1998. An Investigation of Shame Reactions, Motivation, and Achieve-
ment in a Difficult College Course (Unpublished Doctoral Dissertation). Uni-
versity of Texas, Austin.
Turner, J., Waugh, R., 2001. Feelings of shame: capturing the emotion and
investigating concomitant experiences. Paper Presented at the Annual Meeting
of the American Psychological Association, San Francisco.
Uji, M., Nagata, T., Kitamura, T., 2012. Narcissism: its function in modulating self-
conscious emotions. Bulletin of the Menninger Clinic 76, 211–234.
Vater, A., Ritter, K., Schröder-Abé, M., Schütz, A., Lammers, C.-H., Bosson, J.K.,
Roepke, S., 2013a. When grandiosity and vulnerability collide: implicit and
explicit self-esteem in patients with narcissistic personalitiy disorder. Journal of
Behavioral Therapy and Experimental Psychiatry 44, 37–47.
Vater, A., Schröder-Abé, M., Ritter, K., Renneberg, B., Schulze, L., Bosson, J.K., Roepke,
S., 2013b. The narcissistic personality inventory: a useful tool for assessing
pathological narcissism? Evidence from patients with narcissistic personality
disorder. Journal of Personality Assessment 95, 301–308.
Watson, P.J., Hickman, S.E., Morris, R.J., 1996. Self-reported narcissism and shame:
testing the defensive self-esteem and continuum hypothesis. Personality and
Individual Differences 21, 253–259.
Weisman de Mamani, A., 2010. Self-conscious emotions, general emotional distress,
and expressed emotion in family members of patients with schizophrenia.
Journal of Nervous and Mental Disease 198, 305–308.
Westen, D., Shedler, J., Bradley, R., 2006. A prototype approach to personality
disorder diagnosis. American Journal of Psychiatry 163, 846–856.
Wittchen, H.-U., Zaudig, M., Fydrich, T., 1997. Strukturiertes Klinisches Interview für
DSM-IV-TR –Achse I [Structured clinical interview for DSM-IV-TR - Axis I].
Hogrefe, Göttingen.
Young, J.E., Klosko, J.S., Weishaar, M.E., 2003. Schema Therapy. A Practitioner’s
Guide. The Guilford Press, New York, NY.
K. Ritter et al. / Psychiatry Research ∎(∎∎∎∎)∎∎∎–∎∎∎ 9
Please cite this article as: Ritter, K., et al., Shame in patients with narcissistic personality disorder. Psychiatry Research (2013), http://dx.
doi.org/10.1016/j.psychres.2013.11.019i