University of Southern California Dissertations and Theses

Measuring narcissistic cognitions in response to relationship threat using articulated thoughts in simulated situations

(USC Thesis Other)

Measuring narcissistic cognitions in response to relationship threat using articulated thoughts in simulated situations

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

MEASURING NARCISSISTIC COGNITIONS IN RESPONSE TO RELATIONSHIP THREAT
USING ARTICULATED THOUGHTS IN SIMULATED SITUATIONS

by
Kalina Nikolaeva Babeva
________________________________________________________________________

A Thesis Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF ARTS
(PSYCHOLOGY)

May 2011

Copyright 2011 Kalina Nikolaeva Babeva

ii
Table of Contents

List of Tables iii
Abstract v
Chapter 1: Introduction 1
The present study 6
Chapter 2: Research Design and Methods 9
Participants 9
Measures 10
Procedure 11
Chapter 3: Data Analysis and Results 14
Reliability Analyses 14
Evaluation of Normality 15
Evaluation of Effect of the Threat Manipulation 16

Chapter 4: Discussion and Conclusion 24
Caveats 29
Concluding Remarks 29

References 31
Appendices:
Appendix A: Sample Characteristics 34
Figure 1: Distributions for the Total Scores of the Six Coded 36
Variables & Narcissism
Appendix B: ATSS Scenarios 37
Appendix C: Narcissistic Personality Inventory (NPI) 40
Appendix D: ATSS Coding Procedures 43

iii
List of Tables

Table 1. Interrater Reliability (ICCs) Based on Consistency Criterion 15
Table 2. Interrater Reliability (ICCs) Based on Absolute Agreement Criterion 15
Table 3. Results of Normality Analyses 16
Table 4. Descriptive Statistics for Verbal Aggression 17
Table 5. Pairwise Friedman’s Comparisons for Verbal Aggression 17
Table 6. Descriptive Statistics for General Anger 18
Table 7. Pairwise Friedman’s Comparisons for General Anger 18
Table 8. Descriptive Statistics for Distress/Anxiety 18
Table 9. Pairwise Friedman’s Comparisons for Distress/Anxiety 19
Table 10. Descriptive Statistics for Doubts 19
Table 11. Pairwise Friedman’s Comparisons for Doubts 19
Table 12. Descriptive Statistics for Positive Self-Statements 20
Table 13. Descriptive Statistics for Desire to End Relationship 20
Table 14. Pairwise Friedman’s Comparisons for Desire to End Relationship 20

Table 15. Spearman Correlation Coefficients between NPI and Total Scores for the 21
Six Codes

Table 16. Spearman Correlation Coefficients for the Neutral Scenario 22
Table 17. Spearman Correlation Coefficients for the Low Threat Scenario 22
Table 18. Spearman Correlation Coefficients for the High Threat Scenario 23
Table A-1. Participant Gender Ratio 34
Table A-2. Ethnic Composition of Sample 34
Table A-3. Age of Participants 34
Table A-4. Participants’ Narcissistic Personality Inventory Scores 35
Table A-5. Participants’ Year in College 35

iv
Table A-6. Participants’ Current Relationship Status 35
Table A-7. Relationship Duration (in Months) for those “Currently in a Relationship” 35
Table A-8. Participants’ Lifetime Relationship Status 35

v
Abstract
Prior studies document that within the domain of romantic relationships dispositional
narcissism can be associated with relationship damaging behaviors such as game-playing
tendencies (Campbell, Foster & Finkel, 2002) but also with self-reports of relationship preserving
behaviors such as resistance to doubts about a partner’s lack of commitment (Foster & Campbell,
2005). The present study investigated the association between dispositional narcissism and
responses to escalating levels of threatening information regarding partner infidelity in a novel
manner using the Articulated Thoughts in Simulated Situations (ATSS) paradigm. The ATSS is a
think-aloud method that captures individuals’ thought content as a structured, experimenter-
controlled, hypothetical situation unfolds. One hundred sixty-two undergraduate students
verbalized their thoughts in response to three scenarios, in which their romantic partner interacted
with an individual of the opposite sex. Participant responses were content-analyzed for the
presence of verbal aggression, anger, distress, doubts, positive self-statements and desire for
relationship dissolution. Results indicated that level of narcissism was not related to
verbalizations of any of the abovementioned codes in the neutral scenario (“no threat” condition);
that is, in the absence of threat, high narcissists could not be differentiated from low narcissists
based on their articulated cognitions. In the low threat condition, narcissism positively correlated
with verbal aggression. In the high threat condition, narcissism positively correlated with verbal
aggression, positive self-statements and desire for relationship dissolution but negatively
correlated with self-reported distress. The results of the current study suggest that within non-
clinical samples, relatively narcissistic persons are more strongly reactive to threat in a manner
that is consistent with the clinical psychology literature on narcissistic rage and aggression
(Kernberg, 1975). Stated differently, in the current study the hallmark of narcissism in the
presence of threat was verbal aggression. This suggests that high narcissists are not more

vi
emotionally responsive to threat than low narcissists; instead they seem to be more behaviorally
responsive in a manner that allows them to retaliate against the source of threat. These findings
also further support the hypothesis that manifestations of narcissistic behavior vary as a function
of the situation individuals are in.

1
Chapter 1: Introduction
Personality theories posit that individuals have certain predispositions that are fairly
stable across situations. This suggests that an individual’s behavior will be at least somewhat
predictable to the extent that the personality traits he/she possesses are known. In support of these
theories, research shows that certain personality traits (usually assessed using questionnaires) tend
to be fairly stable and can be linked with specific behaviors. For example, psychopathic traits are
frequently associated with violent criminal behaviors in adults (e.g., Porter & Porter, 2006; Stone,
2007) and with aggression and delinquency in adolescents (e.g., Edens, Skeem, Cruise &
Cauffman, 2001; Myers, Burket, Harris, 1995).
Personality, however, does not tell the full story of human behavior. Although traits
might be useful in predicting an individual’s dispositions towards acting in a certain manner,
behavior is also a function of the characteristics of the environment. In fact, situational factors are
so important that a plethora of research documents their powerful influence on both prosocial and
antisocial behavioral domains ranging from administering electric shock to strangers (Milgram,
1963), police arrest decisions (Smith & Visher, 1981) and employee theft (Greenberg, 2002) to
drinking alcohol (Ludwig et al., 1977), consumer behavior (Belk, 1975), and helping strangers
(Darley & Batson, 1973), to name just a few. Thus, behavior can be fully understood only if the
context the individual is in is taken into account.
Walter Mischel (1968), having systematically examined the evidence for the consistency
of personality, also supports the notion that individuals’ behavior, far from being consistent and
stable across situations, is importantly a function of the specific circumstances. He suggests that
behavior can best be viewed as being governed by if-then rules such that certain behavior will
occur only when specific circumstances arise. An example of an “if-then” rule would be: “If
individual X is in a threatening situation, he/she acts aggressively. If individual X is in a neutral

2
situation, he/she acts gregariously.” This conceptualization of personality has the added benefit of
being able to account for the widely variable and even contradictory behaviors that an individual
may display.
Personality psychology, however, does not aim to study the influence of the situation on
the manifestation of personality. Thus, certain domains have only fairly recently been examined
from the perspective of social psychology, which emphasizes the interaction of an individual’s
characteristics with the influence of the situation to produce behavior. Narcissistic personality is
one such domain, in which cross-situational variability is especially pertinent. Dispositional
narcissism is a personality construct that embodies a set of traits that falls on a continuum within
the normal population. The hallmarks of high levels of this personality type are self-absorption,
inflated self-beliefs based on exaggerated accomplishments, perceived superiority to others, lack
of empathy, and an increased need for power (Campbell, 1999), which are thought to be evident
across multiple interpersonal situations. In the psychological literature, individuals are usually
classified as being high or low on this set of traits, with extreme manifestation of the
characteristics in question being diagnosed as a personality disorder (Narcissistic Personality
Disorder) (APA, 2000). It is important to note here that personality disorders in general and
narcissistic personality disorder in particular are defined as "an enduring pattern of [deviant] inner
experience and behavior” by the DSM-IV-TR (2000). Thus, due to its roots in clinical
psychology, which assumes personality disorders to be stable and maladaptive across situations,
narcissism is frequently viewed as a form of situationally inappropriate, inflexible behavior.
However, any adequate conceptualization of narcissism must not overlook the variability
in narcissistic behaviors, which requires incorporating the interaction between features of the
individual and the social situation. For example, studies suggest that some of the characteristics of
narcissism (namely, positive self-views) manifest only in certain domains such as success, status,

3
intelligence and physical appearance and not others such as caring and close intimate
relationships (Gabriel, Critelli, & Ee, 1994; Campbell & Foster, 2007). In these domains,
narcissists strategically maintain their positive and egocentric self-concept via self-enhancement
and aggressive reactions to perceived ego-threats (Campbell, Reeder & Sedikides, 2000; Stucke
& Sporer, 2002; Rhodewalt & Morf, 1998). In particular, individuals high in narcissism
compared to those low in narcissism have been found to both favorably compare themselves to
others (even at the cost of belittling them) and to diminish the importance of tasks on which they
believe they performed poorly (Campbell, Reeder & Sedikides, 2000). This line of research
suggests that high narcissists are likely to verbalize more self-enhancing statements but only in
situations that are perceived as threatening and that are relevant to their self-concept.
Furthermore, the highly positive self-view of narcissists correlate with heightened
sensitivity to criticism in the form of negative evaluations or social rejection because of their
threatening effect on self-perceptions of superiority (Martinez et al, 2008; Twenge & Campbell,
2003). Thus, in evaluative situations in which narcissists are criticized, they may be more likely
to act aggressively towards the source of the evaluation when given the opportunity to do so
(Bushman & Baumeister, 1998). This finding relates to the notion of threatened egotism, which
refers to defending highly positive views of the self when their veracity is challenged; in order to
preserve highly favorable self-perceptions, narcissists attack those who evaluate them negatively.
This type of behavioral response suggests the following “if-then” narcissistic rule of behavior: “If
faced with an ego-threatening situation, aggress against the source of the ego-threat as a means of
preserving positive self-views.”
However, there has been some suggestion that more narcissistic persons may not always
be more sensitive to socially threatening information. In a study on narcissism and romantic
relationships, Foster and Campbell (2005) demonstrated that, in an ambiguous situation implying

4
a lack of commitment of one’s partner, individuals high in narcissism were resistant to doubts
about the feelings of their romantic partners. In particular, high narcissists compared to low
narcissists reported greater difficulty listing reasons why their current dating partner might not be
committed to them and subsequently reported lower levels of: 1) desire to accept a date with
another potential dating partner, 2) aversion to partner dependence and deception, and 3) lack of
commitment. However, when participants were required to list reasons why current dating
partners were committed to them, the effect was reversed. In this case, narcissists were more
likely than non-narcissists to display relationship dysfunction and were more likely to accept a
date with another potential partner. More relevant here, this study seems to suggest that the same
personality trait (narcissism) can be associated with resistance to perceiving social threats to a
relationship and with increased commitment when possible threats are made salient rather than
with aggression.
The findings of Foster and Campbell (2005) with respect to resistance to doubts about
romantic partners directly contradict the clinical literature on narcissism, which implies that
narcissistic rage and increased aggression are the typical response to rejection by others
(Kernberg, 1975). Kernberg (1975) also proposes that narcissists tend to be extremely sensitive to
and to retaliate against even slight criticism or insult. This would then suggest that even a slight
indication of lack of commitment (which can be conceptualized as an ego threat) should elicit
aggressive retaliation. In other words, narcissists should be more sensitive to threat of varying
intensity than non-narcissists. It is thus possible that the task of listing reasons why their current
dating partner might not be committed to them was not engaging and realistic enough to present
participants with sufficient social threat to act aggressively.
Several limitations of the research on narcissism should be noted. First, most studies in
the field utilize questionnaire measures. Although questionnaires provide useful information that

5
is easy to analyze, they limit the variability of participant responses by allowing only very
specific, predetermined answers to items to be endorsed. Furthermore, research suggests that very
different information is elicited based on open versus closed-ended question formats. For
example, in a study on parental values, 61.5% of parents chose “thinking for themselves” as the
most important quality their children could possess for the future when this answer was provided
as part of a list. However, only 4.6% of parents articulated a similar answer in an open response
format (Schuman & Presser, 1981). It is thus evident that answers to questionnaires are strongly
influenced by question wording and the response format, so much so that the information elicited
can differ tremendously (Schwarz, 1999). Furthermore, with the exception of rare behaviors and
those of high significance, most frequent behaviors and emotional reactions are unlikely to stand
out in memory. When asking about these via questionnaires, it is likely that we are instead
tapping into attitudes and generic representations that might lack nuance (Schwarz, 1999). Thus
when trying to evaluate thoughts and behaviors related to a specific situation, it would be
advisable to use both an open response format and a more realistic task that captures ongoing
thoughts and feelings as they occur. This can be achieved by asking participants to imagine that
they are part of an ongoing situation that is presented to them via audio. The current study does
exactly this by using a simulated situation paradigm to examine the variability in narcissistic
thoughts and feelings in response to three scenarios, which vary in their level of threat. This is
done in order to achieve a better understanding of the construct and its variable manifestations
with respect to threat in the relationship realm.
In summary, the interaction of characteristics of the situation and narcissistic traits has
generally not been well examined in the relationship domain by systematically manipulating
experimental conditions within the same study. Various studies provide insight into the correlates
of narcissism, but the subtleties of how the trait predicts responses to varying degrees of

6
relationship threat have been somewhat underappreciated. Thus, it is likely that there might be
more situational variability in narcissistic responses (emotional and behavioral) than meets the
eye.
The Present Study
Given these considerations, the present study seeks to examine the variability in the
reactions of narcissists in response to uncertain versus more obvious threat of partner infidelity
using the Articulated Thoughts in Simulated Situations (ATSS) paradigm, which is a think-aloud
method that captures individuals’ thought content as a structured, experimenter-controlled
situation unfolds (Davison, Robins, & Johnson, 1983). The domain of romantic relationships was
chosen for the scenarios because it is a salient part of life that most individuals (including
undergraduate students) have experience with and it lends itself well to being manipulated to be
threatening in various degrees. Each participant was exposed to three scenarios varying in the
degree of perceived threat in which his/her romantic partner is interacting with an individual of
the opposite sex (please see Appendix B for scenario scripts). He/she was then asked to articulate
his/her ongoing thoughts in reaction to brief segments of the scenario. The purpose of this
manipulation is to assess and compare the verbalized thoughts and reactions to three levels of
infidelity threat generated by individuals who vary on their level of narcissism in the normal
population. Each participant’s verbalizations were examined for the presence of the following
coding categories: 1) verbal aggression; 2) general anger; 3) distress/anxiety; 4) doubts about the
relationship or partner’s commitment; 5) positive self-statements; 6) desire to end the relationship
(please see Appendix D for the coding procedures used). Because the high threat manipulation is
intended to be threatening to areas that are particularly important for narcissists (being admired
by others, having power and autonomy in a relationship) while the low threat manipulation is
intended to be vague and to elicit overconfidence, verbalized cognitions and emotions were

7
expected to vary systematically in the three threat conditions, as outlined in the specific
hypotheses that follow:
(Note: While the author uses the terms “narcissist” and “high narcissist” interchangeably, the
participants in this experiment were not given clinical assessments that diagnose the presence of
Narcissistic Personality Disorder. Rather, they completed the Narcissistic Personality Inventory
(NPI), which measures level of non-clinical narcissistic personality traits. Therefore, in this study
“narcissists/high narcissists” refers to subjects with high NPI scores.)
Main hypothesis: Narcissism will correlate in a different manner with verbal aggression, general
anger, distress/anxiety, positive self-statements, doubts and desire to end the relationship for the
three different threat levels as follows:
a) In the neutral scenario (significant other is talking in an unthreatening manner to a member of
the opposite sex, who is in a committed relationship and is also a good friend of the participant)
individuals will react similarly to each other regardless of their level of narcissism because an ego
threat is not present. In other words, a statistically significant correlation is not predicted to be
present between level of narcissism and the amount of articulated verbal aggression, general
anger, distress/anxiety, doubts, positive self-statements and desire to end the relationship in the
neutral scenario.
b) In the low threat scenario (significant other is talking to a classmate of the opposite sex), an
ambiguous possibility of infidelity threat is present. It is expected that narcissists although will
not perceive the situation as ego threatening and will thus be immune to anger, doubts about the
relationship and relationship dissolution (while non-narcissists will not be). As a result, level of
narcissism will correlate with verbalized emotions and cognitions in the following manner:
i. Level of narcissism will correlate positively with positive self-statements.

8
ii. Level of narcissism will correlate negatively with general anger, distress/anxiety,
doubts, and desire to end the relationship.
iii. A significant correlation between level of narcissism and verbal aggression is not
predicted.
c) In the high threat scenario (significant other is talking to a former girlfriend/boyfriend in a
provocative manner), an explicit ego threat is present that is expected to be especially provoking
for narcissists. Thus, level of narcissism will correlate with verbalized emotions and cognitions in
the following manner:
i. Level of narcissism will correlate positively with articulated verbal aggression,
general anger, doubts, positive self-statements, and desire to end the relationship.
ii. Level of narcissism will correlate negatively with distress/anxiety.
iii. A significant correlation between level of narcissism and articulated doubts is not
predicted.

9
Chapter 2: Research Design and Methods
Overview: Participants were recruited from the undergraduate psychology subject pool at USC.
1

After obtaining informed consent, participants were given a measure of narcissism to complete.
Next, the Articulated Thoughts in Simulated Situations paradigm was administered followed by a
manipulation check and relationship history questions. The obtained data were analyzed using
nonparametric correlations and other techniques to determine whether narcissism is associated
with reactions to a threat of infidelity.
Participants
Participants for the current study were recruited from the undergraduate psychology
subject pool at the University of Southern California. The only exclusion criterion (due to the
nature of the research and the scripts of the simulated situations) was non-heterosexual
orientation. In return for their participation, students received compensation in the form of extra
course credit. Participants were also entered for a lottery to receive one of three gift card prizes.
The analyses are based on 162 individuals, who completed the study. Of these
participants, 119 (73.5%) were female and 43 (26.5%) were male, indicating an unequal gender
ratio. This reflects the gender distribution of the undergraduates who completed the prescreen
procedure as part of the psychology subject pool at USC (74.6% female; 20.4% male). The
sample was ethnically and culturally diverse as it reflects the ethnic distribution of the student
population at USC: 36.4% Caucasian; 30.8% Asian/Pacific Islander; 10.5% African American;
9.9% Hispanic, and 12.4% of mixed/other heritage. All participants reported being proficient in
English when asked verbally and were able to satisfactorily complete the think aloud procedure in
English. (For sample descriptive statistics, please see Appendix A).

1
Students receive extra credit in psychology courses in exchange for their participation in
psychological studies.

10
As expected, the sample was age restricted with most participants falling in the
approximate range of 18 to 22 years of age. Eight participants (4.93%) reported being older than
22 years of age. These participants were retained in the analyses, as no a priori reasons to exclude
them exist.
Measures
Narcissistic Personality Measure. Narcissism was measured using the Narcissistic Personality
Inventory (NPI) (Raskin & Hall, 1979). The NPI is a forty-item, forced-choice questionnaire,
yielding a score in the range of zero to forty (0 – 40). In the current study, narcissism scores were
used as a continuous variable. The NPI is the most commonly used measure of narcissistic
personality traits within the normal population and the measure has been found to have favorable
psychometric properties. A study by Raskin and Terry (1988) using principal components
analysis of the NPI responses of over 1000 individuals provides evidence for a general construct
of narcissism as well as seven first-order components, identified as: Authority, Exhibitionism,
Superiority, Vanity, Exploitativeness, Entitlement, and Self-Sufficiency. A study by Soyer,
Rovenpor, Kopelman, Mullins and Watson (2001) of the construct validity of four narcissism
measures yields an internal reliability value of .83 for the NPI and also provides evidence for
good convergent and divergent validity.
Articulated Thoughts in Simulated Situation. The ATSS was used to collect data about cognitions
(Davison, Robins, & Johnson, 1983). The paradigm has been shown to have good face validity,
construct validity, discriminant validity, and inter-rater reliability (for a review, see Davison,
Vogel, & Coffman, 1997). The thoughts generated during the ATSS were coded for content that
pertains to reactions to partner infidelity and relationship doubts. The six coding dimensions are:
1) verbal aggression and hostile intent (including derogatory statements about partner/other
individual); 2) expression of general anger (including frustration and resentment); 3)

11
distress/anxiety; 4) doubts about relationship and/or partner commitment; 5) positive/self-
enhancing statements; 6) decision to end the current relationship.
The audio recordings of the participants’ articulated thoughts were transcribed and two
independent coders, who were blind to participants’ NPI scores, coded each participant’s
verbalizations. Six coders were trained to ensure adequate reliability and coding meetings were
held periodically to protect against coder drift. Each 30-second segment of participants’
verbalizations was coded for the presence and intensity of the aforementioned coding categories
using a four-point Likert scale [0 – not at all (complete absence of code); 1 – slightly/somewhat
(low presence of code); 2 – moderately (moderate presence of code); 3 – very (high presence of
code)]. Scores were calculated by averaging the two coders’ ratings on each variable. Summary
scores were also calculated by adding up the averaged ratings across all the segments. The mean
score was retained even when the scores given by each coder differed by more than 1 or 2 points
due to the overall high reliability of the coders.
Procedure
The study consisted of a single testing session. When participants arrived at the lab, they
were greeted by an experimenter and provided with an overview of the study. They were told that
they would be participating in a study of personality, dating relationships, and thoughts associated
with them. Participants were given information sheets to read (informed consent was not required
because the study was deemed exempt by the USC institutional review board) and asked if they
were still interested in participating in the study. All of the individuals that came into the lab
agreed to participate and were asked to complete a set of questionnaires (including the NPI and
demographic information questions). Participants were then given instructions (presented both in
audio and written form) that have become standard in the use of the ATSS procedure (cf. Davison
et al., 1997):

12
You are participating in a study of people’s thoughts and feelings in situations
related to romantic relationships. Often, when people are going about their daily affairs,
interacting with others and so forth, they have a kind of internal monologue going
through their heads, a constant stream of thoughts or feelings, which reflect their
reactions to something, which is happening.
What we’d like you to do is to play a part in a couple of situations that we have
taped. You will listen to audio recordings of three stories. Please imagine that this
situation is unfolding right now and that you are a part of it. Every so often the recording
will stop, you will hear a tone, and you will be asked to speak into a microphone for 30
seconds. Simply say out loud whatever is going through your mind. Say as much as you
can until you hear another tone. Of course, there are no right or wrong answers, so
please just say whatever comes to mind without judging whether it is appropriate or not.
The more you can tell us, the better.
Try to imagine as clearly as you can that it is really you in the situation right
now. Note that your task is not to speak back to any of the voices on the tape, as though
you were having a conversation with one of them. Rather, you should tune in to your own
thoughts and say them out loud. Everything that you say will be completely confidential.
Your name will not be associated with the recording in any way.

After these instructions, participants practiced the procedure in the presence of an
experimenter with a neutral scenario. Participants were given feedback on their performance (e.g.
“please try to pretend that you are actually in this situation; try not to say ‘would’”) and were
asked whether they had any questions. The experimenter then left the room and participants
proceeded with the three experimental infidelity ATSS scenarios that were matched for their
gender. The setting for all three simulated situations is a college party. The neutral scenario
depicts a situation in which one’s romantic partner is conversing with a member of the opposite
sex who is in a relationship and is also a good friend of the participant. The low threat scenario
represents an overheard conversation in which one’s romantic partner is arranging to meet with a
member of the opposite sex, who turns out to be a classmate. It is intentionally meant to be

13
somewhat ambiguous. In the high threat scenario, an extremely flirtatious conversation is
overheard in which one’s romantic partner discounts being in a committed relationship. The
reader is referred to Appendix B for the scripts of the three scenarios.
When participants finished with the ATSS procedure, they were asked to fill out a short
ATSS questionnaire asking about the realism of the situations presented, whether the scenarios
were easily imaginable, and how similar their thoughts were to what they believe they would
have in real life situations. Participants were also asked whether they had actually experienced
any real life situations similar to the simulated ones. They were then thanked, debriefed, and
escorted out of the lab.

14
Chapter 3: Data Analysis and Results
The data analysis was conducted using three broad approaches: 1) data summarization, 2)
reliability analyses, and 3) estimation and testing. First, sample characteristics were described
(age, sex, ethnicity, SES). (The reader is referred to Appendix A for descriptive tables and
graphs). Next, analyses were conducted to evaluate the reliability of the obtained data. Finally, in
order to address the three proposed sets of hypotheses, nonparametric correlational and
nonparametric analysis of variance approaches were implemented. Friedman’s nonparametric
repeated measures analysis of variance was conducted in order to evaluate the equality of the
mean ranks of the codes in the three threat conditions. The correlational modeling included an
analysis of the association between the score on a measure of personality (NPI) and performance
on the ATSS.
Reliability Analyses
In order to evaluate the interrater reliability for the six ATSS codes (namely, verbal
aggression, general anger, distress/anxiety, doubts, positive self-statements, desire to end the
relationship), intraclass correlation coefficients were calculated based on the coding scores
obtained from two independent RA coders. ICCs are typically used to assess the consistency of
continuous measurements and/or ratings made by two or more observers reporting on the same
quantity. Values reported in the tables below are based on the two-way random model (average
measures). This model was chosen due to the fact that the two ratings did not come from the same
two coders for each participant. The first table reports values using the absolute agreement
criterion, while the second, the consistency criterion.

15
Table 1. Interrater Reliability (ICCs) Based on Absolute Agreement Criterion

Code Entire Sample
(N=149)
Females
(N=108)
Males
(N=41)
Verbal aggression 0.724 0.695 0.808
General Anger 0.874 0.861 0.912
Distress/Anxiety 0.758 0.756 0.751
Doubts 0.824 0.802 0.884
Positive Self-statements 0.802 0.608 0.882
Desire to End Relationship 0.926 0.931 0.909

Table 2. Interrater Reliability (ICCs) Based on Consistency Criterion

Code Entire Sample
(N=149)
Females
(N=108)
Males
(N=41)
Verbal aggression 0.725 0.695 0.814
General Anger 0.876 0.863 0.915
Distress/Anxiety 0.757 0.756 0.751
Doubts 0.831 0.806 0.900
Positive Self-statements 0.802 0.608 0.882
Desire to End Relationship 0.926 0.932 0.909
The results of the reliability analysis indicate that all six of the coding categories have an
inter-rater reliability over 0.72 for the entire sample (N= 149), over 0.60 for females (N=108) and
over 0.75 for males (N=41). This suggests that the 6 coding categories are sufficiently reliable to
allow for them to be used as variables in further analyses. The lack of noticeable differences
between the ICC values based on absolute agreement and those based on consistency indicates
that the coding disagreements were unsystematic (i.e., there was no main effect of coder).
Evaluation of Normality
The normality assumption was evaluated for the seven variables in the study (namely,
verbal aggression, general anger, distress/anxiety, doubts, positive self-statements, desire to end
the relationship, and narcissism). This was done on the basis of the values for skewness and

16
kurtosis as well as tests of normality, histograms and normal probability plots for each of the
aforementioned variables.
The results of the evaluation of the normality assumption indicate that none of the seven
codes was normally distributed. Skewness and kurtosis were present in the distributions of all
seven codes and the Shapiro-Wilk and Kolmogorov-Smirnov tests of normality were statistically
significant, indicating that the null hypothesis of normality should be rejected. Visual
examination of the histograms also reveals departures from normality. The distributions are
neither symmetric, nor bell-shaped; they are instead significantly asymmetrical and positively
skewed. Therefore the data were analyzed with non-parametric statistical procedures.
Table 3. Results of Normality Analyses

Code Skewness Kurtosis Shapiro-Wilk Kolmogorov-Smirnov
Verbal Aggression 2.4 6.55 W=0.67, p<0.001 D=0.27, p<0.001
General Anger 0.74 0.29 W=0.95, p<0.001 D=0.10, p=0.002
Distress/Anxiety 0.6 -0.28 W=0.96, p<0.001 D=0.10, p=0.001
Doubts 0.66 0.12 W=0.96, p<0.001 D=0.08, p=0.024
Positive Self-Statements 6.29 48.08 W=0.39, p<0.001 D=0.31, p<0.001
Desire to End Relationship 1.52 2.22 W=0.77, p<0.001 D=0.27, p<0.001
Narcissism Score 1.52 2.22 W=0.08, p=0.02 D=0.98, p=0.01
Evaluation of the Effect of the Threat Manipulation
First, it was necessary to check whether the three scenarios indeed differed in the threat
level they portrayed. This was done by evaluating whether the three scenarios successfully
elicited different amounts of verbalized emotions and cognitions in the predicted manner. To
address this question, Friedman’s nonparametric repeated measures comparison was conducted.

17
In particular, the mean ranks for the ATSS code scores for each of the three scenarios were
compared in order to identify statistically significant differences.
Verbal Aggression: Table 4 summarizes the data for the verbal aggression code in the three
scenarios. The results indicate that there is a significant difference in verbal aggression across the
three scenarios (neutral, low and high threat), χ
2
(2) = 59.15, p<0.001. Inspection of the median
values and pairwise Friedman’s tests reveal, as hypothesized, an increase in verbal aggression
from low threat (Md=1.91) to high threat (Md=2.32). However, contrary to the research
hypothesis, there was not a statistically significant difference between the median verbal
aggression score in the neutral (Md=1.77) and low threat (Md=1.91) scenarios (see Table 5
below).
Table 4. Descriptive Statistics for Verbal Aggression

N Median Interquartile Range Minimum Maximum
Neutral Scenario 149 1.77 0.0-0.0 .00 2.00
Low Threat Scenario 149 1.91 0.0-0.0 .00 5.00
High Threat Scenario 149 2.32 0.0-1.0 .00 6.50

Table 5. Pairwise Friedman’s Comparisons for Verbal Aggression
Difference between:
Test Statistic Adjusted Significance
Neutral – Low Threat - 0.14 0.671
Neutral – High Threat -0.54 <0.001
Low – High Threat -0.40 0.002

General Anger: Table 6 summarizes the data for the anger code in the three scenarios. The results
indicate that there was a significant difference among the distributions of general anger in the
three scenarios (based on Friedman’s test, χ
2
(2) = 140.94, p<0.001). Inspection of the median
values and pairwise Friedman’s tests reveal, as hypothesized, that general anger was lowest in the

18
neutral scenario (Md=1.42), higher in the low threat scenario (Md=1.88) and the highest in the
high threat scenario (Md=2.70) (see Table 7 below).
Table 6. Descriptive Statistics for General Anger

N Median Interquartile Range Minimum Maximum
Neutral Scenario 149 1.42 0.0-0.0 .00 9.50
Low Threat Scenario 149 1.88 0.0-3.5 .00 11.00
High Threat Scenario 149 2.70 2.5-7.5 .00 13.00

Table 7. Pairwise Friedman’s Comparisons for General Anger
Difference between:
Test Statistic Adjusted Significance
Neutral – Low Threat - 0.46 <0.001
Neutral – High Threat -1.28 <0.001
Low – High Threat -0.82 <0.001

Distress/Anxiety: Table 8 summarizes the data for the distress/anxiety code in the three scenarios.
The results indicate that there was a significant difference among the distributions of
distress/anxiety in the three scenarios (based on Friedman’s test, χ
2
(2) = 184.35, p<0.001).
Inspection of the median values and pairwise Friedman’s tests reveal, as hypothesized, that
distress/anxiety was lowest in the neutral scenario (Md=1.25), higher in the low threat scenario
(Md=2.01) and the highest in the high threat scenario (Md=2.73) (see Table 9 below).
Table 8. Descriptive Statistics for Distress/Anxiety

N Median Interquartile Range Minimum Maximum
Neutral Scenario 149 1.25 0.0-0.5 .00 5.00
Low Threat Scenario 149 2.01 0.0-3.5 .00 9.50
High Threat Scenario 149 2.73 2.0-6.5 .00 10.50

19
Table 9. Pairwise Friedman’s Comparisons for Distress/Anxiety
Difference between:
Test Statistic Adjusted Significance
Neutral – Low Threat -0.76 <0.001
Neutral – High Threat -1.48 <0.001
Low – High Threat -0.72 <0.001

Doubts: Table 10 summarizes the data for the doubts code in the three scenarios. The results
indicate that there was a significant difference among the distributions of doubts in the three
scenarios (based on Friedman’s test, χ
2
(2) = 253.32, p<0.001). Inspection of the median values
and pairwise Friedman’s tests reveal, as hypothesized, that doubts were lowest in the neutral
scenario (Md=1.11), higher in the low threat scenario (Md=1.99) and the highest in the high
threat scenario (Md=2.90) (see Table 11 below).
Table 10. Descriptive Statistics for Doubts

N Median Interquartile Range Minimum Maximum
Neutral Scenario 149 1.11 0.0-0.5 .00 5.50
Low Threat Scenario 149 1.99 0.5-4.0 .00 11.00
High Threat Scenario 149 2.90 3.6-8.0 .50 11.50

Table 11. Pairwise Friedman’s Comparisons for Doubts
Difference between:
Test Statistic Adjusted Significance
Neutral – Low Threat -0.88 <0.001
Neutral – High Threat -1.80 <0.001
Low – High Threat -0.91 <0.001

Positive Self-Statements: Table 12 summarizes the data for the positive self-statements code in
the three scenarios. The results indicate that there was no significant difference among the
distributions of positive self-statements in the three scenarios (based on Friedman’s test, χ
2
(2) =
2.84, p=0.242). This suggests that the amount of articulated positive self-statements does not
differ significantly based on threat level.

20
Table 12. Descriptive Statistics for Positive Self-Statements

N Median Interquartile Range Minimum Maximum
Neutral Scenario 149 1.99 0.0-0.0 .00 5.50
Low Threat Scenario 149 1.96 0.0-0.0 .00 5.00
High Threat Scenario 149 2.05 0.0-0.0 .00 5.00

Decision to End Relationship: Table 13 summarizes the data for the desire to end the relationship
code in the three scenarios. The results indicate that there was a significant difference among the
distributions of desire to end the relationship in the three scenarios (based on Friedman’s test, χ
2

(2) = 115.32, p<0.001). Inspection of the median values and pairwise Friedman’s tests reveal that,
as hypothesized, more desire to end the relationship was verbalized in the high threat scenario
(Md= 2.45) than either the neutral (Md=1.74) or low threat scenario (Md=1.81). However,
contrary to the research hypothesis, there was not a difference in the median desire to end the
relationship score in the neutral and low threat scenarios (see Table 14 below).
Table 13. Descriptive Statistics for Desire to End Relationship

N Median Interquartile Range Minimum Maximum
Neutral Scenario 149 1.74 0.0-0.0 .00 .50
Low Threat Scenario 149 1.81 0.0-0.0 .00 2.00
High Threat Scenario 149 2.45 0.0-1.0 .00 3.50

Table 14. Pairwise Friedman’s Comparisons for Desire to End Relationship
Difference between:
Test Statistic Adjusted Significance
Neutral – Low Threat - 0.07 1.00
Neutral – High Threat -0.71 <0.001
Low – High Threat -0.63 <0.001
Main Effect of Narcissism: In order to determine whether a main effect of narcissism on
verbalized emotions and cognitions was present, Spearman correlation coefficients were

21
calculated between Narcissistic Personality Inventory (NPI) scores and total scores for the six
coding categories. Please see Table 15 below for the values of the correlation coefficients.
Table 15. Spearman Correlation Coefficients between NPI and Total Scores for the Six Codes

Aggression Anger Distress Doubts
Positive Self-
Statements
Desire to End
Relationship
NPI Score Spearman’s rho .222
**
.156
*
-.080 .086 .157
*
.194
**

Sig. (1-tailed) .003 .029 .168 .151 .029 .009
N 147 147 147 147 147 147
**. Correlation is significant at the 0.01 level (1-tailed).
*. Correlation is significant at the 0.05 level (1-tailed).
The results indicate that narcissism is significantly positively correlated with verbal
aggression, general anger, positive self-statements, and desire to end the relationship (rho=0.19,
p=0.009). Narcissism is not significantly correlated with distress/anxiety or with doubts. This
suggests that verbal aggression, general anger, positive self-statements and desire to end the
relationship tend to increase as narcissism increases, while distress/anxiety and doubts are not
related to level of narcissism in the current sample.
Main Hypothesis: Narcissism will display a different pattern of correlations with verbal
aggression, general anger, distress/anxiety, positive self-statements, doubts and desire to end the
relationship for the three different threat levels.
a) In order to test the hypothesis that in the neutral scenario individuals would react similarly to
each other regardless of their level of narcissism, a correlational analysis was conducted.
Spearman correlation coefficients were calculated and tested (please see Table 16 below).

22
Table 16. Spearman Correlation Coefficients for the Neutral Scenario

Aggression Anger Distress Doubts
Positive Self-
Statements
Desire to End
Relationship
NPI Spearman's rho .127 .094 .059 .121 .144 .000
Sig. (2-tailed) .126 .257 .478 .144 .083 1.000
N 147 147 147 147 147 147
*. Correlation is significant at the 0.05 level (2-tailed).

The results indicate that narcissism is not significantly correlated with any of the
variables of interest, namely verbal aggression (rho=0.13 p=0.126), general anger (rho=0.09,
p=0.257), distress/anxiety (rho=0.06, p=0.478), doubts (rho=0.12, p=0.144), positive self-
statements (rho=0.14, p=0.083), and desire to end the relationship (rho=0.00, p=1.00). This
suggests that verbal aggression, general anger, distress/anxiety, doubts positive self-statements
and desire to end the relationship are not linearly related to level of narcissism in the current
sample.
b) In order to test the hypothesis that in the low threat scenario, narcissism would correlate
positively with positive self-statements and negatively with general anger, distress/anxiety, and
desire to end the relationship, a Spearman correlational analysis was conducted (please see Table
17 below).
Table 17. Spearman Correlation Coefficients for the Low Threat Scenario

Aggression Anger Distress Doubts
Positive Self-
Statements
Desire to End
Relationship
NPI Spearman's rho .199
*
.091 .040 .107 .105 .152
Sig. (1-tailed) .016 .272 .630 .196 .206 .065
N 147 147 147 147 147 147
*. Correlation is significant at the 0.05 level (2-tailed).
The results indicate that narcissism is significantly positively correlated only with verbal
aggression. This suggests that verbal aggression increases as level of narcissism increases while

23
general anger, distress/anxiety, doubts, positive self-statements and desire to end the relationship
are not related systematically to narcissism in the low threat condition.
c) In order to test the hypothesis that in the high threat scenario, narcissism would correlate
positively with articulated verbal aggression, general anger, positive self-statements, and desire to
end the relationship and negatively with distress/anxiety, a correlational analysis was conducted.
Spearman correlation coefficients were calculated and tested (please see Table 18 below).
Table 18. Spearman Correlation Coefficients for the High Threat Scenario

Aggression Anger Distress Doubts
Positive Self-
Statements
Desire to End
Relationship
NPI Spearman's rho .158
*
.130 -.150
*
.017 .175
*
.187
*

Sig. (1-tailed) .028 .058 .035 .420 .017 .012
N 147 147 147 147 147 147
*. Correlation is significant at the 0.05 level (1-tailed).
**. Correlation is significant at the 0.01 level (1-tailed).

The results indicate that, under a condition of high threat, narcissism is significantly
positively correlated with verbal aggression, positive self-statements and desire to end the
relationship and significantly negatively correlated with distress/anxiety. Narcissism is not
significantly correlated with general anger or doubts. This suggests that, when threat is high,
verbal aggression, positive self-statements and desire to end the relationship increase as level of
narcissism increases while distress/anxiety decreases as narcissism increases. General anger and
doubts are not related systematically to narcissism in the current sample.
In order to compare the magnitude of the correlations between NPI scores and the six
codes in the three threat conditions, correlation differences were computed and tested. The results
reveal that the only significant differences in correlation magnitude for the six codes were
between the distress/anxiety code in the neutral and high threat scenario (t=2.05, p=0.02) and
between the distress/anxiety code in the low threat and high threat scenario (t=2.16, p=0.016).

24
Chapter 4: Discussion and Conclusion
The present study portrays narcissists as strongly reactive to threat in a manner that is
consistent with both the clinical literature and studies that examine narcissistic reactions to ego
threat. The hallmark of narcissism in both the low and high threat scenarios was aggression. It is
interesting to note, however, that narcissism was not correlated with anger or doubts. This
suggests that the aggressive reactions of narcissists may not have been fueled by higher levels of
anger or doubts. As a result, it seems that high narcissists are not more emotionally responsive
that low narcissists; they are more behaviorally responsive in a manner that allows them to
retaliate against the source of threat (in this case, their romantic partner) and to regain their
positive self-views. Thus, narcissists didn’t get angrier or more upset than non-narcissists; instead
what differentiated the two groups is that narcissists tried to get even.
The findings also support the “if-then” rule of narcissistic behavior suggested earlier: “If
faced with an ego-threatening situation in the romantic realm, aggress against the source of the
ego-threat and end the relationship as a means of preserving positive self-views.” In other words,
narcissism seems to be best conceptualized as highly responsive to threat in dating relationships
and in other domains. In situations in which narcissists are not threatened, however, they do not
display the same types of cognitions and behaviors because they do not need to protect their
positive self-views. In fact, the literature on narcissism supports the view that narcissists can be
rather charming and even well liked in non-threatening situations (Morf & Rhodewalt, 2001;
Young & Pinsky, 2006).
Thus, the results of the current study partially support the hypothesis that manifestations
of narcissistic behavior vary as a function of the threat level of the situation individuals are in. In
particular, high narcissists were indistinguishable from low narcissists in the neutral situation, i.e.
individuals’ reactions did not vary as a function of their degree of narcissism when threat was not

25
present. However, in the low and high threat situations, systematic differences between
individuals were present based on degree of narcissism. In the low threat situation, as narcissism
increased, so did verbal aggression. In the high threat situation, as narcissism increased so did
verbal aggression, desire to end the relationship and self-enhancing statements. In addition,
distress decreased as narcissism increased. As a result, it appears that degree of narcissism can be
used to predict individuals’ verbal aggression in situations that imply infidelity threat (regardless
of the degree of threat). In addition, these results also suggest that narcissistic behavior is not
necessarily stable across situations. In the three different scenarios in this study the pattern of
narcissism correlates were not the same. This means that individuals’ narcissistic traits were not
the only determinant of their reactions; the situation’s characteristics also had a significant impact
on articulated thoughts and feelings.
It must be noted that the only statistically significant interaction between threat level and
narcissism was found for distress. However, given the relatively small sample size and the
nonparametric nature of the variables, it is possible that the current study did not have the
statistical power to detect other interactions.
The findings in this study also challenge the idea that certain emotional responses are
more characteristic of individuals high on narcissism than those low on narcissism. In particular,
higher levels of anger have been associated with narcissism in past research (e.g. Rhodewalt &
Morf, 1998). This finding, however, was not present in the current study. Anger was not
significantly related to degree of narcissism in any of the three scenarios (neutral, low and high
threat) even though the high threat situation was created to be very provoking and to represent an
ego threat that would be especially salient for narcissists. Furthermore, doubts were also not
systematically related to narcissism in any of the situations. These findings are somewhat
surprising as evidence suggests that narcissism may be beneficial in protecting against doubts

26
when the commitment of one’s partner is questioned (Foster and Campbell, 2005). Based on
Foster and Campbell’s (2005) study, it was expected that narcissism would correlate negatively
with doubts about the relationship and would thus buffer against anger. In other words, since
narcissists have been found to self-enhance in a manner that allows them to preserve their often
exaggerated positive self-views (such as “I am such a wonderful individual that my partner would
never even consider being with someone else”) (Campbell et al., 2000; Morf & Rhodewalt, 2001;
Gabriel, Critelli, & Ee, 1994), it was expected that narcissisms would immunize against doubts
about one’s imaginary partner’s actions. Instead it appears that in the current sample, all three
situations triggered a similar amount of suspicion in all the participants regardless of their
narcissism level.
A possible explanation for this is that a ceiling effect was present. Since the high threat
scenario was created to be provoking and upsetting for everyone, it is possible that all the
participants reported exclusively high levels of anger and doubts. Descriptive statistics and visual
examination of the variable distributions indicate that a ceiling effect was not present; instead,
participants reported a wide range of doubts and anger across the continuum of narcissism.
Exploratory analyses were then conducted to determine if relationship variables assessed in the
study (namely relationship history, current relationship status, current relationship duration) were
correlated with anger and doubts in the high threat scenario. None of these variables correlated
with anger or with doubts. These findings suggest that factors unrelated to narcissism and having
ever been or currently being in a relationship play a role in determining the verbalization of anger
and doubts in the three situations. To reiterate, this means that personality was not as influential
in determining participants’ responses in terms of anger and doubts as suggested by the literature
on narcissism.

27
As mentioned previously, the findings in this study, however, also support the idea that
certain emotional responses are more characteristic of individuals high on narcissism than those
low on narcissism. In the current sample, narcissism was positively correlated with verbal
aggression in both the low and high threat scenarios, as suggested by the literature linking
narcissism with expressions of aggression in situations that represent an ego threat such as social
rejection (Twenge & Campbell, 2003), failure (Rhodewalt & Morf, 1998) or negative feedback
on performance (Martinez, Zeichner, Reidy, & Miller, 2008). Thus, it appears that certain
correlates of narcissism tend to be consistently present in any situation that elicits threat
regardless of the specific level of the threat. In other words, certain emotional and cognitive
correlates of personality traits are likely to be more stable across situations.
Furthermore, desire to end the relationship, distress and positive self-statements were
useful in differentiating high from low narcissists in the high threat situation. Thus, there appears
to have been a threshold effect such that select aspects of narcissism were “activated” only after a
certain level of threat was reached. These finding are consistent with studies suggesting that
narcissists are less invested in their romantic relationships (Campbell, Foster & Finkel, 2002) and
thus are more likely to end them; that narcissism tends to be negatively related to distress, sadness
and depression (Sedikides, et al., 2004); and that narcissists use self-enhancement strategies to
maintain their very positive and egocentric self-concept in the face of threat (Campbell, Reeder &
Sedikides, 2000; Stucke & Sporer, 2002). Thus, the verbalization of less distress, of more self-
enhancing statements and of greater desire to end the relationship in the high threat scenario fits
well with previous research on narcissistic reactions in situations that are perceived as
threatening.
However, degree of narcissism was not associated with distress or positive self-
statements in the low threat scenario, suggesting that the low threat situations was equally

28
distressing for all individuals on the narcissism continuum. In order to confirm this, both the
median values and the distributions of overall distress for high versus low narcissists were tested
for equality. The results indicate that no statistically significant difference were present thus
supporting the claim that the scenarios caused the same amount of verbalized distress in all
individuals regardless of their level of narcissism. This further supports the threshold hypothesis
and also suggests that other factors such as relationship history and current relationship status
might have been more influential in determining the verbalization of distress in the infidelity
scenarios. Indeed, additional analyses revealed that distress was positively correlated with
participants’ self-reported current relationship duration (rho=0.253, p=0.038). This finding
provides further support for the proposition that personality traits are likely to be less influential
than other factors (in this case, relationship duration) in determining the reactions of individuals
in the absence of certain situational characteristics.
It is also noteworthy that when the relation between narcissism and the overall degree of
verbalized emotions across the three scenarios was examined, a slightly different pattern of
correlations emerged. In this case, narcissism was associated positively with verbal aggression,
general anger, positive self-statements, and desire to end the relationship (as would be predicted
by previous research) and not significantly associated with doubts or distress. If the three
situations had not been considered separately, a different view of narcissism would have
emerged, as it would have been easy to conclude that narcissism is always associated with
aggression, anger, self-enhancement, and relationship dissolution. However, when considering
the situations separately, it becomes apparent that anger was not associated with narcissism and
distress and self-enhancing statements were associated with narcissism only in the high threat
scenario. Thus, there appears to be much more variability in emotional responses, such that
certain narcissistic behaviors and cognitions will only manifest when a certain level of threat has

29
been reached. This points to the importance of examining individuals in different situations in
order to accurately capture their emotional and cognitive responses.
Caveats
Despite the use of a paradigm that is more informative than questionnaire measures, all of
the data in the current study are still based on self-report, which may be biased especially when
sensitive issues like infidelity are involved. It is possible that the responses of at least some of the
participants were affected by social desirability or the novelty of the situation, thus attenuating
the strength of the findings. Furthermore, self-report data are not meant to be interpreted as a
perfect predictor of actual behavior. Thus, this study does not attempt to make the extrapolation
that reports of cognitions and of intended behaviors will reliably predict actual behaviors if a
similar, real-life situation were to occur. To achieve this latter aim, future studies should
incorporate a measure of actual behaviors or peer report in order to determine whether self-
reported negative emotions are highly correlated with negative behaviors outside of the lab.
It should also be noted that the study is based on a sample of college students. Due to the
age of the participants and the nature of the simulated situations used, the generalizability of the
findings may be limited to younger adults who have only experienced dating relationships. Thus,
it would be advisable for studies to further examine narcissistic reactions to threat in a more age
diverse sample.
Concluding Remarks
Despite these caveats, the current study is informative as a means of assessing the link
between narcissism and emotional and cognitive responses in threatening situations. It represents
a novel approach to the study of narcissism through its use of the ATSS paradigm. Instead of
assuming that emotional and cognitive responses can be accurately captured through global self-
ratings on questionnaires, the current study examines the ongoing thoughts and feelings of

30
individuals as a simulated situation unravels. As a result, it is not extremely surprising that the
findings obtained differ somewhat from previous questionnaire studies.
In particular, the current study suggests that increases in narcissism in the general
population are likely to not be systematically associated with emotional and cognitive reactions to
a neutral, non-threatening situation. In other words, in the absence of provocation, narcissists are
indistinguishable from non-narcissists based on their articulated thoughts and feelings. However,
as the situations became more threatening, narcissism was related to an increased degree of
verbalizations of aggression. Thus, narcissists appear to be highly sensitive to infidelity threat,
which they seek to counteract by aggressing. In other words, they can be viewed as tending to
have a revenge-seeking orientation, which differentiates them from those low on narcissism in the
presence of threat.
It is thus the interaction between narcissism and situational variables (namely threat) that
makes the personality trait a meaningful predictor of emotions and cognitions. As a result, the
current study highlights the importance of always taking into account situational factors in order
to gain a better understanding of the influence of personality on our lives. Although personality
traits might feel like a defining feature of who we are and how we perceive the world around us,
they become apparent to others only when the social situation allows them to.

31
References
American Psychiatric Association. (2000). Diagnostic and statistical manual of mental
disorders: DSM-IV-TR. Washington, DC: Author.

Belk, R. (1975). Situational variables and consumer behavior. The Journal of Consumer
Research, 2(3),157-164

Bushman, B.J. & Baumeister, R.F. (1990). Threatened egotism, narcissism, self-esteem, and
direct and displaced aggression: Does self-love or self-hate lead to violence? Journal of
Personality and Social Psychology, 75(1), 219-229.

Campbell, W. K. (1999). Narcissism and romantic attraction. Journal of Personality and Social
Psychology, 77(6), 1254-1270.

Campbell, W. K., Foster, C. A., & Finkel, E. J. (2002). Does self-love lead to love for others?: A
story of narcissistic game playing. Journal of Personality and Social Psychology, 83(2),
340-354.

Carroll, L., Hoenigmann-Stovall, G. I., & Whitehead, G. I., III. (1996). The interpersonal
impact of narcissism: A comparative study of entitlement and self-absorption factors.
Journal of Social Behavior & Personality, 11(3), 601-613.

Campbell, K.W. & Foster J.D. (2007). The Narcissistic Self: Background, an extended agency
model, and ongoing controversies. In C. Sedikides & S. Spencer (Eds.), Frontiers in
social psychology: The self. Philadelphia, PA: Psychology Press.

Campbell, W. K., Reeder, G. D., Sedikides, C., & Elliot, A. J. (2000). Narcissism and self-
enhancement strategies. Journal of Research in Personality, 34, 329-347.

Darley, J. & Batson, C.D. (1973). From Jerusalem to Jericho: A study of situational and
dispositional variables in helping behavior. Journal of Personality and Social
Psychology, 27, 100-108.

Davison, G.C., Robins, C., & Johnson, M.K. (1983). Articulated thoughts during simulated
situations: A paradigm for studying cognition in emotion and behavior. Cognitive
Therapy Research, 7, 17-40.

Davison, G. C., Vogel, R. S., & Coffman, S. G. (1997). Think-aloud approaches to cognitive
assessment and the articulated thoughts in simulated situations paradigm. Journal of
Consulting and Clinical Psychology, 65(6), 950-958.

del Rosario, P. M., & White, R. M. (2005). The narcissistic personality inventory: Test-retest
stability and internal consistency. Personality and Individual Differences, 39(6), 1075-
1081.

32
Edens, J., Skeem, J., Cruise, K., & Cauffman, E.(2001). The assessment of juvenile psychopathy
and its association with violence: A critical review. Behavioral Sciences & the Law, 18,
53−80.

Farwell, L., & Wohlwend-Lloyd, R. (1998). Narcissistic processes: Optimistic expectations,
favorable self-statements, and self-enhancing attributions. Journal of Personality, 66,
65–83.

Foster, J. D., & Campbell, W. K. (2005). Narcissism and resistance to doubts about romantic
partners. Journal of Research in Personality, 39(5), 550-557.

Gabriel, M.T., Critelli, J.W., & Ee, J.S. (1994). Narcissistic illusions in self-statements of
intelligence and attractiveness. Journal of Personality, 62, 143-155.

Greenberg, J. (2002). Who stole the money, and when? Individual and situational determinants
of employee theft. Organizational Behavior and Human Decision Processes, 89(1), 985-
1003.

Kernberg, O. (1975). Borderline conditions and pathological narcissism. New York: Jason
Aronson.

Ludwig, A., Cain, R., Wikler, A., Taylor, R., & Bendfeld, T. (1977). Physiological and situational
determinants of drinking behavior. In: M.M. Gross, (Ed.), Alcohol Intoxication and
Withdrawal. Plenum Press, New York, 589–600.

Martinez, M., Zeichner, A., Reidy, D., & Miller, J. (2008). Narcissism and displaced aggression:
Effects of positive, negative, and delayed feedback. Personality and Individual
Differences, 44(1), 140-149.

Milgram, S. (1963). Behavioral study of obedience. Journal of Abnormal and Social Psychology,
67, 371-378.

Mischel, W. (1968). Personality and assessment. New York: Wiley.

Morf, C. C., & Rhodewalt, F. (2001). Unraveling the paradoxes of narcissism: A dynamic self-
regulatory processing model. Psychological Inquiry, 12, 177–196.

Myers, W., Burket, R. & Harris, H. (1995). Adolescent psychopathy in relation to delinquent
behaviors, conduct disorder, and personality disorders. Journal of Forensic Science,
40(3), 435-439.

Porter S., & Porter S. (2006). Psychopathy and violent crime. In: Hervé H, Yuille J. (Eds.) The
psychopath: Theory, research, and practice. Mahwah, New Jersey: Erlbaum.

Raskin, R., & Hall, C. S. (1979). The narcissistic personality inventory. Psychological Reports,
45, 159-161.

33
Raskin, R., & 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(5), 890-902.

Rhodewalt, F., & Morf, C.C. (1998). On self-aggrandizement and anger: A temporal analysis of
narcissism and affective reactions to success and failure. Journal of Personality and
Social Psychology, 45(3), 672-685.

Schwarz, N. (1999) Self-report: How the questions shape the answers, American Psychologist,
54, 93–105.

Schuman, H. & Presser, S. (1981). Questions and Answers in Attitude Surveys: Experiments in
Question Form, Wording, and Context. New York: Academic Press.

Smith, D., & Visher, C. (1981). Street-Level justice: Situational determinants of police arrest
decisions. Social Problems, 29(2), 167-177

Soyer, R. B., Rovenpor, J. L., Kopelman, R. E., Mullins, L. S., & Watson, P. J. (2001). Further
assessment of the construct validity of four measures of narcissism: Replication and
extension. Journal of Psychology: Interdisciplinary and Applied, 135(3), 245-258.

Stone, M. H. (2007). Violent crimes and their relationship to personality disorders. Personality
and Mental Health, 1(2), 138-153.

Twenge, J., & Campbell, K. (2003). “Isn’t it fun to get the respect that we’re going to deserve?"
Narcissism, Social Rejection, and Aggression. Personality and Social Psychology
Bulletin, 29, 261 - 272.

Young, S. M., & Pinsky, D. (2006). Narcissism and celebrity. Journal of Research in Personality,
40, 463–471.
34

Appendix A: Sample Characteristics

Table A-1. Participant Gender Ratio

Frequency Percent Cumulative Percent
Male 43 26.5 26.5
Female 119 73.5 100.0
Total 162 100.0

Table A-2. Ethnic Composition of Sample

Frequency Percent Cumulative Percent
American Indian/Alaska Native 3 1.9 1.9
Asian 49 30.2 32.1
African American 17 10.5 42.6
Pacific Islander 1 .6 43.2
Hispanic 16 9.9 53.1
Caucasian 59 36.4 89.5
Other 17 10.5 100.0
Total 162 100.0

Table A-3. Age of Participants

Frequency Percent Cumulative Percent
18 19 11.7 11.9
19 40 24.7 36.9
20 50 30.9 68.1
21 30 18.5 86.9
22 13 8.0 95.0
23 3 1.9 96.9
24 2 1.2 98.1
25 2 1.2 99.4
29 1 .6 100.0
Total 160 98.8

Missing 2 1.2

Total 162 100.0

35

Table A-4. Participants’ Narcissistic Personality Inventory Scores

N Median Interquartile Range Minimum Maximum
NPI Score 160 16 12-22 4 33

Table A-5. Participants’ Year in College

Frequency Percent Cumulative Percent
1 22 13.6 13.6
2 45 27.8 41.4
3 58 35.8 77.2
4 36 22.2 99.4
6 1 .6 100.0
Total 162 100.0

Table A-6. Participants’ Current Relationship Status

Frequency Percent Cumulative Percent
Not in a serious relationship 89 54.9 54.9
In a serious relationship 73 45.1 100.0
Total 162 100.0

Table A-7. Relationship Duration (in Months) for those “Currently in a Relationship”

N Minimum Maximum Mean Std. Deviation
Duration 73 1 84 17.79 16.411

Table A-8. Participants’ Lifetime Relationship Status

Frequency Percent Cumulative Percent
Never been in a serious relationship 30 18.5 18.5
Been in a serious relationship 132 81.5 100.0
Total 162 100.0

36

Figure 1. Distributions for the Total Scores of the Six Coded Variables (Verbal Aggression,
Anger, Distress/Anxiety, Doubts, Positive Self-Statements, Desire to End Relationship) &
Narcissism
37

Appendix B: ATSS Scenarios
SCENARIOS FOR FEMALES
Neutral
Narrator: It’s Saturday night and you are at a party with your boyfriend. The party is
packed with college kids; as usual there are some familiar faces and some
unfamiliar ones. You ask your boyfriend whether he wants a drink and he says
“yes” so you head over to where the drinks are to get two beers.

Narrator: Since the party is pretty crowded, it takes you a couple of minutes to get the
beers. You head back to where your boyfriend was standing and you catch a
glimpse of him talking to someone whom you vaguely recognize.

Narrator: You are now close enough to overhear the conversation.
Girl: Yeah, this semester has been really busy. I don’t have much time to socialize. I
spend all my free time with my boyfriend.

Girl: On that note, where is your girlfriend? I haven’t seen her in ages! We really
need to catch up.
Boyfriend: Oh, she should be right back. She just went to get drinks.
Girl: Ok. I’ll just wait here with you then.

Narrator: The voice you hear sounds familiar and you realize that your boyfriend is
talking to one of your good friends whom you haven’t seen in a while. They
notice that you are back and your friend looks very happy to see you.

Low Threat
Narrator: You are still at the party. You come back from the bathroom and your boyfriend
is no longer standing where you left him. You look around and catch a glimpse
of him talking to an attractive girl he has class with.

Narrator: As you head over to where they are standing, it looks like they are getting along
pretty well.

Narrator: You are now close enough to overhear part of their conversation.
Boyfriend: It would be great if you could fit me into your schedule.
Girl: Yeah, I think I can make Tuesday work.

Boyfriend: Last time we got together, it really helped…and it was fun.
Girl: I am glad I could help.

Boyfriend: Great. Let’s meet at Starbucks so we can get coffee before we start studying.
Girl: Sounds good. Enjoy the party. I look forward to seeing you on Tuesday.

High Threat
Narrator: You are still at the same party and you’ve been having a pretty good time. Your
phone rings and you see that it’s your roommate. It’s a little noisy inside so you
decide to step out to take the call.
38

Narrator: You are done with your brief conversation and you happen to be standing next
to an open window. You overhear part of a conversation.
Ex-girlfriend: So that girl I saw you with earlier…she is really hot.
Boyfriend: Yeah, she’s alright but she’s nothing special.

Narrator: You recognize your boyfriend’s voice.
Ex-girlfriend: I hear you two are very serious.
Boyfriend: It’s actually very casual.

Narrator: Since it is dark outside, you can see them, but they can’t see you at all. You
catch a glimpse of the girl your boyfriend is talking to and realize it is one of his
really good-looking ex-girlfriends.
Ex-girlfriend: Want to come over for a little after-party fun?
Narrator: Your boyfriend leans over and whispers something into her ear. She blushes and
giggles.

Narrator: Someone drunkenly brushes past you and you turn around for a second. By the
time you look back, the two are nowhere in sight. You call your boyfriend’s cell
and it goes straight to voicemail.

SCENARIOS FOR MALES
Neutral
Narrator: It’s Saturday night and you are at a party with your girlfriend. The party is
packed with college kids; as usual there are some familiar faces and some
unfamiliar ones. You ask your girlfriend whether she wants a drink and she says
“yes” so you head over to where the drinks are to get two beers.

Narrator: Since the party is pretty crowded, it takes you a couple of minutes to get the
beers. You head back to where your girlfriend was standing and you catch a
glimpse of her talking to someone whom you vaguely recognize.

Narrator: You are now close enough to overhear the conversation.
Guy: Yeah, this semester has been really busy. I don’t have much time to socialize. I
spend all my free time with my girlfriend.

Guy: On that note, where is your boyfriend? I haven’t seen him in ages! We really
need to catch up.
Girlfriend: Oh, he should be right back. He just went to get drinks.
Guy: Ok. I’ll just wait here with you then.

Narrator: The voice you hear sounds familiar and you realize that your girlfriend is talking
to one of your good friends whom you haven’t seen in a while. They notice that
you are back and your friend looks very happy to see you.

39

Low Threat
Narrator: You are still at the party. You come back from the bathroom and your girlfriend
is no longer standing where you left her. You look around and catch a glimpse
of her talking to an attractive guy she has class with.

Narrator: As you head over to where they are standing, it looks like they are getting along
pretty well.

Narrator: You are now close enough to overhear part of their conversation.
Girlfriend: It would be great if you could fit me into your schedule.
Guy: Yeah, I think I can make Tuesday work.

Girlfriend: Last time we got together, it really helped…and it was fun.
Guy: I am glad I could help.

Girlfriend: Great. Let’s meet at Starbucks so we can get coffee before we start studying.
Guy: Sounds good. Enjoy the party. See you on Tuesday.

High Threat
Narrator: You are still at the same party and you’ve been having a pretty good time. Your
phone rings and you see that it’s your roommate. It’s a little noisy inside so you
decide to step out to take the call.

Narrator: You are done with your brief conversation and you happen to be standing next
to an open window. You overhear part of a conversation.
Ex-boyfriend: So that guy I saw you with earlier…he is really hot.
Girlfriend: Yeah, he’s alright but he’s nothing special.

Narrator: You recognize your girlfriend’s voice.

Ex-boyfriend: I hear you two are very serious.
Girlfriend: It’s actually very casual.

Narrator: Since it is dark outside, you can see them, but they can’t see you at all. You
catch a glimpse of the guy your girlfriend is talking to and realize it is one of her
really good-looking ex-boyfriends.
Ex-boyfriend: Want to come over for a little after-party fun?
Narrator: Your girlfriend leans over and whispers something into his ear. He smirks.

Narrator: Someone drunkenly brushes past you and you turn around for a second. By the
time you look back, the two are nowhere in sight. You call your girlfriend’s cell
and it goes straight to voicemail.

40

Appendix C: Narcissistic Personality Inventory (NPI)

Instructions: The following are pairs of statements with which you may or may not identify.
Consider this example: A) “I like having authority over people”, versus B) “I don’t mind
following orders”. If you identify more with “liking to have authority over people”, than with
“not minding following orders”, then you would choose option A).
You may identify with both A) and B). In this case you should choose the statement which
seems closer to your personal feelings about yourself. Or, if you do not identify with either
statement, select the one which is least objectionable or remote. Indicate your answer by circling
the letter (“A” or “B”) that corresponds to the statement you pick for each item.

1. A I have a natural talent for influencing people.
B I am not good at influencing people.

2. A Modesty doesn’t become me.
B I am essentially a modest person.

3. A I would do almost anything on a dare.
B I tend to be a fairly cautious person.

4. A When people complement me I sometimes get embarrassed.
B I know that I am good because everybody keeps telling me so.

5. A The thought of ruling the world frightens the hell out of me.
B If I ruled the world, it would be a much better place.

6. A I can usually talk my way out of anything.
B I try to accept the consequences of my behavior.

7. A I prefer to blend in with the crowd.
B I like to be the center of attention.

8. A I will be a success.
B I am not too concerned about success.

9. A I am no better or worse than most people.
B I think I am a special person.

10. A I am not sure if I would make a good leader.
B I see myself as a good leader.

11. A I am assertive.
B I wish I were more assertive.

12. A I like having authority over people.
B I don’t mid following orders.

41

13. A I find it easy to manipulate people.
B I don’t like it when I find myself manipulating people.

14. A I insist upon getting the respect that is due me.
B I usually get the respect I deserve.

15. A I don’t particularly like to show off my body.
B I like to display my body.

16. A I can read people like a book.
B People are sometimes hard to understand.

17. A If I feel I am competent, I am willing to take responsibility for making decisions.
B I like to take responsibility for making decisions.

18. A I just want to be reasonably happy.
B I want to amount to something in the eyes of the world.

19. A My body is nothing special.
B I like to look at my body.

20. A I try not to be a show off.
B I am apt to show off if I get the chance.

21. A I always know what I am doing.
B Sometimes I am not sure of what I am doing.

22. A I sometimes depend on people to get things done.
B I rarely depend on anyone else to get things done.

23. A Sometimes I tell good stories.
B Everybody likes to hear my stories.

24. A I expect a great deal from other people.
B I like to do things for other people.

25. A I will never be satisfied until I get what I deserve.
B I take my satisfactions as they come.

26. A Compliments embarrass me.
B I like to be complimented.

27. A I have a strong will to power.
B Power for its own sake doesn’t interest me.

28. A I don’t very much care about new fads and fashions.
B I like to start new fads and fashions.

42

29. A I like to look at myself in the mirror.
B I am not particularly interested in looking at myself in the mirror.

30. A I really like to be the center of attention.
B It makes me uncomfortable to be the center of attention.

31. A I can live my life in any way I want to.
B People can’t always live their lives in terms of what they want.

32. A Being an authority doesn’t mean that much to me.
B People always seem to recognize my authority.

33. A I would prefer to be a leader.
B It makes little difference to me whether I am a leader or not.

34. A I am going to be a great person.
B I hope I am going to be successful.

35. A People sometimes believe what I tell them.
B I can make anybody believe anything I want them to.

36. A I am a born leader.
B Leadership is a quality that takes a long time to develop.

37. A I wish someone would someday write my biography.
B I don’t like people to pry into my life for any reason.

38. A I get upset when people don’t notice how I look when I go out in public.
B I don’t mind blending into the crowd when I go out in public.

39. A I am more capable than other people.
B There is a lot I can learn from other people.

40. A I am much like everybody else.
B I am an extraordinary person.

43

Appendix D: ATSS Coding Procedures
Please use the following 4-point Likert scale to rate the categories described below.
0 – not at all
1 – slightly/somewhat
2 – moderately
3 – very

Please take into consideration the tone of voice and other verbal cues of the verbalizations,
like how loud they are speaking, when coding!!! These factors help us understand the
meaning of what the person is saying.

Coding Categories:

1. Verbal Aggression and Intent to Harm
This coding category captures statements referring to hostile or violent intentions or
behaviors towards another person in the scenarios. The category also denotes a
readiness to attack or confront another person and includes derogatory statements about
other individuals.

Examples
I would go over and slap him.
I’d like to absolutely destroy him.
I’d confront her later.
There might be a beat down in session for the guy [for trying to overstep his bounds].
This is going to end in a pretty big fight because it’s inappropriate behavior.

2. Expression of General Anger
This coding category captures statements referring to feelings of strong displeasure,
antagonism and/or annoyance. It also denotes frustration and resentment towards other
individuals. This category is related to but is different from the preceding one, Verbal
Aggression and Intent to Harm, because it concerns how the person feels, not any action
that the person is contemplating.

Examples
It’s annoying that he left for a second time and is not watching out for me.
I would be kind of irked by it.
I’d be very pissed off at him.
I don’t need to spend my time on this jerk.
I’m definitely very angry/furious at this point.
Pretty much rage would be a good [description of my feelings].
How dare he!

3. Distress/Anxiety/Hurt
This coding category captures statements referring to feelings of worry, nervousness,
unease, sorrow, pain and emotional suffering.

Examples
I am extremely worried.
This would make me sad and depressed.
44

I am the guy who is “nothing special” and I'm obviously hurt by this.
I would feel like crap for the rest of the party.
I am getting a dreading sensation in my stomach.
I’m very concerned.”
I'm legitimately freaking out because I'm just assuming that she is running off with this
ex-boyfriend.

4. Doubts about relationship and/or partner commitment
This coding category captures statements referring to the feeling/belief that one cannot
rely on one’s partner, that he/she is not trustworthy and/or not committed to the romantic
relationship.

Examples
I would be questioning, I would be doubtful.
I am panicked about my relationship and feel an overwhelming fear of abandonment.
I think she might be leaving me.

5. Positive Self-Evaluations
This coding category captures statements presenting the self in a positive light. These
include expressions of high self-esteem and self-enhancing statements.

Examples
An attractive girl, so what? I’m not unattractive myself, so it’s okay.
I’m pretty smart myself.
He has a girlfriend so he is no threat to me; obviously he wouldn’t be a threat anyway.
There is no way that she can compete with me or that he’s interested in her.

6. Decision to end relationship – Yes/No
This coding category captures explicit statements regarding a decision to end the current
romantic relationship. The category is coded “yes” only if there is an explicit statement
specifically referring to the end of the relationship (as opposed to a general tone of the
verbalizations that might imply a desire to end the relationship, i.e. no inferences should
be made by the coder).

Examples of “Yes” statements
I’m pissed. We're no longer together.
I should dump him.
I would totally break up with him.
We’re done. We’re over.
I would just leave, give him back all his stuff and never talk to him again.

Abstract (if available)

Abstract Prior studies document that within the domain of romantic relationships dispositional narcissism can be associated with relationship damaging behaviors such as game-playing tendencies (Campbell, Foster & Finkel, 2002) but also with self-reports of relationship preserving behaviors such as resistance to doubts about a partner’s lack of commitment (Foster & Campbell, 2005). The present study investigated the association between dispositional narcissism and responses to escalating levels of threatening information regarding partner infidelity in a novel manner using the Articulated Thoughts in Simulated Situations (ATSS) paradigm. The ATSS is a think-aloud method that captures individuals’ thought content as a structured, experimenter-controlled, hypothetical situation unfolds. One hundred sixty-two undergraduate students verbalized their thoughts in response to three scenarios, in which their romantic partner interacted with an individual of the opposite sex. Participant responses were content-analyzed for the presence of verbal aggression, anger, distress, doubts, positive self-statements and desire for relationship dissolution. Results indicated that level of narcissism was not related to verbalizations of any of the abovementioned codes in the neutral scenario (“no threat” condition)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Age differences in jealousy: an ATSS examination of the role of attributions in the emotional reactions of older and younger adults in relationship-threatening situations

PDF

Assessing therapeutic change mechanisms in motivational interviewing using the articulated thoughts in simulated situations paradigm

PDF

Exploring experiences of minority-stress and alcohol use among sexual minorities using articulated thoughts in simulated situations: a proof of concept study

PDF

In-flight turbulence: an articulated thoughts in simulated situations (ATSS) paradigm investigation of air travelers ' experiences

PDF

Personalized normative feedback applied to undergraduates with problem drinking: a comparison with psychoeducation and an examination of cognitive-affective change mechanisms via the articulated ...

PDF

The articulated thoughts of heterosexual male college students in reaction to anti-gay hate speech

PDF

Reacting to the negative consequences of your decisions: an ATSS comparison of cognitive and affective reactions in older versus younger adults

Asset Metadata

Creator Babeva, Kalina Nikolaeva (author)

Core Title Measuring narcissistic cognitions in response to relationship threat using articulated thoughts in simulated situations

School College of Letters, Arts and Sciences

Degree Master of Arts

Degree Program Psychology

Publication Date 02/08/2011

Defense Date 10/27/2010

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag ATSS,narcissism,OAI-PMH Harvest,relationship threat,verbal aggression

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Davison, Gerald C. (committee chair), John, Richard S. (committee member), Lopez, Steven (committee member)

Creator Email babeva@usc.edu,kababeva@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m3647

Unique identifier UC1272114

Identifier etd-Babeva-4266 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-426884 (legacy record id),usctheses-m3647 (legacy record id)

Legacy Identifier etd-Babeva-4266.pdf

Dmrecord 426884

Document Type Thesis

Rights Babeva, Kalina Nikolaeva

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu