Personality and Social Psychology Review Copyright 2006 by

2006, Vol. 10, No. 2, 154165 Lawrence Erlbaum Associates, Inc.

Impulsivity and the Self-Defeating Behavior of Narcissists

Simine Vazire
Department of Psychology
The University of Texas at Austin
David C. Funder
Department of Psychology
University of California, Riverside

Currently prominent models of narcissism (e.g., Morf & Rhodewalt, 2001) primarily
explain narcissistsself-defeating behaviors in terms of conscious cognitive and affec-
tive processes. We propose that the disposition of impulsivity may also play an impor-
tant role. We offer 2 forms of evidence. First, we present a meta-analysis demonstrat-
ing a strong positive relationship between narcissism and impulsivity. Second, we
review and reinterpret the literature on 3 hallmarks of narcissism: self-enhancement,
aggression, and negative long-term outcomes. Our reinterpretation argues that
impulsivity provides a more parsimonious explanation for at least some of narcissists
self-defeating behavior than do existing models. These 2 sources of evidence suggest
that narcissists quest for the status and recognition they so intensely desire is
thwarted, in part, by their lack of the self-control necessary to achieve those goals.

Narcissists are a puzzle. Their bragging and arro- This article begins by proposing that the widely ac-
gance interferes with the attainment of the status and cepted cognitiveaffective processing model presented
recognition they so poignantly desire. Why do they by Morf and Rhodewalt (2001) be extended to include
continually undermine themselves in this way? The re- dispositional impulsivity. We then review the empirical
search literature appears to have achieved some con- evidence for our proposal by presenting a
sensus about the nature of sub-clinical narcissism1 meta-analysis of the relationship between narcissism
with respect to underlying cognitive, social, and affec- and impulsivity, including unpublished results from
tive processes (e.g., Morf & Rhodewalt, 2001). The two large data sets. Although researchers have been
consensual model serves as a solid foundation for inte- tangentially aware of this relationship for decades,
grating narcissism research, and provides a partial ex- none have looked to impulsivity as a central explana-
planation for narcissists perplexing behavior, but it re- tion for narcissists behavior. The final section of the
lies heavily on conscious cognitive processes and article examines the implications of our proposal by re-
omits an important category of explanatory variables: interpreting major findings in the literature in light of
dispositions. We shall argue that one possible key to the relationship between narcissism and impulsivity.
the puzzle posed by narcissists behavior is that they By incorporating dispositional impulsivity into exist-
are dispositionally impulsive: They lack the ing models of narcissism, our proposal seeks both to
self-control necessary to inhibit the behaviors that extend and to simplify the understanding of
thwart the attainment of their goals. narcissism.

Theoretical Framework
1Throughout the paper, we use the term narcissism to refer to
subclinical narcissism. Narcissism is generally seen as deriving from an at-
Portions of this research were supported by National Institute of tempt to regulate and maintain unrealistically high lev-
Mental Health Grant R01MH52391 to James Pennebaker. Data
gathering for the Riverside Accuracy Project was supported by Na-
els of self-esteem (Raskin, Novacek, & Hogan, 1991b;
tional Institute of Mental Health Grant R01MH42427 to David Robins & John, 1997). Narcissists self-views are on
Funder. the one hand lofty (Paulhus & John, 1998), making it
We are grateful to Sam Gosling, Del Paulhus, and Kathleen Vohs difficult for them to find affirmation, and on the other
for their helpful comments on this article, and to Matthias Mehl for hand vulnerable or unstable (Jordan, Spencer, Zanna,
making his data available.
Correspondence should be sent to Simine Vazire, Department of
Hoshino-Browne, & Correll, 2003), making such affir-
Psychology, The University of Texas at Austin, 1 University Station, mation particularly important. This combination of ar-
A800, Austin, TX 78712. E-mail: simine@mail.utexas.edu rogance and vulnerability is one of the paradoxes that

154
 NARCISSISM AND IMPULSIVITY

Morf and Rhodewalt (2001) addressed in their cogni- relationship between narcissism and impulsivity.
tiveaffective processing model. As they and others ar- Then, we review the literature on narcissism and argue
gue (e.g., Westen, 1990), much of narcissists cogni- that much of narcissists behavior can be reinterpreted
tive, affective, and behavioral responses are in the and understood in terms of patterns of behavior charac-
service of defending and affirming an unrealistic teristic of impulsive people. Specifically, we reinter-
self-concept. pret the literature with respect to three self-defeating
How do narcissists go about the difficult task of behaviors that are well-established hallmarks of nar-
maintaining a self-concept that is both overly positive cissism: self-enhancement, aggression, and negative
and highly fragile? Not very successfully. The research long-term outcomes.
literature shows that narcissists engage in behaviors
such as bragging, derogating others, reacting to ego
threats with hostility and aggression, making internal A Meta-Analysis of the Empirical
attributions for success and external attributions for Evidence for our Proposal
failure, and overestimating future outcomes and per-
formance even in the face of disconfirming feedback. In the early days of empirical research on narcis-
Clearly these are not behaviors well-chosen to earn the sism, impulsivity appeared frequently among long lists
respect and esteem of others. Indeed, research has of associated traits. In one of the most cited early arti-
shown that narcissists are disliked by their peers (after cles, Raskin and Terry (1988) described narcissists as
making a fleeting positive first impression; Paulhus, relatively dominant, extraverted, exhibitionistic, ag-
1998), are psychologically maladjusted (Colvin, gressive, impulsive, self-centered, subjectively
Block, & Funder, 1995), and become increasingly un- self-satisfied, self-indulgent, and nonconforming (p.
happy and disengaged from academics over the course 899). Several of these characteristics have been studied
of college (Robins & Beer, 2001). In short, as they extensively (e.g., aggression, Bushman & Baumeister,
yearn and reach for self-affirmation, [narcissists] de- 1998; self-centeredness, Robins & John, 1997), but
stroy the very relationships on which they are depend- impulsivity seems to have largely escaped subsequent
ent (Morf & Rhodewalt, 2001, p. 179). research attention.
Cognitiveaffective processing models (e.g., Morf An exception can be found in the literature on clini-
& Rhodewalt, 2001) maintain that narcissists engage in cal narcissism (Narcissistic Personality Disorder;
ineffective or even counterproductive interpersonal NPD). Although NPD and subclinical narcissism have
strategies because they are insensitive to others con- been traditionally treated as separate lines of research,
cerns. In other words, although their behavior seems the clinical literature provides a strong basis for sus-
self-defeating to the outside observer, it is actually a pecting a central role for impulsivity in subclinical nar-
deliberate, though ill-conceived, strategy that makes cissism. Hollander and Rosen (2000) argued that
sense from the point of view of their internal subjective impulsivity is an important component of many disor-
logic. We propose a more parsimonious explanation ders, including NPD, and that a complete understand-
for at least some of these self-defeating behaviors: The ing and successful treatment of narcissism in particular
behaviors are not strategic at all, narcissists simply requires an understanding of the role of impulsivity.
cant help themselves. Similarly, Casillas and Clark (2002) found that the cor-
We propose that narcissists suffer from a relation between NPD and impulsivity was .35 (p <
dispositional lack of self-control (i.e., impulsivity, a .01) when impulsivity was measured with the Schedule
concept closely akin to ego undercontrol; Block, 2002; for Nonadaptive and Adaptive Personality (SNAP;
Block & Block, 1980), and this contributes to their in- Clark, 1993), and .41 (p < .01) when measured with the
ability to meet the high self-regulatory demands of an Barratt Impulsivity Scale (BIS; Patton, Stanford, &
inflated, unstable self-concept. As a result, they are un- Barratt, 1995). Furthermore, Casillas and Clark found
able to successfully negotiate their social environments that In the case of narcissistic PD, its entire relation-
to obtain the recognition they crave. As we point out in ship with substance abuse was [statistically] explained
our reinterpretation of the narcissism literature, many by [impulsivity and self-harm] (p. 434).
of narcissists behaviors may provide temporary im- Although little or no research has focused directly
mediate gratification of their desire for recognition, but on the role of impulsivity in subclinical narcissism,
it comes at the cost of long-term successthe classic many of the early construct-validation studies of nar-
framework of the concept of delay of gratification cissism scales reported correlations with large batteries
(e.g., Funder, Block & Block, 1983; Mischel & Ayduk, of other measures, including measures of impulsivity.
2002). These correlations were never the primary concern and
In support of our proposal, we present the results often were not even mentioned in the text of the arti-
from a meta-analysis of the published correlations be- cles. The validation studies usually reported all corre-
tween narcissism and impulsivity along with unpub- lations between narcissism and the other constructs as-
lished results from two large data sets examining the sessed, regardless of direction or size. Thus, the

155
 SIMINE AND FUNDER

relationship between narcissism and impulsivity re- formants, and ratings by clinicians. In the first data set
ported in this literature is unlikely to be a biased esti- (Vazire & Mehl, 2004), narcissism scores were ob-
mate (i.e., there is less likely to be a file-drawer prob- tained from participants NPI scores (aggregated
lem of nonsignificant or disconfirming findings that across two assessments) and from the aggregate of
have remained unpublished). three informant ratings of the participants on a 4-item
Our meta-analysis began with a meticulous measure (Kurt & Paulhus, 2004). Both measures of
three-step search. Because we were interested in cast- narcissism had adequate reliabilities (NPI: testretest r
ing a broad net and obtaining empirical articles on nar- = .89; informant-rated narcissism: ICC [1, k] = .73).
cissism, we searched the PsycINFO database. We lim- Impulsivity ratings were obtained from two facets of
ited our search to journal articles and book chapters the NEO Personality InventoryRevised (NEO PIR;
published in English since 1979, when a consistent Costa & McCrae, 1992) both aggregated across two as-
definition of subclinical narcissism began to emerge. sessments: impulsivity, a facet of neuroticism, and de-
In the first step of the search, we searched for papers liberation (reversed), a facet of conscientiousness.
with narciss* (e.g., narcissism or narcissistic), These two facets were combined into a single reliable
NPI, or self-enhancement in the title. To exclude composite; ICC (2, k) = .84.
papers on clinical populations or referring to clinical In the second dataset (The Riverside Accuracy Pro-
narcissism, we excluded papers with the words ject, Phase II2; Funder, 2004), narcissism was also as-
psychoanal* (e.g., psychoanalytic or psycho- sessed using the NPI; ICC (2, k) = .83. Impulsivity
analysis) and disorder in the default search fields scores were obtained from four sources. The first was
(i.e., title, abstract, key phrase, subjects, table of con- participants self-reports on Blocks ego undercontrol
tents, and author). This yielded 347 articles and 35 scale (UC; Letzring, Block, & Funder, in press); ICC
book chapters. (2, k) = .77. The UC scale assesses individual differ-
In the second step of the search, we sought to iden- ences in ego-control (Block & Block, 1980). High
tify those papers reporting a correlation between nar- scorers (undercontrolled individuals) are unable to
cissism and some measure of impulsivity. To do this, delay gratification, have fluctuating emotions, and are
we searched the 382 papers for any of the following spontaneous, easily distracted, and relatively unbound
words or phrases, using the All Text field to search by social norms (Letzring et al., in press). Impulsivity
for impuls* (e.g., impulsivity, impulsiveness, im- scores were also derived from self-reports on the CAQ
pulse-control, etc.), self-control, ego-control, con- (Block, 1961). Participants rated themselves on items
straint, ability to delay gratification, and patience or of the CAQ using a 9-point Likert-type rating method
impatience. However, because we suspected that these (for a description, see Letrzing et al., in press), and
correlations would often be peripheral to the main sub- their profiles were then correlated with the existing
ject of the papers, we also skimmed the abstracts from CAQ prototype for ego undercontrol. This correlation
all 382 papers obtained in the first step of the search (adjusted using Fishers r-to-z formula) was then used
and retained those that included instruments that could as an index of self-rated impulsivity. As a third index of
be used to measure impulsivity (e.g., the CAQ; Block, impulsivity, one of several clinicians also rated each of
1961). From the 382 papers, we retained 81 that could the participants on the CAQ using the Q-sort method.
potentially report correlations between impulsivity and The clinician had conducted 1-hour long life history
narcissism. interview with the participant. In some cases, the rating
In the third step of the search process, we carefully of a second clinician who had viewed a video of the in-
examined all 81 of these papers and retained those with terview was also available. In those cases, the two cli-
an effect size estimate for the relationship between nar- nicians CAQ ratings were aggregated; otherwise, the
cissism and impulsivity. This yielded eight articles single CAQ rating was used. The aggregate was corre-
containing 17 effect size estimates, which are shown in lated with the CAQ ego undercontrol prototype and
Table 1 along with samples sizes and significance lev- these transformed correlations served as our third in-
els. The measures of narcissism and impulsivity came dex of impulsivity. Finally, each participant was also
from self-reports, observer ratings, informant ratings, rated on the CAQ by one or two acquaintances who
and clinician ratings. Some of the narcissism mea- served as informants. The informant CAQ ratings were
sures, such as those based on the CAQ (e.g., Wink, aggregated and the aggregate was correlated with the
1992) contained some content overlap with the CAQ ego undercontrol prototype. This transformed
impulsivity measures, but most, such as the Narcissis- correlation served as our fourth index of impulsivity.
tic Personality Inventory (NPI; Raskin & Terry, 1988)
did not.
Table 1 also includes six previously unpublished 2Data analyses of this large, new data set are in the early stage.
correlations between narcissism and impulsivity from One publication has reported correlates of Blocks measures of
two large data sets. These data sets include a combina- ego-resiliency and ego control (Letzring et al., in press). These anal-
tion of self-reports, reports from well-acquainted in- yses do not overlap with others, completed or planned.

156
Table 1. Correlations Between Narcissism and Impulsivity
Study Measure of Narcissism Measure of Impulsivity N Samplea r

Emmons (1984) NPI 16PF self-control (reversed) 65 1 .03

Raskin and Terry (1988) NPI ACL self-control (reversed) 57 2 .63**
NPI CPI self-control (reversed) 57 2 .36**
NPI Observer rating: ACL patient 57 2 .55**
(reversed)
Raskin and Novacek (1989) NPI MMPI ego-control (reversed) 57 2 .43**b
193 3 .40**
Wink and Gough (1990) CPI narcissism CPI self-control (reversed) 350 4 .67**
MMPI narcissism CPI self-control (reversed) 350 4 .60**
Wink (1991) MMPI narcissism (Factor 2: CPI self-control (reversed) 350 4 .52**
Grandiosity-Exhibitionism)b
MMPI narcissism (Factor 2: Spouse rating: ACL impulsive 152 4 .22**
Grandiosity-Exhibitionism)
Wink (1992) CAQ Narcissism prototype CPI self-control (reversed) 102 5 .30**c
(Willfulness subscale)
350 4 .34**
CAQ Narcissism prototype Spouse rating: ACL impulsive 152 4 .28*
(Willfulness subscale)
CAQ Narcissism prototype Spouse rating: ACL impatient 152 4 .26*
(Willfulness subscale)
Colvin, Block, and Funder Discrepancy between positivity Peer rating: CAQ item 53: unable 30 6 .40**d
(1995) of self and clinician ratings to delay gratification
(on CAQ)
32 .35**
Wink and Donahue (1997) MMPI Narcissism BPS constraint subscale 106 8 .35*
(reversed)e
Vazire and Mehl (2004) NPI NEO PI-R self-report (N5 & 80 9 .27*
C6-reversed)
Informant ratings NEO PI-R self-report (N5 & 78 9 .29*
C6-reversed)
RAP II (Funder, 2004) NPI Ego-undercontrol scale score 196 10 .32**
(self-report)
NPI Ego-undercontrol prototype match 198 10 .39**
(CAQ; self-report)
NPI Ego-undercontrol prototype match 195 10 .35**
(CAQ; clinicians ratings)
NPI Ego-undercontrol prototype match 190 10 .01
(CAQ; informants ratings)
Weighted mean of all 23 3549 .41**
effect sizesf
Weighted mean of effects 1,213g .34**
sizes for the 10
independent samples

Note. All measures are self-reports unless otherwise indicated. RAP II = Riverside Accuracy Project, Phase II, NPI = Narcissistic Personality
Inventory (Raskin & Hall, 1979; 1981), MMPI = Minnesota Multiphasic Personality Inventory, CPI = California Personality Inventory (Gough,
1957; 1987), ACL = Adjective Checklist (Gough & Heilbrun, 1983), CAQ = California Adult Q-set (Block, 1961), BPS = Boredom Proneness
Scale (Farmer & Sundberg, 1986), NEO PIR = NEO Personality InventoryRevised (Costa & McCrae, 1992); N5 = Impulsivity facet of NEO
PIR; C6 = Deliberation facet of NEO PIR.
*p < .05. **p < .01.
aThe 23 effect sizes reported in this table come from 10 independent samples. We have numbered the samples to indicate which effect sizes come

from the same sample.

bWink (1991) characterized the first MMPI narcissism factor as covert narcissism and the second as overt narcissism.
cResults were reported for two separate samples.
dResults reported separately for men (top correlation) and women (bottom correlation).
eThe constraint subscale of the BPS measures: feelings of impulsivity and restlessness in response to external constraints on behavior (Wink &

Donahue, 1997, p. 138).

fThe weighted means are weighted by n 3.
gThe N for the weighted mean reflects the number of unique participants represented in this table (i.e., participants represented in more than one

analysis were only counted once).

157
 SIMINE AND FUNDER

As the results of our meta-analysis demonstrate (Ta- Based on the consistent evidence in the clinical, so-
ble 1), there is strong evidence for a relationship be- cial, and personality literatures, and on our own analy-
tween impulsivity and narcissism. Narcissists are de- ses of two multi-method datasets, it is clear that a rela-
scribed as impulsive, impatient, unable to delay tionship exists between narcissism and impulsivity.
gratification, and lacking in self-control by them- Despite their penchant for self-enhancement, narcis-
selves, their spouses, their peers, and unacquainted ob- sists consistently rate themselves as impulsive, and
servers. All but 2 of the 23 correlations (91%) were sig- they are also seen by others as impulsive.
nificant and in the predicted direction. The consistency Having provided evidence that narcissists are in-
in our findings is particularly impressive considering deed impulsive, we turn to our central point: that this
the breadth of narcissism and impulsivity measures relationship helps explain their self-defeating behav-
that were used. ior. In our review, we shall compare our dispositional
To obtain an estimated effect size based on these explanation to Morf and Rhodewalts (2001) cogni-
correlations, we followed the procedures for tiveaffective processing model. Our goal is not to re-
meta-analysis recommended by Rosenthal (1995). Be- fute the existing model of narcissism, but to show that a
fore averaging the correlations, we transformed them dispositional explanation is in some respects more par-
using Fishers r-to-z formula. However, all of the effect simonious, and should also be included.
size estimates we report have been transformed back
into correlations. The mean of all 23 correlations be-
tween narcissism and impulsivity was .37, the median A Review and Reinterpretation of
was .35, and the weighted mean (weighted by n 3) Narcissists Self-Defeating Behavior
was .41. The 95% confidence interval of this estimate
ranges from .38 to .44, indicating that the effect size es- Most researchers who have examined the behav-
timate is significantly different from zero. ioral responses of narcissists have explained narcis-
However, as we note in Table 1, these are not all in- sists behavior in terms of thoughts and intentions. For
dependent effect sizes. The 23 effect sizes are drawn example, Baumeister, Bushman, and Campbell (2000)
from only 10 independent samples. Thus, we also com- wrote: Another question is what exactly narcissistic
puted the mean effect sizes of the 10 independent sam- people hope to accomplish by responding violently to
ples and conducted a meta-analysis of these effects. As an insult (p. 29). Questions like this implicitly assume
shown in the last row of Table 1, the weighted mean of that narcissists behavior is driven by their cognitive
this set of effect sizes is .34. The 95% confidence inter- appraisals, and even that it may be entirely within their
val for this estimate ranges from .28 to .40, once again conscious understanding and volitional control. In
indicating that the effect size estimate is significantly contrast, a dispositional perspective suggests that nar-
different from zero. cissists react aggressively to an insult because they are
To determine whether the effect sizes from the 10 impulsive; there is no internal subjective logic to their
independent samples are likely to be estimating the behavior, they are simply overcome by impulses that
same population mean, we computed the heterogeneity they fail to contain. In this section we begin by provid-
statistic Q. The value for Q was slightly above what ing a summary of the behavioral and biological corre-
would be expected by chance, 2(9) = 17.6, p < .05. lates of impulsivity to set the foundation for examining
This suggests that the variance in the estimates exceeds the parallels between narcissism and impulsivity.
what would be expected based on sampling error. A Then, we examine and reinterpret narcissists
follow-up analysis shows that this heterogeneity is due self-defeating patterns in three domains which have re-
mostly to a single outlier: of the 10 independent effect ceived the most attention from researchers:
sizes, all range from .25 to .50 except for one self-enhancement, aggression, and negative long-term
(Emmons, 1984; .03). When this effect size is re- outcomes.
moved from the meta-analysis, the heterogeneity drops
to nonsignificant levels, 2(8) = 7.8, n.s.
A Portrait of Impulsivity
This clear pattern of results supports Raskin and
Terrys (1988) suggestion that impulsivity is one of To understand how impulsivity can explain narcis-
the defining characteristics of narcissism. Indeed, in sists self-defeating behavior, we must first understand
Wink and Goughs (1990) study, both measures of the nature of impulsivity. What do impulsive people
narcissism correlated more strongly with CPI do? How strong is the effect of impulsivity on behav-
self-control (rs = .60, .67, ps < .01) than with any ior? In this section we provide a brief summary of the
of the other 19 CPI subscales, and in Raskin and biological, behavioral, and life outcome correlates of
Terrys study narcissism correlated more strongly impulsivity. We will draw on these patterns when rein-
with the ACL item patient (r = .55, p < .01) than terpreting narcissists self-defeating behavior.
with any of the 299 other ACL items except for sub- As mentioned earlier, the construct of impulsivity
missive (r = .57, p < .01). goes by many names, including disinhibition

158
 NARCISSISM AND IMPULSIVITY

(Gorenstein & Newman, 1980), self-control (reversed; summarize the findings in the literature and then exam-
Gough, 1956), ego-control (reversed; Block & Block, ine whether narcissists dispositional impulsivity can
1980), and deliberation (a facet of conscientiousness, provide a parsimonious explanation.
reversed; Costa & McCrae; 1992). It is also closely re-
lated, conceptually and empirically, to other constructs Narcissism and self-enhancement. It is well es-
such as the inability to delay gratification (Funder, tablished that narcissists self-perceptions are overly
Block, & Block, 1983; Wulfert, Block, Ana, Rodri- positive, almost by definition. Moreover, narcissists
guez, & Colsman, 2002), the behavior inhibition and self-deception has a very particular pattern. As Paulhus
activation systems (BIS/BAS; Carver & White, 1994), and John (1998) convincingly argued, narcissists en-
sensation-seeking (Zuckerman, 1993), psychoticism gage primarily in egoistic self-deceptive enhancement.
(Eysenck, 1997), and conscientiousness (reversed; That is, they see themselves as superior on agentic,
Bogg & Roberts, 2004). superhero-like traits such as intelligence, domi-
Self-reports of impulsivity have been validated nance, and emotional stability. In contrast they are not
against informant ratings (e.g., teachers and parents likely to see themselves as especially moral, agreeable,
ratings), cognitive tests (e.g., Stroop test), and behav- or dutiful.
ioral measures (e.g., impatience and restlessness; This pattern of self-enhancement is corroborated in
White, Moffitt, Caspi, Bartusch, & Stouth- many research findings. Narcissism is associated with
amer-Loeber, 1994). Impulsivity has also been shown self-enhancement on agentic traits (Campbell, Rudich,
to have strong biological underpinnings (Eysenck, & Sedikides, 2002), specifically intelligence and abil-
1993; Spinella, 2004; Zuckerman, 2003). For example, ity measures (Campbell et al., 2002; Farwell &
impulsivity is linked with decreased levels of serotonin Wohlwend-Lloyd, 1998; Gabriel, Critelli, & Ee, 1994;
(Harrison, Everitt, & Robbins, 1999) and specific pat- Paulhus, Harms, Bruce, & Lvsy, 2003; Paulhus & Wil-
terns of activity in the prefrontal cortex (Spinella, liams, 2002) and attractiveness (Gabriel et al., 1994).
2004), and has been shown to be substantially heritable However, narcissists do not self-enhance on communal
(Hur & Bouchard, 1997). traits (Campbell et al., 2002) and do not score higher
The behavioral and life-outcome correlates of on social-desirability measures (Raskin, Novacek, &
impulsivity are well documented. Relative to Hogan, 1991a). In addition, narcissists overestimate
nonimpulsive people, impulsive people tend to be their contribution to group tasks (John & Robins,
more delinquent (Krueger, Caspi, Moffitt, White, & 1994), have positively distorted accounts of their be-
Slouthamer-Loeber, 1996; White et al., 1994.), engage havior (Gosling, John, Craik, & Robins, 1998) and of
in more risky sexual behavior (Kahn, Kaplowitz, their romantic histories (Rhodewalt & Eddings, 2002),
Goodman, & Emans, 2002) and driving behavior and are more likely than non-narcissists to make inter-
(Bogg & Roberts, 2004), engage in more substance nal attributions for success and external attributions for
abuse (Bogg & Roberts; Wulfert et al., 2002), have failure (Campbell, Reeder, Sedikides, & Elliot, 2000;
poorer academic performance (Merrell & Tymms, Kernis & Sun, 1994; Rhodewalt & Morf, 1998).
2001), be more aggressive (Krueger et al.; Tangney, There is also evidence that narcissists
Baumeister, & Boone, 2004), be more violent (Bogg & self-delusions are deeply held. Narcissists maintain
Roberts), choose short-term over long-term rewards their overly positive views even in the face of contra-
(Funder, Block, & Block, 1983), be unable to cope dictory feedback (Robins & John, 1997), and hold on
with stress or distress (Tangney et al.), and commit to their self-enhancing attributions for success and fail-
more crimes (Wulfert et al., 2002). ure even when doing so comes at the expense of their
In short, impulsivity is a stable disposition with partners success in a cooperative task (Campbell,
clear biological underpinnings and important and Reeder, Sedikides, & Elliot, 2000). In addition, narcis-
meaningful behavioral outcomes, and is considered a sists continue to self-enhance even when they know
basic dimension of temperament (Clark, in press). This that they will be held accountable for their exaggera-
provides more reason to believe that it may play an im- tions. For example, one study (Paulhus et al., 2003)
portant explanatory role in narcissists self-defeating found that narcissists continued to over-claim (claim-
behaviors. ing to know about nonexistent items) more than
non-narcissists, even when they knew that some of the
items they were asked about were nonexistent. In short,
Review of the Narcissism Literature
narcissists continually overestimate their abilities, pos-
In this section we review the major well-established itively distort their past behaviors, and hold overly pos-
findings on narcissists self-defeating behavior and itive views of their personalities, particularly on
evaluate how well narcissists impulsivity can explain agentic traits. Moreover, they continue to do so even
each. We have identified three global areas of research when such self-presentations put them at social risk.
on narcissism: self-enhancement, aggression, and Why do narcissists exhibit such inappropriate be-
long-term costs. For each of these areas of research, we havior? Morf and Rhodewalts model (2001) empha-

159
 SIMINE AND FUNDER

sizes the role of faulty thinking patterns. For example, is their aggressiveness, particularly in response to ego
they propose that narcissists seek recognition from oth- threats. Narcissists have been found to respond with
ers for affective reasons (i.e., to allay a gnawing con- anger, hostility, and aggression to bogus negative intel-
cern of inadequacy, p. 179) and for cognitive reasons ligence feedback (Smalley & Stake, 1996; Stucke &
(i.e., to complete self-definitional needs, p. 179). Sporer, 2002), social rejection in the laboratory as well
However, several lines of research suggest that as recall of actual social rejection (Twenge & Camp-
self-enhancement is linked to dispositional bell, 2003), and being insulted in a laboratory task
impulsivity. (Bushman & Baumeister, 1998). Evidence in the clini-
First, Paulhus and his colleagues (Paulhus, Graf, & cal literature also suggests that NPD is associated with
Van Selst, 1989; Paulhus & Levitt, 1987) have shown real-world aggression (e.g., Warren et al., 2002).
that self-enhancement increases when people are dis- Furthermore, some preliminary evidence suggests
tracted by attention tasks or affective stimuli. Partici- that narcissists are more likely to endorse aggression
pants in these studies completed self-ratings while per- even in the absence of an ego threat. Specifically, male
forming another task (monitoring digits on the narcissists are more likely than non-narcissists to en-
computer screen) or being presented with an affective dorse rape-supportive statements, enjoy watching a
distractor word. Both manipulations decreased partici- film depicting consensual affectionate activity fol-
pants self-regulatory resources and led to an increase lowed by rape, and punish a female confederate for re-
in the positivity of their self-views. Paulhus and Levitt fusing to read a sexually arousing passage aloud to
suggested that people have an automatic egotism; a them (Bushman, Bonacci, van Dijk, & Baumeister,
tendency to self-enhance that can only be overridden 2003). Narcissists also score higher on self-report mea-
by effortful processes. This experimental manipulation sures of hostility (Rhodewalt & Morf, 1995). In their
sheds light on the potential role of dispositional review of the literature on self-esteem and aggression,
impulsivity in explaining narcissists persistent Baumeister, Smart, and Boden (1996) conclude that
self-enhancement. If overly positive self-views are unrealistically high, vulnerable self-esteem (i.e., nar-
common and automatic, and self-control is necessary cissism) is a contributing factor to aggression in many
to curb peoples tendencies to self-enhance, then nar- domains, including self-reported hostility, murder and
cissists dispositional lack of self-control is likely to assault, rape, domestic violence, juvenile delinquency,
account for their excessive and unconstrained political violence, and even genocide.
self-enhancement. What makes narcissists aggressive? The cogni-
A similar line of research on self-regulation has tive-processes explanation of narcissists aggression is
found that depleting peoples self-regulatory resources based on a fundamental assumption that there must be
induces them to respond more narcissistically. Spe- some internal subjective logic to the aggressive behav-
cifically, Vohs, Baumeister, and Ciarocco (2005) ex- ior of narcissists (e.g., Baumeister et al., 2000). How-
amined the effects of self-regulatory depletion on NPI ever, a strong argument can also be made for a
scores. Their findings show that people whose dispositional influence. Knowing that narcissists are
self-regulatory resources were depleted were more impulsive, that impulsivity is linked to aggression, and
likely to endorse NPI items. Furthermore, that impulsivity is a biologically-based temperament
self-regulatory depletion was associated with de- that exerts strong influences on behavior, it is reason-
creases in socially desirable responding, suggesting able to suppose that narcissists aggression is due in
that depletion is specifically linked to narcissistic part to their impulsive temperament. Furthermore,
self-presentation, not general positive Stucke and Sporer (2002) found that emotional reac-
self-presentation. Moreover, the increase in narcissism tions did not mediate the relationship between narcis-
following depletion was mediated by the decrease in sism and aggression, suggesting that affective apprais-
socially desirable responding, suggesting that deple- als do not account for narcissists aggression. We
tion led to more honest self-presentation, which led to suspect, however, that both impulsivity and cognitive
the leaking out of peoples narcissistic self-views. This and affective appraisals play an important role in nar-
study provides further evidence that narcissists cissists aggression.
self-enhance because of their impulsivity. Like the par-
ticipants in the depletion condition, narcissists do not Narcissism and negative long-term outcomes.
have the self-regulatory resources necessary to curb A third pattern characteristic of narcissists is that they
their inappropriately arrogant self-views. However, in engage in behaviors that provide short-term benefits
the case of narcissists, their self-regulatory depletion is but lead to negative long-term outcomes. This is true in
a chronic, dispositional condition rather than a result of the domains of interpersonal relationships, adjustment,
situational pressures. and achievement.
In the interpersonal domain, narcissists tend to
Narcissism and aggression. Another well-estab- make positive first impressions but become disliked
lished pattern of self-defeating behavior in narcissists once others get to know them. Paulhus (1998) exam-

160
 NARCISSISM AND IMPULSIVITY

ined the quality of narcissists and non-narcissists in- self-serving attributions at the time the predictions
terpersonal relations in newly-formed groups over a were made. However, when the self-enhancers were
7-week period. Narcissism was operationalized in two assessed again over their 4 years of college, they exhib-
ways: as the discrepancy between self-ratings and ited declining levels of self-esteem, well-being, and
friends ratings of personality, and as measured by the engagement in the academic domain relative to
NPI. Both measures of narcissism revealed the same nonself-enhancers.
pattern: narcissists were perceived as agreeable, In summary, narcissists engage in behaviors that
well-adjusted, and competent upon first meeting their provide immediate gratification of their desires for so-
group members, but their reputations deteriorated over cial status, positive affect, and ego-involvement in
time and by Week 7 they were perceived negatively. achievement domains, but they do so at the cost of ful-
These results suggest that the same behaviors that elicit filling these desires in the long-term. In one study,
immediate positive reactions from others lead narcis- Helgeson and Fritz (1999) found that unmitigated
sists to suffer negative long-term consequences in the agency (overly positive self-views on agentic traits,
interpersonal domain. i.e., narcissism) was associated with self-reports of
Paulhus (1998) also found that narcissism was asso- negative interactions with friends and family, hostility,
ciated with higher self-esteem. As he and others have anxiety, negative well-being, and negative health be-
pointed out, however, such self-reports may be affected haviors. This reflects the havoc that narcissism can
by self-enhancement, and other measures of adjust- wreak on the quality of ones interpersonal relations
ment would provide a better test of the relationship be- and psychological and physical health.
tween narcissism and psychological adjustment. Why do narcissists sacrifice the long-term fulfill-
Colvin and colleagues (1995) provided a more rigor- ment of their goals in favor of immediate gratification?
ous test of the long-term consequences of narcissism The most parsimonious explanation would seem to be
on psychological adjustment by examining the rela- that narcissists impulsivity prevents them from delay-
tionship between self-enhancement and clinician-rated ing gratification even when it is beneficial to do so (and
adjustment, each measured 5 years apart. Their find- costly not to). Indeed, the inability to delay gratifica-
ings show that self-enhancement was associated with tion is one of the defining characteristics of
impulsivity, and is linked to many negative outcomes
poorer psychological adjustment 5 years before and 5
(Cooper, Wood, Orcutt, & Albino, 2003; Krueger et al.,
years after self-enhancement was measured.
1996; White et al., 1994). Once again, expanding Morf
Other evidence suggests that narcissistic
and Rhodewalts (2001) model to include dispositional
self-enhancement provides short-term boosts in
impulsivity as a predictor of narcissists behavior ap-
well-being. Robins and Beer (2001) found that narcis-
pears to improve its explanatory power.
sists were more likely to make self-serving attributions
after a group task, and that these positive illusions were
associated with greater positive affect after the task. Summary. Table 2 presents a summary of the
This finding suggests that the same behaviors that are three main areas of research we have reviewed here,
associated with long-term psychological maladjust- and the parallels between narcissism and impulsivity in
ment are also associated with short-term boosts in those three areas. We have focused on the
well-being. self-defeating behavioral patterns of narcissists that
Finally, this pattern can also be found in the aca- have received the most attention in the literature.
demic domain. Robins and Beer (2001) also found that Surely many other studies of narcissists do not fall into
narcissists hold unrealistically high expectations of these three categories, but these represent the most
their own achievement, leading to positive short-term commonly studied and well-established behavioral
outcomes but negative long-term consequences. Spe- patterns.
cifically, those whose predictions about their college Narcissists characteristic patterns of self-enhanc-
GPA were unrealistically high were especially likely to ing, aggressive, and short-sighted behavior are all con-
experience positive affect, high ego-involvement, and sistent with our proposal that narcissists impulsivity

Table 2. Behavioral Parallels Between Narcissism and Impulsivity.

Behavior Narcissism Impulsivity

Self-enhancement Narcissists self-enhance even when Self-regulatory depletion leads to

inappropriate. self-enhancement (automatic egotism).
Aggression Narcissists react aggressively (particularly to Impulsivity is linked to aggression.
ego-threat).
Negative long-term outcomes/ Narcissism has short-term benefits but Impulsivity is linked to the inability to delay
immediate gratification long-term costs in interpersonal, gratification
adjustment, and achievement domains.

161
 SIMINE AND FUNDER

accounts in part for their behavior. Specifically, each of ences between narcissists and nonnarcissists in their
these three patterns of behavior is also typical of peo- studies.
ple high in impulsivity. Given the strong empirical re- Researchers interested in narcissism should also
lationship between narcissism and impulsivity (Table pay special attention to the role that impulsivity plays
1), the most parsimonious explanation for these behav- in the phenomena they study. For example, research in-
iors is that they are influenced, in part, by narcissists terested in the narcissismaggression link should be
impulsivity. Based on the evidence we have provided sure to include measures of impulsivity and examine
here, we suggest that Morf and Rhodewalts (2001) whether impulsivity mediates the link between narcis-
model of narcissism should be expanded to include sism and aggression.
dispositional impulsivity as a predictor of narcissists There are also several unresolved questions about
behaviors. Clearly the predictors that are already in the the nature of the relationship between impulsivity and
model, such as goals, appraisals, and self-concepts, narcissism which should be addressed in future re-
also play an important role, but the model is incom- search. First, does narcissists impulsivity result from
plete without dispositional variables, and specifically an excess of desires (e.g., unusually high need for
impulsivity. power) or from a shortage of restraints (e.g., low
The link between narcissism and impulsivity should self-regulatory resources)? Future research could ad-
also be taken into account when interpreting other (i.e., dress this by administering scales that differentiate be-
nonbehavioral) research on narcissism. For example, tween approach and inhibition systems (e.g., the BIS/
there is a great deal of research on the distinction be- BAS scales; Carver & White, 1994), and examining
tween narcissism and high self-esteem (e.g., Jordan et their relationship to narcissism.
al., 2003; Rhodewalt, Madrian, & Cheney, 1998; Another question to be addressed in future research
Smalley & Stake, 1996). This research has identified is whether the relationship between impulsivity and
several important ways in which narcissism differs narcissism is universal. If it is the case that automatic
from high self-esteem. Impulsivity, which is positively egotism is a universal tendency, we would expect this
correlated with narcissism (Table 1) but not relationship to be true across cultures. However, if
self-esteem (r = .19, n.s.; Vazire & Mehl, 2004), may there are cultural differences in the automaticity and
be another piece of the puzzle. That is, researchers may prevalence of self-enhancement motives, it is likely
be able to use impulsivity as a tool in distinguishing be- that impulsivity would not be associated with narcis-
tween narcissism and high self-esteem. sism in all cultures. That is, impulsivity should be asso-
In summary, we have presented evidence for our ciated with whatever motive is strongest in a given con-
proposal in two steps. First, our meta-analysis demon- text. Future research should examine the relationship
strated an empirical link between narcissism and between impulsivity and narcissism across cultures,
impulsivity. Second, we demonstrated how three im- and continue to examine cultural differences in
portant behavioral patterns of narcissists can be ex- self-enhancement motives (e.g., Heine, Takata, &
plained in terms of narcissists impulsivity. Taken to- Lehman, 2000; Kobayashi & Brown, 2003).
gether, these pieces of evidence suggest that Finally, the directional relationship between
impulsivity provides a parsimonious explanation for impulsivity and narcissistic behavior should be exam-
many of the behavioral (and potentially some of the ined in greater detail. The correlational data presented
nonbehavioral) correlates of narcissism. here cannot address whether impulsivity causes narcis-
sists self-defeating behavior. Although some experi-
Future directions. Our meta-analysis and re- mental research has shown that depleting
view of the literature provide a strong argument for in- self-regulatory resources leads to narcissistic behavior
corporating impulsivity into our understanding of nar- (Vohs et al., 2005), further experimentation is needed
cissism. However, new and improved techniques for to determine the causal nature of the relationship.
assessing impulsivity and self-control have emerged
because many of the studies in our meta-analysis were
conducted. Thus, our first recommendation for future Conclusion
research is that studies examining narcissism should
include multiple well-validated measures of We return now to the original question that moti-
impulsivity. For example, researchers could obtain vates much of the research on narcissism: Why do nar-
self- and informant-reports on the ego undercontrol cissists engage in behaviors that undermine their ur-
scale (Letzring et al., in press) and administer labora- gent goals of power and recognition? We have argued
tory-based tests of self-control (see Vohs et al., 2005, here that sometimes the answer may be very simple:
for examples). This will allow further tests of our hy- Because they cant help it. The paradoxical behavior of
pothesis, and will also allow researchers to examine narcissists, such as their limitless self-enhancement,
whether impulsivity accounts for some of the differ- counterproductive aggression, and preference for

162
 NARCISSISM AND IMPULSIVITY

short-term immediate gratification over long-term ben- Clark, L. A. (in press). Temperament as a unifying basis for per-
efits may be driven in part by their dispositional sonality and psychopathology. Journal of Abnormal
Psychology.
impulsivity. *Colvin, C. R., Block, J., & Funder, D. C. (1995). Overly positive
Although impulsivity has long been recognized as a self-evaluations and personality: Negative implications for
correlate of narcissism, it has been overlooked as an mental health. Journal of Personality and Social Psychology,
important influence on and explanation for narcissists 68, 11521162.
behavior. Our meta-analysis and review of the litera- Cooper, M. L., Wood, P. K., Orcutt, H. K., & Albino, A. (2003). Per-
sonality and the predisposition to engage in risky or problem
ture suggests that impulsivity should be included in behaviors during adolescence. Journal of Personality and So-
models of narcissism, and that researchers should con- cial Psychology, 84, 390410.
sider impulsivity as a possible explanatory variable Costa, P. T., Jr., & McCrae, R. R. (1992). Normal personality assess-
when interpreting narcissists self-defeating behaviors. ment in clinical practice: The NEO Personality Inventory. Psy-
chological Assessment, 4, 513.
*Emmons, R. A. (1984). Factor analysis and construct validity of the
Narcissistic Personality Inventory. Journal of Personality As-
References sessment, 48, 291300.
Eysenck, H. J. (1993). The nature of impulsivity. In W. J. McCown,
References marked with an asterisk indicate studies included in J. L. Johnson, & M. B. Shure (Eds.), Impulsive client: Theory,
the meta-analysis. research, and treatment (pp. 7191). Washington, DC: Ameri-
can Psychological Association.
Baumeister, R. F., Bushman, B. J., & Campbell, W. K. (2000). Nar- Eysenck, S. (1997). Psychoticism as a dimension of personality. In
cissism, self-esteem, and aggression: Does violence result from H. Nyborg (Ed.), Scientific study of human nature: Tribute to
low self-esteem or from threatened egotism? Current Direc- Hans J. Eysenck at eighty (pp. 109121). Amsterdam:
tions in Psychological Science, 9, 2629. Pergamon/Elsevier Science.
Baumeister, R. F., Smart, L., & Boden, J. M. (1996). Relation of Farwell, L., & Wohlwend-Lloyd, R. (1998). Narcissistic processes:
threatened egotism to violence and aggression: The dark side of Optimistic expectations, favorable self-evaluations, and
high self-esteem. Psychological Review, 103, 533. self-enhancing attributions. Journal of Personality, 66, 6583.
Block, J. (1961). The Q-Sort method in personality assessment and *Funder, D. C. (2004). [The Riverside Accuracy Project, Phase II].
psychiatric research. Palo Alto, CA: Consulting Psychologists Unpublished raw data.
Press. Funder, D. C., Block, J. H., & Block, J. (1983). Delay of gratifica-
Block, J. (2002). Personality as an affect-processing system. tion: Some longitudinal personality correlates. Journal of Per-
Mahwah, NJ: Lawrence Erlbaum Associates, Inc. sonality and Social Psychology, 44, 11981213.
Block, J. H., & Block, J. (1980). The role of ego-control and Gabriel, M. T, Critelli, J. W., & Ee, J. S. (1994). Narcissistic illusions
ego-resiliency in the organization of behavior. In W. A. Collins in self-evaluations of intelligence and attractiveness. Journal of
(Ed.), The Minnesota Symposium on Child Psychology: Vol. 13. Personality, 62, 143155.
Development of cognition, affect, and social relations (pp. Gorenstein, E. E., & Newman, J. P. (1980). Disinhibitory
39101). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. psychopathology: A new perspective and a model for research.
Bogg, T., & Roberts, B. W. (2004). Conscientiousness and Psychological Review, 87, 301315.
health-related behaviors: A meta-analysis of the leading behav- Gosling, S. D., John, O. P., Craik, K. H., & Robins, R. W. (1998). Do
ioral contributors to mortality. Psychological Bulletin, 130, people know how they behave? Self-reported act frequencies
887919. compared with on-line codings by observers. Journal of Per-
Bushman, B. J., & Baumeister, R. F. (1998). Threatened egotism, sonality and Social Psychology, 74, 13371349.
narcissism, self-esteem, and direct and displaced aggression: Gough, H. G. (1956). The California Psychological Inventory. Palo
Does self-love or self-hate lead to violence? Journal of Person- Alto, CA: Consulting Psychologists Press.
ality and Social Psychology, 75, 219229. Harrison, A. A., Everitt, B. J., & Robbins, T. W. (1999). Central sero-
Bushman, B. J., Bonacci, A. M., van Dijk, M., & Baumeister, R. F. tonin depletion impairs both the acquisition and performance of
(2003). Narcissism, sexual refusal, and aggression: Testing a symmetrically reinforced go/no-go conditional visual discrimi-
narcissistic reactance model of sexual coercion. Journal of Per- nation. Behavioural Brain Research, 100, 99112.
sonality and Social Psychology, 84, 10271040. Heine, S. J., Takata, T., & Lehman, D. R. (2000). Beyond
Campbell, W. K., Reeder, G. D., Sedikides, C., & Elliot, A. J. (2000). self-presentation: Evidence for self-criticism among Japanese.
Narcissism and comparative self-enhancement strategies. Jour- Personality and Social Psychology Bulletin, 26, 7178.
nal of Research in Personality, 34, 329347. Helgeson, V. S., & Fritz, H. L. (1999). Unmitigated agency and un-
Campbell, W. K., Rudich, E. A., & Sedikides, C. (2002). Narcissism, mitigated communion: Distinctions from agency and commu-
self-esteem, and the positivity of self-views: Two portraits of nion. Journal of Research in Personality, 33, 131158.
self-love. Personality and Social Psychology Bulletin, 28, Hollander, E., & Rosen, J. (2000). Impulsivity. Journal of
358368. Psychopharmacology, 14, 3944.
Carver, C. S., & White, T. L. (1994). Behavioral inhibition, behav- Hur, Y., & Bouchard, T. J., Jr. (1997). The genetic correlation be-
ioral activation, and affective responses to impending reward tween impulsivity and sensation seeking traits. Behavior genet-
and punishment: The BIS/BAS scales. Journal of Personality ics, 27, 455463.
and Social Psychology, 67, 319333. John, O. P., & Robins, R. W. (1994). Accuracy and bias in
Casillas, A., & Clark, L. A. (2002). Dependency, impulsivity, and self-perception: Individual differences in self-enhancement and
self-harm: Traits hypothesized to underlie the association be- the role of narcissism. Journal of Personality and Social Psy-
tween cluster B personality and substance use disorders. Jour- chology, 66, 206219.
nal of Personality Disorders, 16, 424436. Jordan, C. H., Spencer, S. J., Zanna, M. P., Hoshino-Browne, E., &
Clark, L. A. (1993). Schedule for Nonadaptive and Adaptive Person- Correll, J. (2003). Secure and defensive high self-esteem. Jour-
ality (SNAP). Minneapolis: University of Minnesota Press. nal of Personality and Social Psychology, 85, 969978.

163
 SIMINE AND FUNDER

Kahn, J. A., Kaplowitz, R. A., Goodman, E., & Emans, S. J. (2002). and low-narcissistic men. Journal of Research in Personality,
The association between impulsiveness and sexual risk behav- 36, 97116.
iors in adolescent and young adult women. Journal of Adoles- Rhodewalt, F., Madrian, J. C., & Cheney, S. (1998). Narcissism,
cent Health, 30, 229232. self-knowledge organization, and emotional reactivity: The ef-
Kernis, M. H., & Sun, C. R. (1994). Narcissism and reactions to inter- fect of daily experiences on self-esteem and affect. Personality
personal feedback. Journal of Research in Personality, 28, 413. and Social Psychology Bulletin, 24, 7587.
Kobayashi, C., & Brown, J. D. (2003). Self-esteem and Rhodewalt, F., & Morf, C. C. (1995). Self and interpersonal corre-
self-enhancement in Japan and America. Journal of lates of the Narcissistic Personality Inventory: A review and
Cross-Cultural Psychology, 34, 567580. new findings. Journal of Research in Personality, 29, 123.
Krueger, R. F., Caspi. A., Moffitt, T. E., White, J., & Rhodewalt, F., & Morf, C. C. (1998). On self-aggrandizement and
Stouthamer-Loeber, M. (1996). Delay of gratification, anger: A temporal analysis of narcissism and affective reactions
psychopathology, and personality: Is low self-control specific to success and failure. Journal of Personality and Social Psy-
to externalizing problems? Journal of Personality, 64, 107129. chology, 74, 672685.
Kurt, A., & Paulhus, D. L. (2004). Moderators of the adaptive value Robins, R. W., & Beer, J. S. (2001). Positive illusions about the self:
of self-enhancement: Operationalization, domain, and adjust- Short-term benefits and long-term costs. Journal of Personality
ment criterion. Manuscript submitted for publication. and Social Psychology, 80, 340352.
Letzring, T. D., Block, J., & Funder, D. C. (in press). Ego-control and Robins, R. W., & John, O. P. (1997). Effects of visual perspective and
ego-resiliency: Generalization of self-report scales based on narcissism on self-perception: Is seeing believing? Psychologi-
personality descriptions from acquaintances, clinicians, and the cal Science, 8, 3742.
self. Journal of Research in Personality. Rosenthal, R. (1995). Writing meta-analytic reviews. Psychological
Merrell, C., & Tymms, P. B. (2001). Inattention, hyperactivity, and Bulletin, 118, 183192.
impulsiveness: Their impact on academic achievement and Smalley, R. L., & Stake, J. E. (1996). Evaluating sources of
progress. British Journal of Educational Psychology, 71, ego-threatening feedback: Self-esteem and narcissism effects.
4356. Journal of Research in Personality, 30, 483495.
Mischel, W., & Ayduk, O. (2002). Self-regulation in the cogni- Spinella, M. (2004). Neurobehavioral correlates of impulsivity: Evi-
tive-affective personality system: Attentional control in the ser- dence of prefrontal involvement. International Journal of Neu-
vice of the self. Self and Identity, 1, 113120. roscience, 114, 95104.
Morf, C. C., & Rhodewalt, F. (2001). Unraveling the paradoxes of Stucke, T. S., & Sporer, S. L. (2002). When a grandiose self-image is
narcissism: A dynamic self-regulatory processing model. Psy- threatened: Narcissism and self-concept clarity as predictors of
chological Inquiry, 12, 177196. negative emotions and aggression following ego-threat. Jour-
Patton, J. H., Stanford, M. S., & Barratt, E. S. (1995). Factor struc- nal of Personality, 70, 509532.
ture of the Barratt Impulsiveness Scale. Journal of Clinical Psy- Tangney, J. P., Baumeister, R. F., & Boone, A. L. (2004). Higher
chology, 51, 768774. self-control predicts good adjustment, less pathology, better
Paulhus, D. L. (1998). Interpersonal and intrapsychic adaptiveness grades, and interpersonal success. Journal of Personality, 72,
of trait self-enhancement: A mixed blessing? Journal of Per- 271322.
sonality and Social Psychology, 74, 11971208. Twenge, J. M., & Campbell, W. K. (2003). Isnt it fun to get the re-
Paulhus, D. L., Graf, P., & van Selst, M. (1989). Attentional load in- spect that were going to deserve? Narcissism, social rejection,
creases the positivity of self-presentations. Social Cognition, 7, and aggression. Personality and Social Psychology Bulletin, 29,
389400. 261272.
Paulhus, D. L., Harms, P. D., Bruce, M. N., & Lvsy, D. C. (2003). *Vazire, S., & Mehl, M. R.(2004). [Self-reports, informant reports,
The over-claiming technique: Measuring self-enhancement in- and the prediction of behavior]. Unpublished raw data.
dependent of ability. Journal of Personality and Social Psychol- Vohs, K. D., Baumeister, R. F., & Ciarocco, N. J. (2005).
ogy, 84, 890904. Self-regulation and self-presentation: Regulatory resource de-
Paulhus, D. L., & John, O. P. (1998). Egoistic and moralistic bias in pletion impairs impression management and effortful
self-perception: The interplay of self-deceptive styles with ba- self-presentation depletes self-regulatory resources. Journal of
sic traits and motives. Journal of Personality, 66, 10251060. Personality and Social Psychology, 88, 632657.
Paulhus, D. L., & Levitt, K. (1987). Desirable responding triggered Warren, J. I., Burnette, M., South, S. C., Chauhan, P., Bale, R., &
by affect: Automatic egotism? Journal of Personality and So- Friend, R. (2002). Personality disorders and violence among fe-
cial Psychology, 52, 245259. male prison inmates. Journal of the American Academy of Psy-
Paulhus, D. L., & Williams, K. M. (2002). The dark triad of personal- chiatry and the Law, 30, 502509.
ity: Narcissism, machiavellianism, and psychopathy. Journal of Westen, D. (1990). The relations among narcissism, egocentrism,
Research in Personality, 36, 556563. self-concept, and self-esteem. Psychoanalysis and Contempo-
*Raskin, R., & Novacek, J. (1989). An MMPI description of the nar- rary Thought, 13, 185241.
cissistic personality. Journal of Personality Assessment, 53, White, J. L., Moffitt, T. E., Caspi. A., Bartusch, D. J., &
6680. Stouthamer-Loeber, M. (1994). Measuring impulsivity and ex-
Raskin, R., Novacek, J., & Hogan, R. (1991a). Narcissism, amining its relationship to delinquency. Journal of Abnormal
self-esteem, and defensive self-enhancement. Journal of Per- Psychology, 103, 192205.
sonality, 59, 1938. *Wink, P. (1991). Two faces of narcissism. Journal of Personality
Raskin, R., Novacek, J., & Hogan, R. (1991b). Narcissistic and Social Psychology, 61, 590597.
self-esteem management. Journal of Personality and Social *Wink, P. (1992). Three narcissism scales for the California Q-Set.
Psychology, 60, 911918. Journal of Personality Assessment, 58, 5166.
*Raskin, R., & Terry, H. (1988). A principal-components analysis of *Wink, P., & Donahue, K. (1997). The relation between two types of
the Narcissistic Personality Inventory and further evidence of narcissism and boredom. Journal of Research in Personality,
its construct validity. Journal of Personality and Social Psy- 31, 136140.
chology, 54, 890902. *Wink, P., & Gough, H. G. (1990). New narcissism scales for the
Rhodewalt, F., & Eddings, S. K. (2002). Narcissus reflects: Memory California Psychological Inventory and the MMPI. Journal of
distortion in response to ego-relevant feedback among high- Personality Assessment, 54, 446462.

164
 NARCISSISM AND IMPULSIVITY

Wulfert, E., Block, J. A., Ana, E. S., Rodriguez, M. L., & Colsman, & M. B. Shure (Eds.) Impulsive client: Theory, research, and
M. (2002). Delay of gratification: Impulsive choices and prob- treatment (pp. 7191). Washington, DC: American Psychologi-
lem behaviors in early and late adolescence. Journal of Person- cal Association.
ality, 70, 533552. Zuckerman, M. (2003). Biological bases of personality. In T. Millon
Zuckerman, M. (1993). Sensation seeking and impulsivity: A mar- & M. J. Lerner (Eds.), Handbook of psychology: Vol. 5. Person-
riage of traits made in biology? In W. J. McCown, J. L. Johnson, ality and social psychology (pp. 85116). New York: Wiley.

165