Lawrence Ian Reed
Lily Cooke
Isabella Kasaba1
Jill M. Hooley- 1Department of Psychology, New York University, New York, NY, United States
- 2Department of Psychology, La Salle University, Philadelphia, PA, United States
- 3Department of Psychology, Harvard University, Cambridge, MA, United States
Introduction: The characteristic behaviors we use to define personality pathology arise from specific interpersonal interactions. In an effort to create a laboratory-based context in which behavior might be expected to be influenced by particular personality traits, we used four 2-person, 2-choice games (the Prisoner’s Dilemma, Chicken, Leader, and Hero games) to create a simulated interaction and focused specifically on narcissism and dependency.
Method: An online sample of 1137 (35% male, M age = 38.46 years, SD age = 13.20) participants completed brief, self-reported measures of trait narcissism and dependency and played one of the four games. Before deciding how to act or react, participants received either no message, a promise to cooperate, or a threat to defect from a (confederate) partner.
Results: When receiving no message, those who cooperated in the Prisoner’s Dilemma had lower trait narcissism, while those who defected in the Chicken and Leader games had higher trait narcissism. Also with no message, participants who cooperated in the Hero game had higher trait dependency. Promises only affected the relationship between trait narcissism in the Leader game while threats only affected the relationship between trait dependency in the Chicken game.
Discussion: These findings add to the limited behavioral research on personality pathology and largely support established interpersonal conceptualizations and models. Future work might extend these findings using even more ecologically valid approaches to explore the behavioral correlates of personality traits that have important implications for interpersonal interactions.
1 Introduction
1.1 The interpersonal perspective
Interpersonal dysfunction is a core feature of personality pathology that encompasses many of the peripheral features used to define the personality disorders. These include characteristic views of ourselves, others, and the world (1–3), motivations and emotions (4–7), as well as beliefs and attitudes (8). For example, a sense of entitlement (in narcissistic personality disorder) or a lack of remorse (in antisocial personality disorder), might motivate opportunistic behaviors that disregard the welfare of others. In contrast, the fear of losing a caretaker’s support (in dependent personality disorder) might motivate self-sacrificing behaviors that forgo one’s personal needs in order to keep others happy. As such, it has been argued that the most useful conceptualizations of personality disorders are interpersonal (9–13).
Traditional interpersonal perspectives propose that personality pathology is best understood as a stable pattern of thoughts, feelings, and motivations within the self that lead to characteristic behaviors when interacting with others (13). Furthermore, the interpersonal perspective assumes that the ultimate motivation behind these interpersonal processes and behaviors is an attempt to alleviate anxiety and maintain a positive self-image (13, 14). From this view, interpersonal behavior not only defines personality disorders, but also distinguishes them from other forms of psychopathology (15–17).
Contemporary integrative interpersonal theory [CIIT; (18–20)] retains the emphasis on the self in relation to others. Here, adaptive interpersonal functioning is defined by goal-oriented behavior that satisfies the needs of the self and another relative to developmental context and culture (20). In contrast, interpersonal dysfunction is defined as maladaptive social behavior resulting from a breakdown in regulating the self, emotions, or the interaction with another. It is argued that this dysfunction is a defining feature of personality pathology and can be captured in the interpersonal situation, which relates the self and another through perceptions and behaviors (13, 17, 20, 21).
The long tradition and continued influence of the interpersonal perspective is supported in the theoretical and empirical literature. Measures of psychosocial functioning significantly differ between those with and without personality disorders (22–25). This reinforces the idea that interpersonal dysfunction is a central component of personality pathology. The perspective has also formed a useful basis for understanding personality pathology. For example, empirically derived models such as the interpersonal situation (13, 17, 21) and interpersonal circumplex (14, 26) describe affects and motivations within the self in relation to another.
1.1.1 Narcissism
Specifically, narcissistic personality disorder (NPD) is defined as a pervasive pattern of grandiosity, need for admiration, and lack of empathy (27). The diagnostic criteria for NPD capture the interpersonal dysfunction defined above. These include criteria directly related to the self in relation to another (grandiose sense of self-importance, belief that they are special and unique, requiring excessive admiration, a sense of entitlement, a lack of empathy, and an envy of others) as well as behaviors towards another (interpersonally exploitative and showing arrogant or haughty behaviors).
From the interpersonal perspective, those with NPD prioritize individual gains in status, esteem, and accomplishment at the expense of forming close, cooperative relationships with others (28–33). This overarching theme is consistent with self-reported data finding that those high in NPD rate themselves as more domineering (34), assertive (35), and antagonistic (35, 36). It is also consistent with interpersonal behaviors leaving those with NPD mired in conflict. For example, NPD is associated with problematic interpersonal behaviors such as violence (37) hostility (38) and aggression (39), often as a result of provocation from others. Consequently, those with NPD often have impaired social and occupational functioning (40) and significant distress (41).
1.1.2 Dependency
Dependent personality disorder is defined as a pervasive and excessive need to be taken care of that leads to submissive and clinging behavior and fears of separation (27). As with NPD, many of the diagnostic criteria capture the interpersonal dysfunction described above. Criteria related to the self in relation to another include a need for reassurance from others, a need for others to assume responsibility for major areas in their life and feelings of discomfort and helplessness when alone. Criteria related to behaviors towards another include difficulty expressing disagreement, going to excessive lengths to obtain nurturance and support from others, discomfort when alone, and seeking another relationship when a close one ends. Taken together, the diagnostic criteria for DPD emphasize the inadequate and helpless features of the disorder (42).
These features create an interpersonal profile that contrasts with that of NPD. The characteristic beliefs of inadequacy and feelings of helpless motivate those with DPD to prioritize cooperation and harmony in interpersonal relationships. Those with DPD have been described by clinicians as conscientious and eager to please (43). Similarly, self-reported data from those with DPD suggest they are agreeable, amiable, and eager to acquiesce to others’ demands and expectations (44). Unfortunately, this comes at the expense of individual needs, which can result in interpersonal dysfunction (45–47). This includes difficulties with social and occupational functioning (43, 48) high risk sexual behavior (49) and domestic violence (42).
1.2 Integrating clinical science and game theory
Despite the emphasis on interpersonal dysfunction within the diagnostic criteria as well as within the theoretical and empirical literature, there have been few studies examining how those with personality disorders behave in social situations. Most work on personality pathology has utilized self-report measures, rating scales, checklists, clinical interviews, or projective techniques (50) in lieu of behavioral measures. Furthermore, there is relatively little research on dependent and narcissistic personality disorders relative to other personality disorders across all empirical works (51). In sum, the difficulty capturing social behavior in personality pathology and the relative lack of objective behavioral research methods has impeded the development of new knowledge in this area.
A recent advance in the study of personality pathology has come from the integration of game theory into clinical science. Economic games allow for the examination of interpersonal behavior in formalized social situations designed to capture those experienced in the real world. These situations are objective and easily compared across studies. Furthermore, they are not susceptible to the biases inherent in recollection, self-report, or clinical judgement.
The interpersonal situations that individuals with personality pathology encounter can be closely modeled using games. For example, individuals with narcissistic personality disorder might expect favorable treatment and exploit others. This might result in contentious behavior in competitive interactions such as the Prisoner’s Dilemma (52), Chicken (53, 54), or Auction (55). In contrast, individuals with dependent personality disorder might have difficulty expressing disagreement or assuming responsibility. This might result in passive or selfless behavior in cooperative games such as Hero or The Battle of the Sexes (52). Moreover, although these models represent simplified interactions, they may be used to predict behaviors in real life situations and provide primary treatment targets (56).
Interpersonal behavior in borderline personality disorder has been examined using the trust game (57). Here, participants must decide whether to invest in an investee. If the participant invests and the investee reciprocates, both can profit. However, if the participant invests and the investee does not reciprocate, the participant will suffer a loss. When playing this game once with another participant, those with BPD have been shown to invest less in the trustee in comparison to control participants (58). When played repeatedly, those with BPD were less likely to invest over repeated exchanges (59).
Narcissistic personality disorder has been examined using the commons dilemma (60). Here, participants must decide whether to take an action benefiting themselves at the cost of the group, or to benefit the group at the cost of the self. It has been found that trait narcissism, as measured by the Narcissistic Personality Inventory (NPI-40; (61)) was positively associated with behavior beneficial to the self at the cost of the group.
Narcissism has also been examined in the Prisoner’s Dilemma (52). The Prisoner’s Dilemma is a 2-person, 2-choice game similar to the commons dilemma. Here, participants must decide whether to cooperate or defect. The outcome of the game is dependent upon the decision that each player makes. If both players cooperate, they both receive a modest payoff. If they both defect, they both receive a smaller payoff. However, if one player cooperates and the other player defects, the cooperating player earns nothing while the defecting player earns the highest possible payoff. It has been found that criminal psychopaths from psychiatric hospitals were greater than 7 times more likely to defect in comparison to men chosen from the general population (62). Similarly, participants’ scores on a measure of psychopathy were positively associated with defection over time in a repeated Prisoner’s Dilemma game (63).
Finally, trait narcissism and dependency have been modeled using the Battle of the Sexes (52). Here, the interests of both players are shared, though not perfectly so. That is, participants share an interest in coordinating an outcome (either A or B). However, one player prefers to coordinate on Outcome A, whereas the other prefers they both coordinate on Outcome B. Results suggest that trait narcissism predicted behaving in accordance with one’s own interest while trait dependency predicted behaving in accordance with the other’s interests. Crucially, these effects were only found when participants were provoked with a message (either a promise to act in accordance with the other’s interests or a threat to act in accordance with one’s own interests) before making their decision (64).
The aim of the current investigation was to expand upon the empirical studies of interpersonal behavior in economic games. Here, we assessed participants’ trait narcissism and dependency and examined their behavioral decisions in four archetypal 2-person, 2-choice games: the one-shot Prisoner’s Dilemma (presented here as the Closed Bag Exchange) (52), Chicken (53, 54), Leader (65); and Hero (52) games. Based on previous literature suggesting that personality pathology is evoked in response to provocation (64, 66) each game was played in one of three communication conditions. In the first condition, participants received no message from their purported partner. In the second and third conditions, participants received either a preemptive promise of cooperation or threat of defection, respectively. Based on the interpersonal perspective of personality disorders as well as the extant empirical work, we hypothesized that participants who scored more highly in trait narcissism would be more likely to defect in each game whereas participants who scored more highly in trait dependency would be more likely to cooperate in each game.
This study was preregistered on the Open Science Framework1 and we report all measures, manipulations, data exclusions, and sample size determinations. De-identified data are available on OSF.2
2 Method
2.1 Participants
We recruited 1,260 participants through Prolific, which generates high quality data (67) and can provide good participant reimbursement. We planned to collect a sample size of approximately 100 per condition, and we did not analyze the data until all of the responses were collected. Assuming that cooperation and defection rates would be equal, this sample size provides 80% power to detect an effect size of d = 0.57 in a two-tailed test with a 5% false-positive rate (calculated with G*Power). One-hundred and twenty-three participants were excluded (120 failed at least one attention check and 3 revoked consent), leaving 1,137 participants’ (396 males, 738 females, 3 other). Participant’s mean age was 38.46 years (SD = 13.20) and their ethnicities were as follows: 75 Asian, 34 Black, 35 mixed, 968 White, and 25 other). Participants were currently residing in the United Kingdom (992) or in the United States (145). This study was approved by the New York University Institutional Review Board (IRB-FY2021-5214).
2.2 Measures
2.2.1 Personality
We chose self-report measures of trait narcissism and dependency that were brief and applicable to both clinical and non-clinical samples. Trait narcissism was measured using the Narcissistic Personality Inventory [NPI-40; (61)]. In the NPI, participants are presented with 40 pairs of opposing statements and must choose the one that best describes them. For example, “I find it easy to manipulate people” (narcissistic response) and “I do not like it when I find myself manipulating other people” (non-narcissistic response). The NPI-40 assesses trait narcissism continuously on a range of non-clinical and clinical populations. Potential scores on the NPI range from 0 to 40. The NPI has shown good reliability and validity (68, 69). A recent meta-analysis has reported a mean population reliability coefficient of 0.82 (70).
Trait dependency was measured using the Dependent Personality Questionnaire (DPQ; 71). In the DPQ, participants are presented with eight questionnaire items and must rate the extent that it applies to them on a 4-point scale ranging from yes, definitely to no, not at all. Example items include “I am an independent person” and “I rely a lot on my family and friends.” Potential scores in the DPQ range from 0 to 24.
2.2.2 Games
Participants played one of four symmetrical 2 -person, 2-choice games in which the outcome jointly depended upon each player’s independent decision to cooperate or defect (see Figure 1). As with previous works (72), we use the terms cooperate and defect throughout for convenience. They are not accurate in describing behavior in all the games. We chose the four 2 × 2 games described by Rapoport (65) as motivating distinct behaviors: the Prisoner’s Dilemma (N = 282), Chicken (N = 288), Leader (N = 279), and Hero (N = 288) games. Each game was defined by the relative rank of four ordinal payoffs (73) which have been referred in the literature as the Reward (R), Punishment (P), Temptation to defect (T), and the Sucker’s Payoff (S). Here, each payoff was set at either 0 cents, 10 cents, 20 cents, or 30 cents.
Figure 1. Game payoffs of four archetypal 2 × 2 games. Column payoffs presented first within each cell. Shaded cells are natural outcomes.
The payoff structure for each game is presented in Table 1. The decision that will minimize the player’s loss should they incur a loss is referred to as the maximin strategy. When each player chooses the maximin strategy, the results is the natural outcome represented in each payoff matrix. The motivation to switch from the maximin strategy is different in each game (see below).
Table 1. Payoff Structures for closed bag exchange (Prisoner’s Dilemma), chicken, leader, and hero games.
The order of payoffs in the Prisoner’s Dilemma is S > P > R > T. The Prisoner’s Dilemma is unique among the 4 games in that it has a dominant strategy (i.e., Nash equilibrium) in which each player will gain a greater individual payoff by switching from the minimax strategy regardless of their partner’s decision.
The Chicken, Leader, and Hero games also differ from the Prisoner’s Dilemma in that each player could benefit by switching from the minimax strategy if the other player does not. In the Chicken game (T > R > S > P), a player who switches (defects) benefits themselves at the expense of the other player. In the Leader game (T > S > R > P), a player who switches (defects) benefits themselves and the other player with the larger benefit being conferred upon the self. Finally, in the Hero game (S > T > R > P), a player who switches (cooperates) benefits themselves and the other player with the larger benefit being conferred upon the other player.
In a between-subjects design, each game was presented in one of three conditions varying in communication. In the no communication condition, participants did not receive any message from the other player. In the promise and threat conditions, participants received a message (that was presented as having come from the other player) communicating an intention to act cooperatively or selfishly, respectively. Participants were told that the message may or may not have reflected the other player’s actual decision.
2.3 Procedure
Participants gave consent and read a description of the procedure. Participants read that they would answer questions about their personality and play a game with another Prolific participant. The order that participants completed the personality measures personality measures and game was counterbalanced to reduce any potential order effects. Participants earned $1.93 for completing the task, and they could earn an additional bonus depending on the decisions made by them and the other participant. Most participants completed the task in under 10 min (M = 8.20, SD = 5.20).
A set of instructions was presented before each game. Participants read:
In this game, you will be matched with another Prolific participant. Although the story behind the game is fictional, the earnings are real. The decision that you make will affect your earnings in the game. Any money you earn in the game will be paid to you as a bonus payment.
These instructions were followed by a backstory and description of earnings unique to each game.
Closed Bag Exchange (Prisoner’s Dilemma).
The backstory was written as follows:
You and the other participant are meeting to exchange closed briefcases with the understanding that each of them contains valuable goods. Both you and the other participant can choose to either honor the agreement by putting the goods into the briefcase or dishonor the agreement by leaving the briefcase empty.
The earnings were described as follows:
If you and the other participant both honor the agreement, you both get the agreed upon goods and earn 20 cents.
If you and the other participant both choose to dishonor the agreement, you both keep your own goods without benefiting from the exchange and earn 10 cents.
If you choose to honor the agreement and the other participant chooses to dishonor the agreement, you end up with nothing and earn 0 cents.
If you choose to dishonor the agreement and the other participant chooses to honor the agreement, you end up with your goods and your partner’s goods and earn 30 cents.
In those conditions with a purported message, participants also read:
Before you decide, you’ll receive a brief, typical, written message from the other participant stating their choice. This message may or may not accurately reflect their actual decision.
After reading the instructions, participants answered 2 comprehension questions (e.g., “Let us say that the other participant decides to honor the agreement. How much money will you earn in bonus payments if you decide to dishonor the agreement?”). If they answered incorrectly, they could try again. Participants had to answer correctly in order to proceed.
Before making their decision, participants were shown a payoff matrix outlining the earnings described above. In those conditions with a purported message, participants then read “Here is a typical message from your partner.” This was followed by “I promise to honor the agreement and put the goods in the briefcase!” in the promise condition and “No way am I honoring the agreement! I’m leaving my briefcase empty!” in the threat condition. Participants then decided whether to honor or dishonor the agreement.
Chicken.
The backstory was written as follows:
You and the other participant are driving towards a single lane bridge from opposite directions. Both you and the other participant can choose either to go straight or to swerve away.
The earnings were described as follows:
If you and the other participant both swerve away, you will both show weakness and earn 20 cents each.
If you and the other participant both go straight, you crash and both earn 0 cents.
If you choose to swerve and the other participant chooses to go straight, you lose face and earn 10 cents.
If you choose to go straight and the other participant chooses to swerve, you show more strength and earn 30 cents.
In those versions with a purported message, participants also read:
Before you decide, you’ll receive a brief, typical, written message from the other participant stating their choice. This message may or may not accurately reflect their actual decision.
As in the Closed Bag Exchange, participants had to correctly answer 2 comprehension questions in order to proceed. Participants were shown a payoff matrix outlining the earnings described above. In those conditions with a purported message, participants then read “Here is a typical message from your partner.” This was followed by “I’m definitely going to swerve!” in the promise condition and “No way I’m swerving! I’m going straight!” in the threat condition. Finally, participants decided whether to swerve or go straight.
Leader.
The backstory was written as follows:
You and the other participant are driving in the same direction in separate lanes and must merge into a single lane. Both you and the other participant can choose to either wait or drive forward.
The earnings were described as follows:
If you and the other participant both wait, you lose time and earn 10 cents.
If you and the other participant both drive forward, you crash and earn 0 cents.
If you choose to wait and the other participant chooses to drive forward, your wait is minimal and you earn 20 cents.
If you choose to drive forward and the other participant chooses to wait, you move along quickly and you earn 30 cents.
In those versions with a purported message, participants also read:
Before you decide, you’ll receive a brief, typical, written message from the other participant stating their choice. This message may or may not accurately reflect their actual decision.
As in the other games, participants had to correctly answer 2 comprehension questions in order to proceed. Participants were shown a payoff matrix outlining the earnings described above. In those conditions with a purported message, participants then read “Here is a typical message from your partner.” This was followed by “I’ll wait!” in the promise condition and “I’m driving forward! Wait!” in the threat condition. Finally, participants decided whether to wait or drive forward.
Hero.
The backstory was written as follows:
You and the other participant are in an argument. Both you and the other participant can choose to either go to the other person’s house and apologize or wait at your house for an apology from the other participant.
The earnings were described as follows:
If you and the other participant both go to the other’s house to apologize, you both find an empty house and cannot apologize. The argument continues and you both earn 0 cents.
If you and the other participant both wait for the other to come to your house to apologizer, the argument continues. You both save face and earn 10 cents.
If you choose to go apologize and the other participants waits at their house for the apology, you lose face. But, the argument is resolved and you earn 20 cents.
If you choose to wait at your house for an apology and the other participant goes to your house to apologize, you save face. In addition, the argument is resolved and you earn 30 cents.
As in the other games, participants had to correctly answer 2 comprehension questions in order to proceed. Participants were shown a payoff matrix outlining the earnings described above. In those conditions with a purported message, participants then read “Here is a typical message from your partner.” This was followed by “I’m going to your house to apologize!” in the promise condition and “No way I’m going to apologize! I’m waiting at home for your apology!” in the threat condition. Finally, participants decided to go apologize or wait for an apology.
3 Results
3.1 Preliminary results
As preliminary analyses, we first examined cooperation rates in each game with no communication. To examine the effect of purported messages, we then compared cooperation rates with no communication to those in response to the purported promise and threat. In the Closed Bag Exchange, participants were more likely to cooperate (75/96, 78%) than defect (21/96, 22%) with no communication, Χ2 (1, N = 96) = 30.357, p < 0.01, ϕ = 0.563. In comparison to no communication, participants were no more likely to cooperate (72/93, 76%) than defect (21/93, 23%) in response to the promise, Χ2 (1, N = 189) = 0.014, p > 0.10, ϕ = 0.008, and less likely to cooperate (34/93, 37%) than defect (59/93, 57%) in response to the threat, Χ2 (1, N = 189) = 33.433, p < 0.01, ϕ = 0.421.
In the Chicken game, participants were more likely to cooperate (69/98, 70%) than defect (29/98, 30%) with no communication, Χ2 (1, N = 98) = 16.327, p < 0.01, ϕ = 0.408. In comparison to no communication, participants were no more likely to cooperate (59/94, 63%) than defect (35/94, 37%) in response to the promise, Χ2 (1, N = 192) = 1.261, p > 0.10, ϕ = 0.081, or in response to the threat, Χ2 (1, N = 192) = 0.304, p > 0.10, ϕ = 0.040.
In the Leader game, participants were more likely to cooperate (65/95, 68%) than defect (30/95, 32%) with no communication, Χ2 (1, N = 95) = 12.895, p < 0.01, ϕ = 0.368. In comparison to no communication, participants were less likely to cooperate (16/92, 17%) than defect (76/92, 83%), in response to the promise, Χ2 (1, N = 187) = 49.569, p > 0.01, ϕ = 0.515, and more likely to cooperate (84/92, 91%) than defect (8/92, 9%) in response to the threat, Χ2 (1, N = 187) = 15.115, p > 0.01, ϕ = 0.284.
In the Hero game, participants were less likely to cooperate (21/93, 23%) than defect (72/93, 77%) with no communication, Χ2 (1, N = 93) = 27.968, p < 0.01, ϕ = 0.548. In comparison to no communication, participants were less likely to cooperate (7/97, 7%) than defect (92/97, 95%), in response to the promise, Χ2 (1, N = 192) = 9.261, p < 0.01, ϕ = 0.220, and more likely to cooperate (56/96, 58%) than defect (40/96, 42%), in response to the threat, Χ2 (1, N = 189) = 25.010, p < 0.01, ϕ = 0.364.
3.2 Primary analyses
The primary analyses examining group differences in trait narcissism among participants who decided to cooperate and defect in each game are depicted in Table 2. In our sample, scores on the NPI ranged from 0–36 (M = 8.93, SD = 6.58) and scores on the DPQ ranged from 0–24 (M = 9.15, SD = 3.68). Consistent with our hypotheses, trait narcissism was higher among those choosing to defect in the Closed Bag Exchange, Chicken, and Leader games with no communication. These effects were in the moderate range (Cohen’s d = −0.469 -- -0.546). However, trait narcissism was no higher among those choosing to defect in the Hero game with no communication. Furthermore, trait narcissism was no higher among those choosing the selfish decision in all but one of the games with communication: the Leader game with the promise.
Table 2. Narcissistic personality inventory scores by game, communication, and decision.
The primary analyses examining group differences in trait dependency among participants who decided to cooperate and defect in each game are depicted in Table 3. Contrary to our hypotheses, trait dependency was no higher among those choosing the selfish decision in the Bag Exchange, Chicken, or Hero games with no communication. Consistent with our hypothesis, trait dependency was higher among participants who decided to cooperate in the Hero game with no communication. This effect was near the moderate range (Cohen’s d = 0.468). Finally, trait dependency was no higher among those choosing the cooperative decision in all but one of the games with communication: the Chicken game with the threat.
Table 3. Dependent personality disorder questionnaire scores by game, communication, and decision.
4 Discussion
We examined participants’ behavior in four 2-person, 2-choice games in relation to trait narcissism and dependency in one of three preemptive message conditions. We hypothesized that participants with high trait narcissism would be more likely to defect while participants with high trait dependency would be more likely to cooperate in each game across conditions. Consistent with our hypotheses, trait narcissism was higher among participants who defected in the Closed Bag Exchange (Prisoner’s Dilemma), Chicken, and Leader games in comparison to those who cooperated when given no preemptive message. However, trait narcissism did not differ by behavior in the Hero game with no preemptive message. In response to preemptive messages, trait narcissism was no higher among participants who defected across games with one exception: those who defected in the Leader game reported higher trait narcissism compared to those who cooperated in response to a promise of cooperation.
These results are consistent with interpersonal conceptualizations emphasizing behaviors that prioritize the self over the other (13, 14, 74). Results from the no communication condition are also consistent with previous studies showing that psychopaths are more likely to defect in Commons (60) and Prisoner’s Dilemma (62, 63) games. They are, however, only partially consistent with previous research examining trait narcissism and dependency in the Battle of the Sexes (BoS) game. This research found that trait narcissism predicted participant defection in the BoS game only when participants were provoked with a preemptive threat from their partner (64). These results might be reconciled when considering participants’ partners. Here, participants were told they were playing with another Prolific participant while Simon and Reed (64) told participants to imagine they were playing with someone who was very close to them.
Also consistent with our hypotheses, trait dependency was higher among participants who cooperated in the Hero game in comparison to those who defected when given no preemptive message. However, trait dependency was no higher among participants who cooperated in the Closed Bag Exchange (Prisoner’s Dilemma), Chicken, or Leader games in comparison to those who defected when given no preemptive message. In response to preemptive messages, trait dependency was no higher among those who cooperated with one exception: those who cooperated in the Chicken game reported higher trait dependency compared to those who defected in response to the threat.
These results add nuance to interpersonal conceptualizations positing that individuals with dependent personality disorder behave submissively (43, 75, 76). They are also partially consistent with previous research. Simon and Reed (64) found that trait dependency predicted participant cooperation only when participants were provoked with a preemptive threat from their partner. Here, we found the opposite: trait dependency predicted cooperation only in the no communication condition. As with trait narcissism in the BoS game, these differences could be the result of partners’ portrayal (77, 78).
Clinically, these behaviors can be used to model those behaviors characteristic of individuals with trait narcissism and dependency in real life situations. For example, the selfish behaviors seen here among those high in trait narcissism in the Closed Bag Exchange (Prisoner’s Dilemma) game may relate to dysfunctional behavior in intimate relationships (79). In contrast, the selfless behaviors seen here among those high in trait dependency may related to the increased risk for physical abuse from spouses (80). Our findings also support the value of adopting a more dimensional and trait-focused approach to personality (81, 82) – an approach that is now receiving considerable theoretical attention (83, 84).
Our results should be interpreted within the context of the methodology and sample. The payoff structure inherent in each game created a unique interpersonal situation for participants to behave in a range of ways. This could account, in part, for varying behaviors among individuals with similar personality pathology across different games. Furthermore, as shown in the preliminary analyses, preemptive promises and/or threats affected participants’ behavior in every game aside from Chicken. Additionally, factors such as group identity, anonymity, repetition, and the relative magnitude of game payoffs have all been shown to affect behavior (85–87). Furthermore, our online sample was a non-clinical population consisting mostly of females. Females have been shown to differ from males in the prevalence and presentation of NPD ((88–90) and DPD (88). We also do not know if our participants would have behaved differently had they interacted with their partners in person. That said, it is possible that not seeing their interactional partner allowed participants to behave in a more authentic manner than might otherwise have been the case. It also warrants mention that many online interactions are characterized by anonymity and a lack of face-to-face interaction. As such our findings may have more ecological validity than might be apparent at first glance. Finally, an updated validation of the NPI-40 might further contextualize the validity and reliability of our data.
In sum, interpersonal behavior in economic games can be used to test interpersonal conceptualizations of personality pathology. Here, we found that those behaving selfishly in the Chicken and Leader games reported higher trait narcissism while those behaving cooperatively in the Prisoner’s Dilemma game reported lower trait narcissism in instances without communication. This suggests a narcissistic preference for self-interested behavior regardless of its effect on the other. In addition, we found that those behaving cooperatively in the Hero game reported higher trait dependency in instances without communication. This suggests a dependent preference to sacrifice a benefit in order to help another. Further research utilizing similar methods stands to increase our understanding of personality pathology across social situations and levels of communication.
Data availability statement
Publicly available datasets were analyzed in this study. The de-identified data can be found on the Open Science Framework website: https://osf.io/xturq/?view_only=e193967749354534b9c06cfc587ddbc3; https://osf.io/hxzse/?view_only=82f73da95df549bf95822181b7b05df5. Further inquiries can be directed to the corresponding author.
Ethics statement
The studies involving humans were approved by New York University Institutional Review Board. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.
Author contributions
LR: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing – original draft, Writing – review & editing. LC: Writing – review & editing. IK: Writing – review & editing. EH: Writing – review & editing. JH: Conceptualization, Supervision, Writing – review & editing.
Funding
The author(s) declare that no financial support was received for the research, authorship, and/or publication of this article.
Conflict of interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Footnotes
1. ^https://osf.io/xturq/?view_only=e193967749354534b9c06cfc587ddbc3
2. ^https://osf.io/hxzse/?view_only=82f73da95df549bf95822181b7b05df5
References
1. Blatt, SJ, and Lerner, H. The psychological assessment of object representation. J Pers Assess. (1983) 47:7–28. doi: 10.1207/s15327752jpa4701_2
2. Kernberg, OF. An Ego psychology-object relations theory approach to the transference. Psychoanal Q. (1987) 56:197–221. doi: 10.1080/21674086.1987.11927172
3. Livesley, WJ, and Jang, K. Toward an empirically based classification of personality disord. J Pers Disord. (2000) 14:15. doi: 10.1521/pedi.2000.14.2.137
4. Baskin-Sommers, A, Krusemark, E, and Ronningstam, E. Empathy in narcissistic personality disorder: from clinical and empirical perspectives. Personal Disord Theory Res Treat. (2014) 5:323–33. doi: 10.1037/per0000061
5. Ebner-Priemer, UW, Houben, M, Santangelo, P, Kleindienst, N, Tuerlinckx, F, Oravecz, Z, et al. Unraveling affective dysregulation in borderline personality disorder: a theoretical model and empirical evidence. J Abnorm Psychol. (2015) 124:186–98. doi: 10.1037/abn0000021
6. Reed, LI, Fitzmaurice, G, and Zanarini, MC. The course of positive affective and cognitive states in borderline personality disorder: a 10-year follow-up study: positive affective and cognitive states in BPD. Personal Ment Health. (2012) 6:281–91. doi: 10.1002/pmh.1197
7. Reed, LI, Fitzmaurice, G, and Zanarini, MC. The relationship between childhood adversity and dysphoric inner states among borderline patients followed prospectively for 10 years. J Personal Disord. (2015) 29:408–17. doi: 10.1521/pedi_2013_27_086
8. Beck, AT, Butler, AC, Brown, GK, Dahlsgaard, KK, Newman, CF, and Beck, JS. Dysfunctional beliefs discriminate personality disorders. Behav Res Ther. (2001) 39:1213–25. doi: 10.1016/S0005-7967(00)00099-1
9. Benjamin, L. Interpersonal diagnosis and treatment of personality disorders. New York: Guilford Press (1996).
10. Benjamin, L. Interpersonal reconstructive therapy: Promoting change in nonresponders. New York: Guilford Press (2003).
12. Horowitz, L. Interpersonal foundations of psychopathology. Washington, DC: American Psychological Association (2004).
15. Bender, DS, Morey, LC, and Skodol, AE. Toward a model for assessing level of personality functioning in DSM–5, part I: a review of theory and methods. J Pers Assess. (2011) 93:332–46. doi: 10.1080/00223891.2011.583808
16. Hopwood, CJ, Wright, AGC, Ansell, EB, and Pincus, AL. The interpersonal Core of personality pathology. J Personal Disord. (2013) 27:270–95. doi: 10.1521/pedi.2013.27.3.270
17. Pincus, AL, Cain, NM, and Halberstadt, AL. Importance of self and other in defining personality pathology. Psychopathology. (2020) 53:133–40. doi: 10.1159/000506313
18. Pincus, AL. A contemporary integrative interpersonal theory of personality disorders In: MF Lenzenweger and JF Clarkin, editors. Major theories of personality disorder. 2nd ed. New York: Guilford Press (2005). 282–331.
19. Pincus, AL, and Hopwood, CJ. A contemporary interpersonal model of personality pathology and personality disorder In: TA Widiger, editor. The Oxford handbook of personality disorders. New York, NY: Oxford University Press (2012)
20. Wright, AGC, Ringwald, WR, Hopwood, CJ, and Pincus, AL. It’s time to replace the personality disorders with the interpersonal disorders. Am Psychol. (2022) 77:1085–99. doi: 10.31234/osf.io/7syvf
21. Hopwood, CJ, Pincus, AL, and Wright, AGC. The interpersonal situation: integrating personality assessment, case formulation, and intervention In: D Samuel and D Lynam, editors. Using basic personality research to inform personality pathology. New York, NY: Oxford University Press (2019). 94–121.
22. Beeney, JE, Lazarus, SA, Hallquist, MN, Stepp, SD, Wright, AGC, Scott, LN, et al. Detecting the presence of a personality disorder using interpersonal and self-dysfunction. J Personal Disord. (2019) 33:229–48. doi: 10.1521/pedi_2018_32_345
23. DeFife, JA, Goldberg, M, and Westen, D. Dimensional assessment of self- and interpersonal functioning in adolescents: implications for DSM-5 ‘s general definition of personality disorder. J Personal Disord. (2015) 29:248–60. doi: 10.1521/pedi_2013_27_085
24. Kampe, L, Zimmermann, J, Bender, D, Caligor, E, Borowski, A-L, Ehrenthal, JC, et al. Comparison of the structured DSM–5 clinical interview for the level of personality functioning scale with the structured interview of personality organization. J Pers Assess. (2018) 100:642–9. doi: 10.1080/00223891.2018.1489257
25. Morey, LC, Bender, DS, and Skodol, AE. Validating the proposed diagnostic and statistical manual of mental disorders, 5th edition, severity Indicator for personality disorder. J Nerv Ment Dis. (2013) 201:729–35. doi: 10.1097/NMD.0b013e3182a20ea8
26. Freedman, MB, Leary, TF, Ossorio, AG, and Goffey, HS. The interpersonal dimension of Personality1. J Pers. (1951) 20:143–61. doi: 10.1111/j.1467-6494.1951.tb01518.x
27. American Psychiatric Association, D. S. M. T. F., and American Psychiatric Association. Diagnostic and statistical manual of mental disorders: DSM-5. (Vol. 5). Washington, DC: American psychiatric association (2013).
28. Bradlee, PM, and Emmons, RA. Locating narcissism within the interpersonal Circumplex and the five-factor model. Personal Individ Differ. (1992) 13:821–30. doi: 10.1016/0191-8869(92)90056-U
29. Campbell, WK, Bush, CP, Brunell, AB, and Shelton, J. Understanding the social costs of narcissism: the case of the tragedy of the commons. Personal Soc Psychol Bull. (2005) 31:1358–68. doi: 10.1177/0146167205274855
30. Campbell, WK, Reeder, GD, Sedikides, C, and Elliot, AJ. Narcissism and comparative self-enhancement strategies. J Res Pers. (2000) 34:329–47. doi: 10.1006/jrpe.2000.2282
31. Carroll, L. A study of narcissism, affiliation, intimacy, and power motives among students in business administration. Psychol Rep. (1987) 61:355–8. doi: 10.2466/pr0.1987.61.2.355
32. Emmons, RA. Factor analysis and construct validity of the narcissistic personality inventory. J Pers Assess. (1984) 48:291–300. doi: 10.1207/s15327752jpa4803_11
33. Morf, CC, Weir, C, and Davidov, M. Narcissism and intrinsic motivation: the role of goal congruence. J Exp Soc Psychol. (2000) 36:424–38. doi: 10.1006/jesp.1999.1421
34. Pincus, AL, and Wiggins, JS. Interpersonal problems and conceptions of personality disorders. J Personal Disord. (1990) 4:342–52. doi: 10.1521/pedi.1990.4.4.342
35. Samuel, DB, and Widiger, TA. Convergence of narcissism measures from the perspective of general personality functioning. Assessment. (2008) 15:364–74. doi: 10.1177/1073191108314278
36. Wright, AGC, Stepp, SD, Scott, LN, Hallquist, MN, Beeney, JE, Lazarus, SA, et al. The effect of pathological narcissism on interpersonal and affective processes in social interactions. J Abnorm Psychol. (2017) 126:898–910. doi: 10.1037/abn0000286
37. Bushman, BJ, and Baumeister, RF. Threatened egotism, narcissism, self-esteem, and direct and displaced aggression: does self-love or self-hate Lead to violence? J Pers Soc Psychol. (1998) 11:219–29. doi: 10.1037//0022-3514.75.1.219
38. Rasmussen, K. Entitled vengeance: a meta‐analysis relating narcissism to provoked aggression. Aggress Behav. (2016) 42:362–79. doi: 10.1002/ab.21632
39. Thomaes, S, Bushman, BJ, Stegge, H, and Olthof, T. Trumping shame by blasts of noise: narcissism, self-esteem, shame, and aggression in young adolescents. Child Dev. (2008) 79:1792–801. doi: 10.1111/j.1467-8624.2008.01226.x
40. Miller, JD, Keith Campbell, W, and Pilkonis, PA. Narcissistic personality disorder: relations with distress and functional impairment. Compr Psychiatry. (2007) 48:170–7. doi: 10.1016/j.comppsych.2006.10.003
41. Pincus, AL, Ansell, EB, Pimentel, CA, Cain, NM, Wright, AGC, and Levy, KN. Initial construction and validation Fo the pathological narcissism inventory. Psychol Assess. (2009) 21:365–79. doi: 10.1037/a0016530
42. Bornstein, RF. The complex relationship between dependency and domestic violence: converging psychological factors and social forces. Am Psychol. (2006) 61:595–606. doi: 10.1037/0003-066X.61.6.595
43. Bornstein, RF. The dependent patient: A Practitioner’s guide. Washington, DC: American Psychological Association (2005).
44. Lowe, JR, Edmundson, M, and Widiger, TA. Assessment of dependency, agreeableness, and their relationship. Psychol Assess. (2009) 21:543–53. doi: 10.1037/a0016899
45. Zuroff, DC, and Duncan, N. Self-criticism and conflict resolution in romantic couples. Can J Behav Sci. (1999) 31:137–49. doi: 10.1037/h0087082
46. Bornstein, RF. Heisenberg, Kandinsky, and the heteromethod convergence problem: Lessons from within and beyond psychology. J Pers Assess. (2009) 91:1–8.
47. Wilberg, T, Karterud, S, Pedersen, G, and Urnes, Ø. The impact of avoidant personality disorder on psychosocial impairment is substantial. Nord J Psychiatry. (2009) 63:390–6. doi: 10.1080/08039480902831322
49. Lavan, H, and Johnson, JG. The association between axis I and II psychiatric symptoms and high-risk sexual behavior during adolescence. J Pers Disord. (2002) 16:73–94.
50. Millon, T, Millon, CM, Meagher, SE, Grossman, SD, and Ramnath, R. Personality Disorders in Modern Life. Germany: Wiley (2004).
54. Smith, JM, and Price, G. The logic of animal conflict. Nature. (1973) 246:15–8. doi: 10.1038/246015a0
55. Shubik, M. The Dollar auction game: a paradox in noncooperative behavior and escalation. J Confl Resolut. (1971) 15:109–11. doi: 10.1177/002200277101500111
56. Monaghan, C, and Bizumic, B. Dimensional models of personality disorders: challenges and opportunities. Front Psych. (2023) 14:1098452. doi: 10.3389/fpsyt.2023.1098452
57. Berg, J, Dickhaut, J, and McCabe, K. Trust, reciprocity, and social history. Games Econ Behav. (1995) 10:122–42.
58. Unoka, Z, Seres, I, Áspán, N, Bódi, N, and Kéri, S. Trust game reveals restricted interpersonal transactions in patients with borderline personality disorder. J Personal Disord. (2009) 23:399–409. doi: 10.1521/pedi.2009.23.4.399
59. King-Casas, B, Sharp, C, Lomax-Bream, L, Lohrenz, T, Fonagy, P, and Montague, P. The rupture and repair of cooperation in borderline personality disorder. Science. (2008) 321:806–10. doi: 10.1126/science.1156902
60. Hardin, G. The tragedy of the commons: the population problem has no technical solution; it requires a fundamental extension in morality (1968) 162:1243–8. doi: 10.1126/science.162.3859.1243
61. Raskin, RN, and Hall, CS. A narcissistic personality inventory. Psychol Rep. (1979) 45:590–10. doi: 10.2466/pr0.1979.45.2.590
62. Mokros, A, Menner, B, Eisenbarth, H, Alpers, GW, Lange, KW, and Osterheider, M. Diminished cooperativeness of psychopaths in a Prisoner’s dilemma game yields higher rewards. J Abnorm Psychol. (2008) 117:406–13. doi: 10.1037/0021-843X.117.2.406
63. Rilling, JK, Glenn, AL, Jairam, MR, Pagnoni, G, Goldsmith, DR, Elfenbein, HA, et al. Neural correlates of social cooperation and non-cooperation as a function of psychopathy. Biol Psychiatry. (2007) 61:1260–71. doi: 10.1016/j.biopsych.2006.07.021
64. Simon, L, and Reed, LI. Narcissistic and dependent behaviors in the Battle of the sexes game. Personal Disord Theory Res Treat. (2021) 12:286–90. doi: 10.1037/per0000437
65. Rapoport, A. Exploiter, leader, Hero, and martyr: the four archetypes of the 2 x 2 game. Behav Sci. (1967) 12:81–4. doi: 10.1002/bs.3830120202
66. Kenrick, DT, and Funder, DC. Profiting from controversy. Am Psychol. (1988) 12:23–34. doi: 10.1037//0003-066x.43.1.23
67. Peer, E, Rothschild, D, Gordon, A, Evernden, Z, and Damer, E. Data quality of platforms and panels for online behavioral research. Behav Res Methods. (2022) 1.
68. Raskin, R, and Terry, H. A principal-components analysis of the Narcissistic Personality Inventory and further evidence of its construct validity. J Pers Soc Psychol. (1988) 54:890–902.
69. Rhodewalt, F, and Morf, CC. Self and interpersonal correlates of the Narcissistic Personality Inventory: A review and new findings. J Res Pers. (1995) 29:1–23.
70. Miller, BK, Nicols, KM, Clark, S, Daniels, A, and Grant, W. Meta-analysis of coefficient alpha for scores on the narcissistic personality inventory. PLoS One. (2018) 13:e0208331. doi: 10.1371/journal.pone.0208331
71. Tyrer, P, Morgan, J, and Cicchetti, D. The Dependent Personality Questionnaire (DPQ): a screening instrument for dependent personality. Int J Soc Psychiatry. (2004) 50:10–7.
72. Browning, L, and Colman, AM. Evolution of coordinated alternating reciprocity in repeated dyadic games. J Theor Biol. (2004) 229:549–57.
73. Guyer, M, and Rapoport, A. Information effects in two mixed‐motive games. Behav Sci. (1969) 14:467–82.
74. Hopwood, CJ. Interpersonal dynamics in personality and personality disorders. Eur J Pers. (2018) 32:499–524.
75. Overholser, J. The dependent personality and interpersonal problems. J Nerv Ment Dis. (1996) 184:8–16. doi: 10.1097/00005053-199601000-00003
76. Widiger, TA, and Presnal, JR. Pathological altruism and personality disorder. Pathological Altruism. Oxford University Press. (2012):85–93.
77. Honeycutt, JM. Imagined interaction conflict-linkage theory: Explaining the persistence and resolution of interpersonal conflict in everyday life. Imagin Cogn Pers. (2003) 23:3–26.
78. Wallenfelsz, KP, and Hample, D. The role of taking conflict personally in imagined interactions about conflict. South Commun J. (2010) 75:471–87.
79. Day, NJS, Townsend, ML, and Grenyer, BFS. Pathological narcissism: an analysis of interpersonal dysfunction within intimate relationships. Personal Ment Health. (2022) 16:204–16. doi: 10.1002/pmh.1532
80. Loas, G, Cormier, J, and Perez-Diaz, F. Dependent personality disorder and physical abuse. Psychiatry Res. (2011) 185:167–70. doi: 10.1016/j.psychres.2009.06.011
81. Crego, C, Oltmanns, JR, and Widiger, TA. Obtaining and losing the bipolarity of the five-factor model through factor analysis. Personal Disord Theory Res Treat. (2020) 11:119–30. doi: 10.1037/per0000378
82. Crego, C, and Widiger, TA. Core traits of psychopathy. Personal Disord Theory Res Treat. (2022) 13:674–84. doi: 10.1037/per0000550
83. Clark, LA, and Watson, D. The trait model of the DSM–5 alternative model of personality disorder (AMPD): a structural review. Personal Disord Theory Res Treat. (2022) 13:328–36. doi: 10.1037/per0000568
84. Widiger, TA, and Hines, A. The diagnostic and statistical manual of mental disorders, fifth edition alternative model of personality disorder. Personal Disord Theory Res Treat. (2022) 13:347–55. doi: 10.1037/per0000524
85. Gallo, PS Jr, and McClintock, CG. Cooperative and competitive behavior in mixed-motive games. J Conflict Resol. (1965) 9:68–78.
86. Sally, D. Conversation and cooperation in social dilemmas: A meta-analysis of experiments from 1958 to 1992. Ration Soc. (1995) 7:58–92.
87. Balliet, D. Communication and cooperation in social dilemmas: A meta-analytic review. J Conflict Resol. (2010) 54:39–57.
88. Pulay, AJ, Stinson, FS, Dawson, DA, Goldstein, RB, Chou, SP, Huang, B, et al. Prevalence, correlates, disability, and comorbidity of DSM-IV schizotypal personality disorder: results from the wave 2 national epidemiologic survey on alcohol and related conditions. Prim Care Companion J Clin Psychiatry. (2009) 11:53–67.
89. Kubarych, TS, Aggen, SH, Kendler, KS, Torgersen, S, Reichborn‐Kjennerud, T, and Neale, MC. Measurement non‐invariance of DSM‐IV narcissistic personality disorder criteria across age and sex in a population‐based sample of Norwegian twins. Int J Methods Psychiatr Res. (2010) 19:156–66.
Keywords: personality disorders, narcissistic personality disorder, dependent personality disorder, game theory, interpersonal
Citation: Reed LI, Cooke L, Kasaba I, Harrison E and Hooley JM (2024) Narcissistic and dependent traits and behavior in four archetypal 2-person, 2-choice games. Front. Psychiatry. 14:1275403. doi: 10.3389/fpsyt.2023.1275403
Edited by:
Massimiliano Beghi, Azienda Unità Sanitaria Locale (AUSL) della Romagna, ItalyReviewed by:
Ilaria Maria Antonietta Benzi, University of Pavia, ItalyMaria Paola Rapagnani, Local Health Unit (AUSL) of Romagna, Italy
Copyright © 2024 Reed, Cooke, Kasaba, Harrison and Hooley. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Lawrence Ian Reed, bHIxMTNAbnl1LmVkdQ==