Abstract
The present study was a direct test of the cheater–hawk hypothesis which argues that psychopathy is related to two potentially adaptive interpersonal strategies: cheating and aggression. As expected, the measures of cheater and hawk behaviors comprised a single factor, according to a maximum-likelihood factor analysis. As hypothesized, psychopathic traits exhibited large positive correlations with measures of both cheater (entitlement, exploitiveness, and short-term mating orientation) and hawk (vengeance and aggression) behaviors. Furthermore, psychopathic traits were associated with the tendency of using individualistic and competitive tactics in an altruism game and being less likely to act in a prosocial manner. Finally, scores on the combined–cheater hawk variable were significantly correlated with psychopathic traits. As hypothesized, individuals scoring high on Factor 1 of psychopathy were more likely to utilize behaviors and strategies associated with the cheater–hawk designation, whether or not they scored high on Factor 2 of psychopathy. In general, the findings support the conceptualization that psychopathy represents a fast life-history strategy characterized by seeking personal gain over others, including exploitiveness (cheater), aggression (hawk), and risk taking. Results also indicate that cheater and hawk behaviors are part of a single strategy more often employed by those higher on psychopathic traits. Implications for treatment success are discussed.
Psychopathy is often described as a personality disorder encompassing affective, interpersonal, and behavioral features (Hare, 2003). However, from an evolutionary perspective, psychopathy can instead be described as an adaptive competitive strategy that facilitates reproductive success. It should be noted that approaching psychopathy from an evolutionary standpoint does not suggest that the behaviors are currently adaptive, only that they conferred some fitness benefit in our evolutionary history (Buss, 2004). It also does not preclude negative impacts on self and/or others. However, what society might label as a disorder, or antisocial, due to its societal costs may in fact be of some benefit to the individuals who are engaging in that behavior. For example, adolescent and young male risk taking have been traditionally labeled as maladaptive behavior, but there are evolutionary reasons why such socially maladaptive behaviors may in fact be individually adaptive (Daly & Wilson, 1988; Ellis et al., 2012). This may be the case for psychopathy, which has undeniably large costs for society but may potentially be beneficial for its practitioners (Babiak & Hare, 2006).
Research does, in fact, suggest that psychopaths tend to utilize a fast life-history strategy that may be individually adaptive (Mealey, 1995). This strategy is characterized by using a combination of short-term mating tactics and opportunistic and exploitive interpersonal interactions to bolster reproductive fitness. With regard to mating, psychopathic traits have been related to promiscuous sexual behavior characterized by an uncommitted short-term mating style, engagement in multiple short-term marital relationships, and having sexual relationships at an earlier age, with a greater number of partners over the life span (Halpern, Campbell, Agnew, Thompson, & Udry, 2002; Hare, 2003; Harris, Rice, Hilton, Lalumière, & Quinsey, 2007; Jonason, Koenig, & Tost, 2010; Jonason, Li, Webster, & Schmitt, 2009; Lalumière & Quinsey, 1996). There is even empirical evidence that short-term mating orientation (STMO) affords psychopathic offenders some reproductive benefit, in that they have as many or even more offspring than non-psychopathic controls (Harris et al., 2007). Psychopath’s evolutionary fitness may be further increased by an adaptive tendency to avoid victimizing one’s genetic relatives (Krupp, Sewall, Lalumière, Sheriff, & Harris, 2012).
Furthermore, psychopathic traits appear to be related to the use of opportunistic, aggressive, and exploitive tactics in their interactions with others. More specifically, psychopathy is related to two potentially adaptive interpersonal strategies: cheating and aggression. This is known as the cheater–hawk hypothesis of psychopathy (Book & Quinsey, 2004). Psychopaths are therein described as engaging in both manipulation and exploitation of others (cheaters) as well as using violence or aggression to achieve their interpersonal and mating goals (hawks). The purpose of the current study was to further validate the cheater–hawk hypothesis across a larger number of indicators (including a behavioral measure of prosociality) using a non-forensic sample. Successful validation could have implications in the development of treatment programs for psychopathic individuals. Wong and Hare (2005) have suggested that a successful treatment program would capitalize on the strengths of the individual and would not focus on empathy training, as previous attempts to treat psychopaths in this manner have been unsuccessful and have even increased their chances of violent recidivism (Harris, Rice, & Cormier, 1991). The cheater–hawk hypothesis, should it be supported, would suggest that Wong and Hare are looking in the right direction, and that successful treatment would involve convincing the psychopathic individual of the benefits (to themselves) of prosocial methods of achieving their goals, rather than focusing on empathy and remorse, which they are clearly lacking (Cleckley, 1941).
Psychopathy
Psychopathy has been conceptualized as a personality disorder that is comprised of interpersonal, affective, and antisocial traits and behaviors (Hare, 2003). The most common tool in assessing psychopathy is the Psychopathy Checklist–Revised (PCL-R; Hare, 2003). Factor analyses of the PCL-R have consistently demonstrated two separate but related factors: Factor 1 includes interpersonal and affective traits (i.e., lack of empathy and remorse, superficial charm, and the tendency to exploit and manipulate others), whereas Factor 2 consists of behavioral traits (antisocial behavior, impulsivity, and lack of responsibility). A four-factor solution has also been found, essentially subdividing Factors 1 and 2 into Callous Affect and Interpersonal Manipulation on one hand, and Erratic Lifestyle and Antisocial Behavior on the other (Hare & Neumann, 2008).
Psychopathy has generally been viewed as a disorder (reviewed in Mealey, 1995). Psychiatrist Herve Cleckley’s (1941) seminal book on psychopathy, The Mask of Sanity, framed psychopathy as a seemingly untreatable psychiatric disorder. More recently, however, numerous researchers have suggested that psychopathy can be seen as adaptive through the lens of evolutionary psychology (Book & Quinsey, 2004; Glenn, Kurzban, & Raine, 2011; Hare, 2001; Mealey, 1995). Krupp, Sewall, Lalumière, Sheriff, and Harris (2013) have suggested that there are certain characteristics of psychopathy that call into question the notion that psychopathy is in fact a maladaptive disorder. These researchers suggest that psychopathy is not associated with the dysfunction, reproductive disadvantage, or the intellectual or developmental instability indicative of a psychological disorder (Krupp et al., 2013; Lalumière, Harris, & Rice, 2001). In addition, researchers have referred to psychopaths as “social predators” (Hare, 2001) who exploit the cooperation of others for personal gain (Mealey, 1995). From this perspective, psychopathy is seen as an alternate life-history strategy that confers some reproductive benefits rather than as a purely maladaptive disorder (Harris, Rice, & Quinsey, 1994; Mealey, 1995; Seto, Khattar, Lalumière, & Quinsey, 1997).
Further evidence for the life-history approach to psychopathy comes from research on the Dark Triad. The Dark Triad is the combined study of the three socially averse personality traits of psychopathy, Machiavellianism, and narcissism (Paulhus & Williams, 2002). Machiavellians are similar to psychopaths in their manipulative nature and their lack of concern for others, though they do not possess the impulsivity and irresponsibility characteristics associated with Factor 2 of psychopathy (Jones & Paulhus, 2011). Narcissists set themselves apart from the other Dark Triad personalities with their internal focus or self-absorption, with the goal of feeding their own egos rather than on external gain (Jones & Paulhus, 2013). In terms of the life-history approach, psychopathy, narcissism, and Machiavellianism have all been found to be associated with the tendency to engage in exploitative short-term mating strategies (Book, Visser, & Volk, 2015; Jonason et al., 2009).
Furthermore, within the Dark Triad, psychopathy has consistently demonstrated stronger relations to a fast life-history strategy than has either narcissism or Machiavellianism. For example, psychopathy is found to be associated with the facilitation of multiple, simultaneous, short-term relationships (Jonason & Kavanagh, 2010; Jones & Weiser, 2014). Jonason and colleagues (2010) further reported that psychopathy but not Machiavellianism or narcissism was associated with short-term strategizing and short-term mating efforts. Visser, DeBow, Pozzebon, Bogaert, and Book (2015) found that psychopathy was associated with fantasizing about unromantic sex with uncommitted and anonymous partners, and that at higher levels of psychopathy, people were more likely to report that they had engaged in behavior consistent with their fantasies. This research suggests that psychopathy is characterized by a preference for unrestricted sexual activity as opposed to an inability to maintain long-term relationships. Furthermore, Jonason and Schmitt (2012) found that men high in psychopathy tended to select similarly volatile friends who could facilitate their short-term mating efforts while acting as protective barriers to those who would punish them for implementing cheating strategies. Finally, Krupp et al. (2012) found evidence for an evolutionary genetic bias wherein psychopathic individuals were less likely to exploit individuals if they were genetically related to the psychopath.
Taken together, this body of research suggests that the formulation of psychopathy as an adaptive life-history strategy is plausible. Two evolutionary “arms races” have been described that may be directly relevant to the behaviors observed in people with psychopathic traits: (a) cheaters versus cooperators and (b) doves versus hawks (e.g., Dawkins, 1976; Westneat, 2010).
Most of the previous research on psychopathy has treated the construct as a taxon, grouping participants into psychopath/non-psychopath. While early research suggested that psychopathy was a discrete class (e.g., Harris et al., 1994; Vasey, Kotov, Frick, & Loney, 2005), more recent studies and meta-analyses have found that psychopathic traits are continuous in nature (e.g., Edens, Marcus, Lilienfeld, & Poythress, 2006; Guay, Ruscio, Knight, & Hare, 2007; Marcus, John, & Edens, 2004). It should be noted that early studies finding a taxon typically focused on Factor 2 (antisocial behavior). Recent research suggests that psychopathic traits are dimensional when considered in their entirety (e.g., Hare & Neumann, 2008), and this is particularly likely to be the case for the personality traits associated with psychopathy. Hare and Neumann (2008) state that psychopathy should be viewed as a continuum, which encourages research on non-forensic populations (e.g., 97% of the variance in self-reported psychopathy scores can be explained by a common measure of personality; Book et al., 2015).
Cheaters
From an evolutionary perspective, psychopaths have been seen as cheaters exploiting the cooperation of others (e.g., Book & Quinsey, 2004; Mealey, 1995). Evolutionary theory supports cheating as a potentially adaptive strategy, but its success depends on the proportion of cheaters in a given population (vs. cooperators; Dawkins, 1976). For a behavior to be considered adaptive, however, it must have “plausibly conferred a fitness advantage in ancestral environments” (Book & Quinsey, 2004, p. 34). Cheating (or exploiting the cooperation of others) has obvious biological benefits, in that the cheater gains resources without the effort or expense of compromise/cooperation (Dawkins, 1976; Westneat, 2010). Early descriptions of psychopaths highlight their exploitive nature (e.g., Cleckley, 1941), and Hare (1993) has written that “psychopaths view any social exchange as a ‘feeding opportunity,’ a contest or a test of wills in which there can be only one winner. Their motives are to manipulate and take, ruthlessly, and without remorse” (p. 145). Empirical research supports these labels, as psychopaths are in fact likely to exploit and manipulate others (Jonason & Webster, 2012; Mealey, 1995) and use deception across numerous social contexts (Jonason, Lyons, Baughman, & Vernon, 2014; Seto et al., 1997). Thus, we are confident in labeling psychopaths as cheaters, but are they also hawks?
Hawks
As well as being deceptive and manipulative, psychopaths tend to be impulsive and aggressive (Cima & Raine, 2009; Hare, 2003), which is not explained by a purely cheating strategy. Dawkins (1976) described another evolutionary arms race that pits hawks against doves. The hawk–dove scenario describes two different strategies for winning a resource. Hawks are persistently aggressive and will attack until an encounter is won, whereas doves will posture aggressively but retreat when attacked. If pitted against one another, the hawk will consistently win the contest by forcing the dove to concede. However, the hawk may incur a significant cost of injury should it face another hawk who also refuses to back down from confrontation. Thus, a hawk can benefit if it encounters numerous doves, but it will likely suffer if it encounters numerous hawks. This makes the success of one or the other strategy dependent upon the frequency of other strategies in the population, leading to general evolutionary instability in determining which strategy is best (Dawkins, 1976). A hawk strategy could therefore explain the potentially violent nature of psychopaths but not their use of deception and manipulation.
Cheater–Hawk Hypothesis
Separately then, neither the cheater nor hawk hypothesis fully explains both Factor 1 (e.g., manipulation, callous disregard of others) and Factor 2 (e.g., impulsive, aggressive tendencies) of psychopathy. However, a combination of cheater and hawk might explain both sets of traits, leading Book and Quinsey (2004) to propose the cheater–hawk hypothesis. Their results showed that psychopathy was related to measures of indignation and aggression, and that psychopaths showed a lower ratio of behavioral inhibition to activation than non-psychopaths. However, contrary to prediction, psychopaths were not less likely to engage in empathy and altruism. The cheater–hawk hypothesis was also tested by Coyne and Thomas (2008), who found that psychopathic traits were related to indirect and direct aggression, and cheating. In the latter study, support was found for the cheater–hawk hypothesis for primary but not secondary psychopathy (Coyne & Thomas, 2008). Primary psychopathy is roughly defined by Factor 1 traits (i.e., manipulativeness, lack of empathy and remorse, general callousness, and superficial charm), whereas secondary psychopathy encompasses Factor 2 traits (i.e., impulsivity, irresponsibility, and antisocial behavior; Hare, 2003).
Purpose
The purpose of the current study was to therefore further test the cheater–hawk hypothesis using a larger number of indicators than used in previous studies. We expected to find that cheater and hawk behaviors are complementary aspects of a single strategy, and that those behaviors are employed more often by those who score higher on psychopathic traits. In other words, for the cheater approach, people with psychopathic traits should be more exploitive and entitled, and display less altruism than other individuals (i.e., cheaters). The exploitation should also be evident in a STMO, which would support the idea of a fast life-history strategy (Jonason et al., 2010; Jonason et al., 2009). For the implementation of the hawk approach, psychopathic traits should be related to aggression, the tendency to be vengeful, and a propensity to engage in risk taking (specifically related to status). We should note that the tendency to be vengeful was previously housed in the cheater realm (Book & Quinsey, 2004), but given the aggressive nature of the items, we decided that it was more appropriately housed in the hawk realm.
Given that we are suggesting that the cheater–hawk approach is an evolutionary strategy, we also expected that it would partially explain the tendency of people with psychopathic traits to engage in short-term mating.
As well, given the tendency to focus solely on male participants when studying psychopathy, there is a paucity of research on female participants (research reviewed in Hare, 2003). We, therefore, decided to include men and women in our sample, increasing the generalizability of the findings.
Method
Participants and Procedure
Five hundred seventeen undergraduate students (312 women, 205 men) from three Canadian universities participated in the present study. Participants ranged in age from 18 to 45 years (M = 19.83, SD = 3.29). Data collection was conducted online (via Qualtrics), and course credit was received by participants in exchange for participation.
Materials
Psychopathic traits
The 64-item Self-Report Psychopathy (SRP-III; Paulhus, Hemphill, & Hare, 2015) scale was used to assess psychopathic traits. The SRP-III includes four subscales (Interpersonal Manipulation, Callous Affect, Erratic Lifestyle, and Antisocial Behavior), and also yields two factor scores: Factor 1 combines Interpersonal Manipulation (e.g., “I purposely flatter people to get them on my side”) and Callous Affect (e.g., “I never feel guilty over hurting others”), and Factor 2 combines Erratic Lifestyle (e.g., “I enjoy doing wild things”) and Antisocial Behavior (e.g., “Every now and then, I carry a weapon, knife, or gun”) and an overall psychopathy score. Participants responded to items on a 5-point scale from 1 (strongly disagree) to 5 (strongly agree). Cronbach’s alphas in the present study indicated high levels of internal consistency (α = .86, .82, .82, and .80, for Interpersonal Manipulation, Callous Affect, Erratic Lifestyle, and Antisocial Behavior, respectively).
Cheater variables
Exploitation and entitlement
Two subscales from the Narcissistic Personality Inventory (NPI; Raskin & Hall, 1979) were included in this study: Exploitation (five items) and Entitlement (seven items). Items are forced choice, with one point assigned for every exploitive response. For example, the exploitive choice in the following item would be “A. I find it easy to manipulate people,” as opposed to “B. I don’t like it when I find myself manipulating people.” Internal consistency was relatively low in the present study but adequate for research purposes (α coefficients were .60 and .65 for entitlement and exploitation, respectively).
Altruism
The tendency toward prosociality was measured using an altruism game developed by Van Lange, Otten, De Bruin, and Joireman (1997) that contains nine items. For each item, participants are given three options and asked to determine the number of points allocated to themselves and to an unknown other. The actual amounts change for each item but follow the same format. The instructions to participants are as follows:
In this task, we ask you to imagine that you have been randomly paired with another person, whom we will refer to simply as the “Other.” This other person is someone you do not know and that you will not knowingly meet in the future. Both you and the “Other” person will be making choices by circling either the letter A, B, or C. Your own choices will produce points for both yourself and the “Other” person. Likewise, the other’s choice will produce points for him or her and for you. Every point has value: The more points you receive, the better for you, and the more points the “Other” receives, the better for him or her. Here is an example of how this task works:
You get Other gets
A 500 100
B 500 500
C 550 300
In this example, if you chose A, you would receive 500 points and the other would receive 100 points; if you chose B, you would receive 500 points and the other 500; and if you chose C, you would receive 550 points and the other 300. So, you see that your choice influences both the number of points you receive and the number of points the other receives. Before you begin making choices, please keep in mind that there are no right or wrong answers—choose the option that you, for whatever reason, prefer most.
In this example, Option A is the competitive choice due to the large difference between the outcomes. Option B is the prosocial option. Finally, C is individualistic (i.e., your outcome is better than the other person). The number of competitive, individualistic, and prosocial choices determines the participant’s dominant category. That is, someone who chooses more competitively will be placed in the “competitive” category, whereas someone who chooses to be prosocial most often will be placed in the “prosocial” category.
Hawk variables
Aggression
The Reactive-Proactive Aggression Questionnaire (RPAQ; Raine et al., 2006) contains 23 items pertaining to reactive (e.g., “Reacted angrily when provoked by others”) and proactive aggression (e.g., “Had fights with others to show who was on top”; 11 and 12 items, respectively). Responses are on a 3-point scale (0 = never, 1 = sometimes, 2 = often). The RPAQ subscales and total score showed excellent internal consistency in the present study (α = .84 and .85 for reactive and proactive aggression, respectively).
Vengeance
The 20-item Vengeance Scale (Stuckless & Goranson, 1992) measures the tendency for individuals to desire retribution in response to perceived mistreatment. Responses are provided on a 7-point Likert-type scale (1 = strongly disagree; 7 = strongly agree). The scale displayed excellent internal consistency in the present study (Cronbach’s α = .90). Whereas Book and Quinsey (2004) included this scale to represent part of the cheating strategy, it does have several items relating to aggression and may be more closely related to the hawk approach. Example items include “Revenge is morally wrong” and “Anyone who provokes me deserves the punishment I give them.”
Short-Term Mating
We included 10 items measuring STMO from the Sociosexuality Scale (Jackson & Kirkpatrick, 2007). Items on this measure are answered on a scale from 1 (strongly disagree) to 5 (strongly agree). Internal consistency for this subscale was excellent (α = .94).
Results
Descriptive statistics for all study variables are detailed in Table 1. All variables met the assumption of normality, and examination of scatterplots revealed linear relationships. To test the cheater–hawk hypothesis, correlations and ANOVAs were first conducted. Furthermore, a factor analysis was conducted to determine whether the cheater and hawk behaviors are a single strategy, and we then examined the relationships with psychopathic traits.
Descriptive Statistics for Study Variables (N = 490).
Note. N = 490. SRP-III = Self-Report Psychopathy scale (Paulhus et al., 2015); STMO = Short-Term Mating Orientation of the Sociosexuality scale (Jackson & Kirkpatrick, 2007).
Bivariate correlations were conducted to determine whether psychopathic traits were related to cheater (exploitiveness, entitlement, and short-term mating orientation) and hawk (aggression, vengeance, and status-driven risk taking [SDRT]) behaviors. Correlations were all significant (p < .001) and were moderate to large in size, indicating that people who scored higher on psychopathic traits were more likely to report using cheater and hawk behaviors (see Table 2).
Relationships Between Psychopathic Traits and Study Variables (N = 490).
Note. All correlations are significant at p < .001. IPM = interpersonal manipulation; CA = callous affect; ELS = erratic lifestyle; F1 = Factor 1; F2 = Factor 2; SRP Total = total score of the Self-Report Psychopathy scale (Paulhus et al., 2015); NPI = Narcissistic Personality Inventory (Raskin & Hall, 1979); STMO = short-term mating orientation.
Because the altruism task results in a nominal variable (prosocial, individualistic, or competitive tendencies), a series of ANOVAs were conducted to examine the prediction that psychopathy (total score, factor scores, and facet scores) would be negatively related to altruism. Contrasts comparing prosocial with competitive and individualistic participants (combined) were also conducted. That is, we examined whether prosocial people scored lower on psychopathic traits than those who were not categorized as prosocial.
Homogeneity of variance can be assumed, as Levene’s tests were non-significant (ps > .01). As expected, we found that total psychopathy, both factors, and all four facets were significantly lower in those individuals categorized as prosocial compared with those who were either individualistic or competitive (see Table 3 for tests and effect sizes). Furthermore, the three groups (competitive, individualistic, and prosocial) all differed significantly from one another on Factor 1 scores in post hoc tests using Tukey’s honest significant difference (HSD; ps < .05). Factor 2 and total SRP-III scores were not significantly different between competitive and individualistic individuals. These results suggest that while people with psychopathic traits in general are less likely to be prosocial, people with higher Factor 1 scores are most likely to choose competitive responses that give them a smaller payout but maximize the difference between their payout and the “other” payout. These results are displayed visually in Figure 1.
Psychopathic Traits in Prosocial, Individualistic, and Competitive Individuals.
Note. All tests were significant at p < .001. Contrast compares prosocial responders with individualistic and competitive responders. IPM = interpersonal manipulation; CA = callous affect; ELS = erratic lifestyle; F1 = Factor 1; F2 = Factor 2; SRP total = total score of the Self-Report Psychopathy scale (Paulhus et al., 2015).

Standardized Factor 1, Factor 2, and total Self-Report Psychopathy (SRP-III; Paulhus et al., 2015) scores in each of the altruism game response categories.
Factor Analysis
To determine whether cheater and hawk behaviors represent a single strategy, we conducted a maximum-likelihood factor analysis. For the altruism variable, we used the number of times the person chose a prosocial act. Only one eigenvalue was above 1.0 (2.95), and the second eigenvalue was well below that at 0.88. Examination of the scree plot further supported one factor.
The single factor was named “Cheater–Hawk,” and loadings for each of the study variables can be found in Table 4. Loadings ranged from .42 to .77, with all variables except altruism loading in a positive direction. Factor scores were calculated and will be referred to as cheater–hawk scores. Total SRP scores and all subscale and Factor scores were highly correlated with this new variable (see Table 5), with total SRP correlating at .70 (p < .001). Given the size of these effects, one might even be tempted to equate the cheater–hawk strategy with psychopathic traits.
Factor Loadings for Cheater–Hawk Scores.
Correlations Between Cheater–Hawk Scores and Psychopathic Traits.
Note. N = 490. SRP = Self-Report Psychopathy scale. All correlations significant at p < .001.
We also predicted that the cheater–hawk strategy may partially explain the relationship between psychopathic traits and the tendency to engage in short-term mating. To examine this, we conducted two mediation analyses using Process (Hayes, 2012; for Factor 1 and Factor 2 traits, respectively). As expected, Factor 1 was a significant predictor of STMO, b = .22 (.15-.29), t = 6.23, p < .001, as was the cheater–hawk score, b = 1.87 (.54-3.19), t = 2.78, p = .005. When the cheater–hawk score was taken into account, the effect of Factor 1 dropped to .07, which was a significant reduction according to the Sobel test (z = 2.75, p = .006; R2 = .16), indicating that the relationship between Factor 1 and short-term mating is partly explained by cheater–hawk behaviors. Also as predicted, Factor 2 was a significant predictor of STMO, b = .26 (.19-.33), t = 7.02, p < .001, as was the cheater–hawk score, b = 2.06 (.85-3.27), t = 3.36, p < .001. When cheater–hawk score was taken into account, the effect of Factor 2 dropped to .007, which was a significant reduction according to the Sobel test (z = 3.28, p = .001; R2 = .15), indicating that the relationship between Factor 2 and short-term mating is partly explained by cheater–hawk behaviors.
Discussion
The cheater–hawk hypothesis was originally introduced to test an adaptive view of psychopathy (Book & Quinsey, 2004). The purpose of the present study was to replicate previous findings using a larger number of indicators and with a non-criminal population. Given previous research (e.g., Book & Quinsey, 2004), we hypothesized that psychopathic traits would be associated with both cheating (e.g., entitlement, exploitiveness, competition vs. altruism) and hawk behaviors (e.g., aggression, vengeance). Cheating behaviors were associated with all subscales and total psychopathy scores, supporting our hypothesis. Furthermore, people who chose to use individualistic/competitive tactics rather than prosocial ones in the altruism games scored significantly higher on all measures of psychopathic traits. These findings are congruent with the research of Coyne and Thomas (2008), who found that psychopathic traits were associated with cheating behaviors, and are slightly in contrast to the findings of Book and Quinsey (2004) who found that individuals higher in psychopathy did not measure any lower on indicators of empathy and altruism. Finally, psychopathy was positively correlated with STMO, which is consistent with the research that suggests that psychopaths employ a fast life-history strategy characterized by short-term mating strategies (Jonason & Kavanagh, 2010; Jonason et al., 2010; Jonason et al., 2009; Jones & Weiser, 2014). From an evolutionary perspective, this strategy could be adaptive, allowing individuals to cheat on social contracts, conventions, and expectations to facilitate sexual activity with multiple partners, thereby enhancing their potential for reproductive success (Jonason et al., 2009).
The hawk prediction was also supported, in that bivariate correlations between psychopathy and aggression, vengeance, and SDRT were all significant, positive, and large. These results support the findings of Coyne and Thomas (2008) and Book and Quinsey (2004), who found that psychopathy was associated with aggression. These results are also consistent with general findings, indicating that psychopathy is associated with aggression (Cima & Raine, 2009) and risk taking (Crysel, Crosier, & Webster, 2013; Lyons, 2015; Swogger, Walsh, Lejuez, & Kosson, 2010). From an adaptive perspective, this suggests that psychopaths are set against being exploited by others. Whether this is due to a greater degree of vigilance or vengeance remains to be tested, but our results do suggest that the latter explanation may be more likely (particularly in light of psychopaths’ possible lack of fear; for example, Fowles & Dindo, 2006; Lykken, 1995).
We also predicted that the cheater and hawk variables may represent a single adaptive strategy, and in that vein, we conducted a maximum-likelihood factor analysis. As expected, all of the cheater and hawk variables loaded highly on one factor; now coined cheater–hawk score. This score was also highly correlated with subscale, Factor, and total scores on the SRP-III (Paulhus et al., 2015).
Given that the cheater–hawk behaviors are purported to be an adaptive strategy that is often utilized by those with psychopathic traits, we thought that the tendency of people with psychopathic traits to engage in short-term mating may be partially explained by the use of this strategy. Mediation analyses for Factors 1 and 2 supported this assertion, suggesting that when cheater–hawk behaviors are accounted for, the relationship between psychopathic traits and STMO is significantly reduced.
In general, our research provides support for the cheater–hawk hypothesis put forth by Book and Quinsey (2004), which indicated that psychopathy is an evolutionary strategy that blends a cheating strategy with a hawk strategy. Furthermore, we have extended the findings to include female participants, which is rare in the psychopathy literature.
Although the results of the present study provide compelling evidence in support of the cheater–hawk hypothesis, there are some limitations that need to be addressed. First, and foremost, the sample was made up entirely of undergraduate students. Although our sample was quite large, there remains an obvious need to replicate the study with a forensic sample where psychopathic traits are more likely to be present. Using an undergraduate sample necessarily restricts the range of psychopathy scores.
A related issue is our use of a Canadian undergraduate sample which always raises concerns regarding generalizability with regard to both the general population and more specific forensic populations (see above). One strength of our sample, as mentioned above, was the fact that we included a large number of female participants. One of the current issues in the psychopathy literature is its general reliance on male samples, and our results provide some support for the generalizability to women.
Although our sample was a strength, in that we were able to establish that the cheater–hawk hypothesis was not specific to offender populations, and in our inclusion of female participants, there may very well be cultural differences that we were unable to examine. For example, Visser, Pozzebon, and Reina-Tamayo (2014) found that subclinical psychopathy showed substantial overlap with SDRT (Ashton, Lee, Pozzebon, Visser, & Worth, 2010), which is defined as a willingness to risk great physical harm when there is a status or resources benefit. The authors suggest that SDRT explains the high mortality and morbidity rates cross-culturally in young men, but suggest that SDRT would be particularly prominent in young, unmarried men in societies with great economic disparities. In such environments, cheater–hawk tendencies might have the potential to reap far greater rewards than in western societies, where presumably individuals can prosper through honest methods. Future research should seek to examine this prediction.
Second, all data (except for the altruism game) were self-reported, and therefore socially desirable responding is a potential problem. This refers to the fact that participants may be motivated to present themselves in a more positive light. It should be noted, however, that a recent meta-analysis reported that psychopathy was actually negatively related to positive impression management which may attenuate this concern (Ray et al., 2013). Regardless, future research should attempt to implement behavioral tasks (where possible) to mitigate potential self-report response biases as well as the possibility of inflated intercorrelations due to shared method variance. This is highlighted by the fact that Book and Quinsey (2004) failed to find an effect of altruism when using a self-reported measure, while the effect in the current study using a more behavioral task was quite large.
The use of self-report methods with an undergraduate sample is in keeping with substantial research that psychopathy is dimensional in nature—that is, psychopaths differ from subclinical psychopaths in a qualitative fashion rather than in a categorical fashion (e.g., Edens et al., 2006; Guay et al., 2007). Future research should attempt to replicate these results in a forensic context.
Third, a direct test of the strategies employed by individuals high in psychopathic traits utilizing an experimental paradigm would be both interesting and informative. Future studies should implement real-time games that allow players to cheat, behave altruistically, retaliate, and be forgiven to ensure that individuals high in psychopathic traits are truly following a cheater–hawk strategy as suggested by this and previous studies (Book & Quinsey, 2004; Coyne & Thomas, 2008).
Finally, although the cheater–hawk hypothesis explains some of the advantages of a psychopathic adaptation, it also suggests several serious costs to using these strategies. In general, there are definite negative consequences to employing exploitive and aggressive tactics, including loss of reputation that is described in social contract theory (Binmore, 1998) and game theory (Dawkins, 1976; Westneat, 2010; which would affect the likelihood and tenor of future interactions with others), loss of future resources (as a direct result of loss of those interactions), and even physical harm that would be associated with the use of physical aggression (Westneat, 2010). The short-term and selfish nature of psychopathy is thus likely to come with significant social and personal costs, which is in line with research finding deficits and problems associated with psychopathic traits. For example, several studies have found deficits in emotion processing, which can negatively affect social interactions (e.g., Fowles & Dindo, 2006; Lykken, 1995).
Importantly, although there may be an adaptive advantage to some of the traits and behaviors associated with psychopathy, there are obvious drawbacks to using these strategies, and they cannot be considered to be adaptive in a broader social context. Psychopaths are more resistant to offender-based treatment programs (Olver, Lewis, & Wong, 2013; Olver & Wong, 2009), are more likely to recidivate (Leistico, Salekin, DeCoster, & Rogers, 2008; Yang, Wong, & Coid, 2010), and are less likely to have success in interpersonal relationships (Ali & Chamorro-Premuzic, 2010). So though potentially advantageous from an evolutionary perspective, these strategies are not adaptive in a societal sense as they tend to cause significant harm to others while limiting the normative success of psychopaths. Overall, from the perspective of the cheater–hawk hypothesis, psychopathy represents a risky short-term strategy that relies on short-term gains to offset long-term losses. The specific gains and losses will likely depend on individual attributes (e.g., strength, intelligence) as well as more global factors (e.g., the prevalence of hawks vs. doves in a population). Further research should therefore investigate the ecological conditions that relate to the costs and benefits associated with a psychopathic cheater–hawk strategy.
These costs and benefits may have very important implications for the successful treatment of psychopathic individuals. If psychopaths are enjoying some individual benefits from their personality strategies, this will make altering those strategies more difficult, as is seen in treating addictions to rewarding substances, such as cocaine (Kirby, Marlowe, Lamb, & Platt, 1997). As mentioned in the introduction, we believe that Wong and Hare’s (2005) approach is appropriate, namely, that a successful treatment program would capitalize on the individual’s strengths and would attempt to convince the individual that they can meet their goals by prosocial means. This is further supported by the lack of success in focusing on empathy and remorse (reviewed in Wong & Hare, 2005) and by similar evolutionarily informed strategies used to reduce other antisocial behaviors (e.g., Ellis, Volk, Gonzalez, & Embry, 2016). From a personality psychology perspective, empathy training strategies that aim to alter the empathy of psychopaths may not be targeting the most salient personality cues—a tendency to cheat others (e.g., low Honesty-Humility) and respond aggressively (e.g., low Agreeableness) versus low empathy (e.g., the sentimentality aspect of Emotionality; Book et al., 2016). Thus, the cheater–hawk hypothesis may serve as a clearer theoretical guide to informing future interventions.
Footnotes
Authors’ Note
Data can be accessed (within ethical guidelines) by contacting the corresponding author.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
