Personality and Individual Differences 36 (2004) 575–586

www.elsevier.com/locate/paid

Is social dominance a sex-specific strategy for infidelity?
Vincent Egan*, Sarah Angus
Department of Psychology, Glasgow Caledonian University, Cowcaddens Road, Glasgow G42 9SF, Scotland, UK
Received 20 August 2002; received in revised form 16 January 2003; accepted 26 February 2003

Abstract
The current study investigated personality, psychopathy and mating effort of 84 adults recruited from a
large office setting who admitted infidelity whilst involved in another relationship. These were compared
with individuals who had not been unfaithful. Measurement scales were reduced by principal components
analysis to three general factors; social dominance, manipulativeness, and openness. There was no sex
difference in social dominance or openness. Males were higher on the manipulativeness factor. There were
no differences in the social dominance or openness factors for individuals admitting affairs compared to
those who had not; males who admitted affairs were higher in social dominance. There was an interaction
between sex and having had an affair (or not) for the social dominance, this indicated males who had
committed infidelity were higher on the social dominance dimension than females who were also unfaithful, the reverse was the case for males and females who had not had affairs. Manipulativeness predicted the
number of affairs had and their emphasis on sexuality, whereas social dominance did not. These results
suggest male and female infidelity is underpinned by differential personality types as well as differential
sexual strategies.
# 2003 Elsevier Ltd. All rights reserved.
Keywords: Infidelity; Affairs; Extra-pair copulation; Personality; Evolution

1. Introduction
Infidelity is the main reason cited when petitioning for a divorce (Betzig, 1989), and those who
are unfaithful tend to be cautious about discussing such activity. This may be because it involves
elements of behaviour participants may not be comfortable admitting to; unmet sexual and
emotional needs, opportunism, irresponsibility, and the execution of intentional deceit. As such,
infidelity provides a means of examining the darker and more problematic side of human
* Corresponding author. Tel.: +44-141-331-3037; fax: +44-141-331-3636.
E-mail address: v.egan@gcal.ac.uk (V. Egan).
0191-8869/03/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0191-8869(03)00116-8

Agyei. Disapproval of affairs. relationship satisfaction. 1999. Marital commitment does not change matters. Fejfar. and VanYperen (1993) tested the assumption that following AIDS awareness ‘‘individuals have become more cautious with regard to sexual relations outside their marriage’’ (p. Females typically value emotional attachment. and Clayton (2000) found that low straightforwardness (one facet of the A construct within Costa and McCrae’s ‘‘Big Five’’ model. Leukefeld. it is rarely socially acceptable. 2000). and their sexual relationships are no exception. Meta-analysis of 45 studies examining personality factors underlying sexual risk behaviour found high agreeableness (A) and high conscientiousness (C) reliably correlated with lower sexual risk taking (Hoyle. 2000). Ellis & Symonds. This may reflect an evolutionary need for females to be generally sure of a male’s personal commitment and his willingness to share resources (as well as genes) before having sex with him. not everybody necessarily follows the evolutionary script (Buss. as were desired and actual extramarital affairs. 1997). and these are more likely to involve strangers. have a greater desire for anonymous sexual encounters. Sexual fantasies reiterate this sexual disparity. 1995). examining susceptibility to infidelity within that first year. A study of normative disapproval to extramarital affairs by Prins. and double-standards exist between public expression and private practice (Feldman & Cauffman. deceitful behaviour) was also a strong predictor of sexual promiscuity. This study did not indicate that low C is related to sexual promiscuity. and how likely the participants were to commit infidelity. Sheppard. Some aspects of infidelity reflect sex differences in attitudes to sex itself. Gaulin. & Gladue 1994. men have more sexually focussed fantasies than women. Miller. Women’s sexual fantasies are more likely to have some personal content (e.g. such as sexual satisfaction.576 V. S. a separate measure of narcissism and contextual variables predictive of susceptibility to sexual infidelity. 2000). although sensation seeking and impulsivity (sharing constructs with low C) are strongly predictive of sexual risk taking. including multiple partners. for low C the strongest correlation was with unprotected sex. While infidelity is universal. Nevertheless. Angus / Personality and Individual Differences 36 (2004) 575–586 experience without having to examine actual criminal offending (Spitzberg & Cupach. 1998). While neither normative disapproval nor actual affair frequency differed between the sexes. thus males typically place less emphasis on emotional attachment for sexual relationships. Egan. Low A was correlated negatively with greater sexual risk taking. 1990). Each participant reported . was not correlated with desire and actual behaviour. General disapproval for infidelity and disapproval of infidelity due to risk of HIV was measured. this theory is powerful and the current study seeks to examine the individual differences in personality and mating effort in people who admit to having actually been unfaithful using an evolutionary model. Buunk. Wright & Reise. knowledge of their fantasy partner. characterised by manipulative. 1994. Infidelity and the desire for more than one partner are clearly important factors of sexual risk taking and strongly underpinned by individual differences. the desire to have an affair was significantly higher for males. Zimmerman. 1998. and desire a greater number of sexual partners than females (Bailey. Buss and Shackelford (1997) studied 107 couples married for less than a year. and sources of conflict. While infidelity affects deep human emotions and bonds driven by evolutionary factors. Shifts from this stereotype reflect research with adolescents and young adults rather than the general population (Feldman & Cauffman. Logan. Lynam. ‘‘Normative disapproval’’ showed significant negative correlations for both sexes on desire and behaviour. & Miller. Nelson. 39). They examined both Eysenckian and ‘‘Big Five’’ dimensions of personality. even in the light of AIDS awareness. & Andreoli-Mathie.

In keeping with this thesis. self-reported ‘‘psychopathic’’ traits. Lynam. better or more diverse genes. Openness (O) was a predictor of susceptibility to infidelity in men. This overlap is because the means employed by males in attracting a sexual partner involve competition. 1997. 2000).. toughminded hostility) was significantly and positively correlated with infidelity for both men and women. Miller. Vazsonyi. Given the differential sexual motivation of males relative to females. 107). & Figueredo. Traits such as high narcissism [akin to Machiavellanism (M) and primary psychopathy]. they may also be more delinquent and inclined to social failure (Rowe. these were better predictors in females. who also completed the scales. for some. and high N. 1995. 1995. the actual behaviour must occur (Buss & Schmitt. . We expected mating effort to be greater in individuals who were unfaithful. Rowe et al. Most studies have examined the desire to commit infidelity rather than actual acts. whereas secondary psychopathy involved low C. although the sexual strategies theory suggests that for these strategies to become internalised via a result of sexual selection and adaptation. 1993). an option to mate-switch and increased mate manipulation (Buss. M has a positive correlation with primary and secondary psychopathy (McHoskey. and delinquency (in an adolescent) or criminal behaviour (in an adult) may be. she found that the preference for physical attractiveness in females was specific to the sexual context. efforts to attract a sexual partner) is higher in males (Rowe. Angus / Personality and Individual Differences 36 (2004) 575–586 577 on themselves and their partner. female short-term sexual strategies are more complex. We also expected to replicate classic findings such as infidelity to be greater in males and sex-related differences in motivation for unfaithfulness. Secondary psychopathic characteristics such as a lack of long-term goals and an impulsive lifestyle are less obviously relevant. an effective (or only) tool for acquiring ‘‘resources with which to attract a mate’’ (Rowe et al. The aim of the present study was to examine the personality of self-admitted adulterers in relation to the five-factor model of personality. a lack of empathy.e. 1997). While male short-term sexual strategies are perhaps driven primarily by hedonic factors. 1997). low A. reasoning that physical health underlies attractiveness. & Szyarto. A person high in mating effort is more likely to commit infidelity and be sexually promiscuous. Widiger. Relating the five-factor model of personality to the psychopathy construct. Egan. it is unsurprising that mating effort (i. p. Worzel. 1997). women preferring men with good character as colleagues. The core characteristics of primary psychopathy (lying and manipulativeness. as they may bestow a range of benefits including greater resources. While low C and high Narcissism generally predicted the desire to commit infidelity. 1998). reasoning that these traits all overlap. S. and sexual selection strategies reflecting mating effort. Wright & Reise. and grandiosity) might be thought helpful attributes for infidelity. Psychoticism (P. 1999.V. Greiling & Buss. Scheib (2001) examined the particular thesis that relationship infidelity may enable a female to obtain ‘‘good genes’’. and Leukefeld (2001) found primary psychopathy reflected a combination of low A and low neuroticism (N). hostile and manipulative factors to be greater in those who commit adultery. and this may have been more important for the offspring than the milieu provided by a male with good character. This design enabled each person to give a self-rating and to have an intimate peer rating. low A and high P might thus apply equally to individuals with antisocial personality and those who are sociosexual (Mealey.. and perhaps as longterm partners.

3. To assure participants of confidentiality. The sample was an opportunistic broad screening of workers in a large nonacademic office. 23 (27. Of the 84 participants.3%) of the sample were single. and whether they were male or female.5%) female. Participants Of 120 questionnaires issued 87 (72. and thus ethical practice was paramount. We explicitly did not confine this question to the current relationship.5%) were male and 55 (65.4%) married. sexual preference and occupation. this reduced the problem of highly skewed data attributable to several very extreme individuals. it was highly likely that they may have been in the past. Design Participants completed a brief screening questionnaire and four psychological scales. Measures 1. 2. Egan. 29 (34. 2. Method 2.578 V.5%) were returned. Occupationally. Angus / Personality and Individual Differences 36 (2004) 575–586 2. the study was passed by the departmental ethics committee. No participants were unemployed. and 7 (8. individuals were classified by sex and whether they had been unfaithful or not.2. S. with instructions to place the completed scales in an envelope and thence into a sealed collecting unit. which was not opened until the study had been completed. sex. Persons who had done this more than six times were given the value of 7 (two participants). an additional envelope was provided with the questionnaires. 2. Ethics The current study examined some very personal matters.4.1%) were full-time employed in the office while 15 (17. marital status. To examine whether the infidelity was driven by sexual or . Brief screening instrument and infidelity index Participants completed a covering sheet which assessed age. Two participants had homosexual sexual preferences.1. Three questionnaires were excluded from analysis due to incomplete data.9%) were students working over the Summer. 69 (82. This measure also asked about infidelity.3%) divorced. The dependent variables were the measures completed by each participant. 54 (64. While inexact. Information was also collected to examine whether the motivation for the affair in those who admitted to such behaviour was driven by emotional or sexual factors. reasoning that while participants may not currently be in a relationship. The mean age of participants was 30 (range=17–53). Based on information obtained in the screening instrument. The independent variables were thus whether the person had had an affair or not. asking whether the participant had engaged in sexual relations with someone other than their regular long-term partner. Participants stated the estimated number of relationships in which they had committed infidelity. subject to close adherence to the principles of confidentiality and anonymity.

manipulative and egocentric behaviour). these difficulties can be partially overcome statistically. the scales scored and data coded. & Austin. The NEO-FFI is a short form of the more lengthy NEO-PI and assesses the ‘‘Big Five’’ personality dimensions of N. The NEO-FFI has limitations. Statistics Initial analysis involved an examination of the raw scores on the instruments broken down by age and sex. while a further 10 items test secondary psychopathy (i. Smith. The Mating Effort Scale (MES. 1991) for non-offending populations. Rowe et al. 4. Responses are given on a five-point scale from strongly disagree to strongly agree. The NEO-FFI (Costa & McCrae. Individuals were asked to complete them in their own time and to return them to the sealed unit. For many cases affairs could be driven by both factors. 5. A and C. & Cory. extraversion (E). and provides a self-report alternative for the PCL-R (Hare. five primarily emotional. Levenson. Levenson et al. O. and consists of 13 items and responses are limited to true or false. 2. 3. Procedure Questionnaires were given out with a consent form and an information sheet to the large workplace. S. The short versions of the MCSD correlate highly with the original scale (Loo & Thorpe.. the core psychopathic traits of callous. The MES is a reliable 10-item scale used to examine intra-sexual competition in both males and females. A parametric analysis was selected after exploratory analysis of the data found no . Egan. then entered into SPSS. in which a response of one would indicate the affair was primarily sexually driven. A and C are more reliable than the O and E dimensions (Egan. 2000). (1995) argue that if psychopathy is a consistent construct then the cluster of behaviours and characteristics associated with it should not be confined to the criminal population. However. The concordance of the LSRP and the PCL-R is generally fair (Brinkley. and the scale was included to investigate this possibility.V. Angus / Personality and Individual Differences 36 (2004) 575–586 579 emotional factors a single five-point Likert scale was used. lifestyle-driven anti-social behaviour). Response bias is a limitation in most studies that use self-report data. and using a widely used instrument enables some extrapolation of results to other studies using similar measures. Schmitt. Reynolds. 2001). Sixteen of the items test primary psychopathy (i. & Newman. 3. as has been the case in this research area. The Self Report Psychopathy Scale (LSRP. and a response of three would signify this possibility.5.e. 1997). After several months the box was removed from the work setting.e. 1982) The MCSD tests for socially desirable response bias. 2000). Deary. 1992) The NEO-FFI was used to assess general personality traits in the participants. and the dimensions of N. in particular the five personality dimensions it indexes are not orthogonal. Kiehl. 2. 1995) The LSRP measured psychopathic (or at least Machiavellian) attributes with a 26-item inventory designed to examine characteristics associated with primary and secondary psychopathy in the general population. The Marlowe–Crowne Social Desirability Scale (MCSD.

4. S. Of these. again reiterating what might be expected from the literature. 12 male) reported having had an affair compared to 50 (33 female. When the analysis was confined to the 34 participants who had been unfaithful. and higher on primary psychopathy and the MES. and an interaction between these factors would indicate differential sexual strategies between the sexes. The most parsimonious way of analysis was deemed to be a two-way ANOVA in which the sex of the participant and whether or not they had had an affair were the independent variables. and stem and leaf analysis found few genuine outliers. Angus / Personality and Individual Differences 36 (2004) 575–586 highly skewed distributions. 34 (40. 4. emotional/sexual infidelity and the measures of behaviour and personality were correlated with one another using Pearson’s r. in the current data set.4%) individuals (22 female. Perhaps of most interest is the interaction between sex and having had an affair or not. These results show that males are higher than females on measures of self-rated primary and secondary psychopathy and mating effort. The resulting correlation matrix is shown in .. This conservative analysis was conducted to avoid getting significant effects via the same predictors due to the indirect effects of prior correlated factors. Males were lower than females on measures of A and social desirability. This was done using principal components analysis with Promax rotation (which does not assume the independence of derived factors). This result shows that. 4. There was no difference in the frequency of males and females reporting affairs compared with control subjects (w 2=0. n.). single people were more likely to answer no. and married people were approximately equal in their responses. Results Eighty-four participants completed the questionnaire package. Relationships within the measures The rate of infidelity. Given the NEO-FFI does not represent five orthogonal personality dimensions (Egan et al. Egan. These results are not unexpected given what is already known about sex differences in personality. from which factor scores for each of the dimensions were then calculated.05. Interactions were observed between these variables and N. the correlation matrix was reduced to a simpler set of general personality factors.s. 17 male) who had not.2. These factor scores were then analysed using the two-way ANCOVA model used to analyse the raw scores on the scales. males were not more likely to commit infidelity than females. E and A indicating sex-differential effects on personality.1. males showed higher rates of infidelity than females. Divorced people were more likely to answer yes to sexual infidelity. 2000). To explore simple relationships between the scales (some of which have not been examined in relation to each other before) Pearson’s r correlations were calculated. Sex and affair differences on the basic measures To examine the combined effects of sex differences and whether the participant had had an affair before. A comparison based on whether individuals had previously had an affair or not indicated that those who had committed infidelity were lower in A and social desirability. we ran a series of two-way ANOVAs in which we examined individual variables by sex (male or female) and affair (no or yes).580 V. Table 1 presents these results expressed as F-ratios and their associated significance.

38** 0. this measure does not measure five independent dimensions of personality (Egan et al. Angus / Personality and Individual Differences 36 (2004) 575–586 Table 1 F-ratios for two-way ANOVA dividing the sample by sex and whether they have had an affair or not Sex Affair P< DF (83.40 0.s..025 n.02 0.s.24** 0..59** 0.s.s.06 0. 0.19 0. n.02 0.92 0.00 O 0.s.s.04 0.25** 0.s.001 n. more highly (and negatively) with A.00 0.07 0.49 1.009 n.09 1.00 MES 0.001 0.17 1. as previously observed.40** 0.23 1.19 3.23** 0. lower A and higher scores on the MES. In the current study N.75 5.71 1.59.76 0. 0.23 0. Associated with primary and. for which there was no association at all.001 0. P<0.27** 0.11 0. n. 0. indicating that.04 1.31** 1. MCSD 0. n. 2000). n. S.s.31** 0. Primary psychopathy was correlated moderately with secondary psychopathy. 0.23 0.s. 0.61 0. 1) 4. supporting views that social desirability measures reflect personality variance as much as dissimulation.25** 0.57 5.20 . in keeping with what is known about the personality features of formally recognised offenders.31 4. 0. Table 2.02 0.s.24** 0.18 0.03 0. 1) Primary psychopathy Secondary psychopathy Neuroticism Extraversion Openness Agreeableness Conscientiousness Mating effort Social desirability Sexaffair P< (83.23** 1. E and C all overlapped.18 0.s. only correlations significant at ** (P< 0.32** 1.28** 0.04 0.s.00 To avoid noting spurious associations. n.41** 0.01 0.03 2. n.00 E A 0. n.s. Table 2 Correlation matrix of measures (n=84) Rate LSRP-P LSRP-S N E A O C MES LSRP -P LSRP -S 0.00 0.19 0. 1) 45.00 0.13 0. Frequency of affairs was positively correlated with primary psychopathy.05 P< (83.001 n. and substantially (r=0.90 34.21** 0.11 7.11 8.s.01 or higher) are noted.s.96 15.06 1. Egan. Secondary psychopathy (with a greater emphasis on antisocial activity rather than interpersonal manipulativeness) was correlated with higher N and lower scores on all personality measures bar O.24** 0.00 C 0. Within the NEO-FFI 6 of the 10 intercorrelations were significant.38 11.45** 0.06 12. and with lower self-rated N and psychopathy.43** 0.00 N 0.02 0.44** 0.05 n.09 0.35** 0.001) with the MES. Social desirability was associated with greater A and C. the MES was also associated with lower A.66 0. to a lesser extent secondary psychopathy.s.00 n. non-significant. n. n.03 0.001 0.581 V.004 0.33** 1.23** 0.27** 1. all supporting the broad hypothesis that higher negative personal traits will be related to greater frequency of infidelity.70 4.20 0.04 n.80 0.

91 0. Classification of data The systematic associations between the different measures in the study suggest that there are general underlying variables that account for much of the observed variance.17 3. S.60 1. respectively. Sex and affair differences on the derived factors The Promax rotation assumes the factors are correlated. 4.69 0.28 0.5 were used to define the derived dimensions. C and social desirability.10 0. The third factor had high positive loadings for E and O.00 0.79 0. These correlations. To optimise conservatism of factor interpretation. This was labelled social dominance.77 0. and required six iterations to converge (Table 3).31 0. and was labelled manipulativeness. A.68 0. Factor 1 had high negative loadings for secondary psychopathy and N.60 . and this was confirmed during the analysis.68 18. only loadings of over 0.11 0.40 1. as would varimax rotation.24 0.00 0.63 40. an interaction was found Table 3 Principal components analysis (with Promax rotation) of measures for all participants (n=84) Rotated factors (six iterations) F1 ‘social dominance’ Primary psychopathy Secondary psychopathy Neuroticism Extraversion Openness Agreeableness Conscientiousness Mating effort Social desirability Eigenvalue % Variance F2 ‘manipulativeness’ F3 ‘outgoingness’ 0.20 0.582 V.00 0. The oblique Promax rotation was used to accommodate the correlated nature of the underlying variables. this method was legitimate as there were nine times as many subjects as there were variables put into the analysis. and a high negative loading on A.16 0.55 0. To test this a principal factor analysis with Promax rotation was conducted on the four scales used in the study.3. indicate that social dominance is associated with outgoingness and negatively associated with manipulativeness. Even after correction. and high loadings for E. though not huge.48 0.3.18 0. and reflected outgoingness. Using the smaller set of dimensions we tested for interactions between sex and whether the person had committed infidelity or not using a two-way ANCOVA (Table 4).40 0.30 0. correcting each factor for the influence of the other two factors . optimising the likely stability of the solution.90 0.74 0.34.28 and 0.45 0.14 12. Factor 2 and factor 3 were uncorrelated to one another.84 0. with factor 1 correlating with factors 2 and 3 at 0. which avoided imposing a false model which assumed and optimised independence of the factors. The three components were defined by their high positive or negative loadings on the test measures. Factor 2 was defined by high positive loadings for primary psychopathy and the MES.6% of the observed test variance.54 0. Angus / Personality and Individual Differences 36 (2004) 575–586 4.1. accounting for 71. Egan. This analysis generated three principal components.86 0.

0. correlations were calculated between the derived factors and the emotional–sexual trade-off score and the frequency of infidelity measure.s.13 13. This suggested that males who committed infidelity were higher in social dominance than those who had not. 4. A similar analysis (correcting for the influence of the other factors) for manipulativeness showed strong significant effects for sex and whether the person had engaged in an affair (both which may be expected).3.s. n. males had higher rates than females.001) and frequency of infidelity (r=0. n. P<0. females were higher in N.04 n.001 n.s. Those who committed infidelity were predictably higher than those who did not on measures of primary .41. 4. Only in the case of the manipulativeness factor was there an association with increased sexual emphasis (r=0.71 Affair P< (83. Following the previous literature we expected sex differences between males and females in the degree of infidelity. The same analysis for the outgoingness (controlling for social dominance and manipulativeness) found no differences by sex or sexual infidelity.s.s. This indicates that sex-related parental investment factors are in operation even when infidelity occurs.71 0. P< 0. manipulativeness or outgoingness would lead to the individual committing more infidelity or emphasising the sexual element of their infidelity. but no interaction. Correlations between the derived factors and infidelity To examine whether greater dominance. whereas this relationship reversed for females. and mating effort than females.001). This result suggests a differential sexual strategy in operation for males and females who decide to have sex with a person who is not their regular partner.s. Discussion The current study examined self-reported acts of sexual infidelity in individuals engaged in ongoing relationships in relation to personality. These results suggest a specific differential personality effect across the sexes for persons who have affairs. Angus / Personality and Individual Differences 36 (2004) 575–586 583 Table 4 F-ratios for two-way ANOVA dividing the sample by sex and whether they have had an affair or not for the derived factor scores. psychopathy and mating effort. 1) n. between sex and having an affair for social dominance.V. S. 1) F1: dominance F2: manipulativeness F3: openness 0. controlling each factor by the other two factors Sex P< DF (83.. We found males were higher in primary and secondary psychopathy.001 n. 0. non-significant. Egan. When analysis was confined exclusively to the participants who had committed infidelity.45.33 0.32 0.82 1. with females being fussier about whom they are willing to be unfaithful with. 5.2.64 0. 0.06 Sex affair P< (83. 1) n. Females who admitted to having had a sexual relationship with another person whilst ostensibly committed to another person were lower in social dominance than those who had not.47 44.s. A and social desirability.

and to find a way to have sex with another person despite the other person’s mate-guarding strategies. mating effort and social desirability were all substantially correlated and easily reduced to three underlying factors—social dominance. and that the personalities of male and female adulterers differ relative to males and females who remain faithful. as were less socially dominant women. we do not know if these males were also materially successful and powerful. Buss and Shackelford (1997) and Hoyle et al (2000) note that personality traits like manipulation. While our data does not provide answers to the more complex questions of rationale behind infidelity. more socially dominant women) was the case for those that did not. Unfaithful females are lower on this dimension. our results provide further evidence that psychopathic-type self-reports are markedly associated with short-term. and are thus more likely to be low in E. it does give evidence to suggest that individual differences in personality may be salient to such behaviour. high N. the controversial and ethical complexities of this research area make such a study understandably difficult to conduct. Females attach to cues in males reflecting greater potential resource acquisition (Buss. We observed significant correlations between high mating effort. openness did not differentiate sex or adulterers. This is relatively straightforward to understand. Nor could we examine if the infidelity reflected dissatisfaction in the unfaithful person’s main relationship. A. The strength of the current study is our examination of actual sexual infidelity in normal adults rather than the hypothesised desire for multiple partners in student populations. lower in C and A and indifferent to expressing themselves in a socially desirable way. opportunistic and exploitative sexual strategies (Rowe. We found no interaction between sex and infidelity for such traits. psychopathy. low A and high primary psychopathy and frequency of infidelity. Females at the other end of the same dimension may seek to increase their access to such characteristics. The present study shows personality and behaviour to be important predictors of sexual infidelity in humans.584 V. low N. or the interaction of these variables. Within this experimental design the study was unable to determine whether the personalities of individuals who are unfaithful are actually complementary. deceitfulness. the reverse (less socially dominant men. This is a strong test of hypotheses underlying infidelity. This showed that more socially dominant men were more likely to have extra-pair copulation. Whilst such a natural history approach would clarify whether our cross-sectional observations were correct. irrespective of biological sex. In addition. While we could establish that the men who were unfaithful were socially dominant. Angus / Personality and Individual Differences 36 (2004) 575–586 psychopathy and mating effort. S. C and seeking to present themselves in a socially desirable way. Taken in conjunction with the association seen between mating effort and primary psychopathy. The scales of personality. and were also lower in A and social desirability. 1989). low A and high psychopathy are more common in individuals who are more likely to be unfaithful. Egan. the skills needed to carry out infidelity and mate-stealing require an active willingness to betray another’s trust and intimacy. One way of interpreting this result is that males with such attributes are seen as more dominant and successful. and perhaps . and openness. Manipulativeness was greater in males and in individuals who had affairs. manipulativeness. A sex by infidelity ANCOVA found that it was only in the case of stable dominance that a differential sex-related strategy occurred. Unfaithful males are higher on a dimension defined by E. but did not show any interaction. This because we did not ask about current extra-relationship affairs or the personalities of the dyad involved. and males are higher than females on these traits. we collected descriptive information assessing the personalities of the individuals in a range of domains to examine whether particular types of males and females were more likely to carry out this behaviour. 1995).

Buss. J. R. Deary. S. 907–920. Loo. Women’s sexual strategies: the hidden dimension of extra-pair mating. (2000). T. Journal of Sex Research. T. 35. (2000). Buss. & Newman. 151–158. 1081–1093. J. Journal of Youth and Adolescence. Behavioural and Brain Sciences. Buss. (2000). R. & Shackelford. S. T. D. W. Kiehl. 204–232. Journal of Personality & Social Psychology. Costa.. Levenson. R. 19–31. M. Confirmatory factor analyses of the full and short versions of the Marlowe–Crowne Social Desirability Scale.. A. 192–210. Agyei. 28. Sex differences in sexual fantasy: an evolutionary approach. S. P. A. Causes of conjugal dissolution: a cross-cultural study. Schmitt. M. Personality and Individual Differences. Journal of Personality and Social Psychology. W. and thus are in some way driven to seek out such individuals. Egan. Greiling. & Cauffman. The revised NEO personality inventory (NEO-PI-R) and NEO five factor inventory (NEO-FFI).. M. (1991). The NEO-FFI: emerging British norms and an item-level analysis suggest N. D. D. B.. K. 29. Assessing psychopathic attributes in a noninstitutionalized population. (1999). C. 100. 527–555. The sociobiology of sociopathy. FL: Psychological Assessment Resources. References Bailey. Odessa. D. (1995).. Machiavellianism and psychopathy. Males following a short-term sexual strategy will not be concerned with personality characteristics that might make for a good mother. A. H. The (un)acceptability of betrayal: a study of students’ evaluations of sexual betrayal by a romantic partner and betrayal of a friend’s confidence. (1992). S. H. 18. M. & Cauffman. . & Buss. & McCrae. E. & Leukefeld. (1998).. 523–599. (1989). A. R. Lynam. I. Worzel.. Mealey. Gaulin. D. London: Bloomsbury. & Symons. & Miller. Betzig. J.. 31. Miller. & Cory. 68. Sexual betrayal among late adolescents: perspectives of the perpetrator and the aggrieved. R. J. Buss. D. Current Anthropology. The evolution of desire: strategies of human mating.. Journal of Sex Research. & Gladue. One way to test this hypothesis might be to examine how long someone is unfaithful with the same partner. S. Buss. M. L. 1021–1038. 628–635.. 28. D. S. D. C. W. Hare. (1998). Angus / Personality and Individual Differences 36 (2004) 575–586 585 even seek genes that increase the probability of imparting such characteristics to their offspring. (1997). S. Journal of Youth and Adolescence. M. J. (2000). Journal of Research in Personality. (1989).. Personality and Individual Differences. Journal of Social Psychology.. Effects of gender and sexual orientation on evolutionary relevant aspects of human mating psychology. M. Fejfar. 499–523.. Y. (1993).. C. (1990).. D. 68. & Schmitt. D. The dangerous passion: why jealousy is as necessary as love or sex.. 929–963. Feldman. M. E. Suceptibility to infidelity in the first year of marriage. 654–676. & Szyarto. Toronto: Multi-Health Systems. Smith. London. Personality disorders as extreme variants of common personality dimensions: can the five-factor model adequately represent psychopathy? Journal of Personality. D. UK: Allyn and Bacon... Sex differences in human mate preferences: evolutionary hypotheses tested in 37 cultures. & Austin. J. 140. 1203–1231. Buss. 193–221. Psychological Review. Construct validation of a self-report psychopathy scale: does Levenson’s self-report psychopathy scale measure the same constructs as Hare’s psychopathy checklist-revised? Personality and Individual Differences. E. F. R. K. D. Buss. (1994). 12. D. (2000). Behavioural and Brain Sciences. Evolutionary psychology: the new science of the mind. Personality and sexual risk taking: a quantitative review. V. Sexual strategies theory: an evolutionary perspective on human mating. (1994). B. A. professional manual. McHoskey. 235–258. P. (2000). 66. R. New York: Basic Books. Brinkley. M. C. M. 253–275.. Sexual strategies theory: historical origins and current status.. 30. (2001). (1999). S. The Hare psychopathy checklist-revised. 4. Hoyle. P. an integrated evolutionary model. Egan. Widiger. (1995). Journal of Personality and Social Psychology. K. 74. L. 1–49. M. 29. M. & Thorpe. 31.V. A and C are more reliable than O and E. D.. Journal of Personality... Ellis. 69. (2001). Feldman...

.. Personal Relationships. Equity. Mating-effort in adolescence: a conditional or alternative strategy. Personality and unrestricted sexual behaviour: correlations of sociosexuality in Caucasian and Asian college students. T. Development of a reliable and valid short form of the Marlowe–Crowne Social Desirability Scale. 202–212. J. & Reise. A... C. T. D. (1997). R. Spitzberg. (1998). 38. D. E. 31. D. mating effort and crime.. Context-specific mate choice criteria: women’s trade-offs in the contexts of long-term and extrapair mateships. A. (1995). Lynam. London. Wright. T... (1993). P. (1995). W. Sheppard.. 10. Dating relationships and infidelity: attitudes and behaviour. V. 8. 18. In Meeting of the American Psychological Association. E.. Journal of Clinical Psychology. (2001). 371–389. 573–574. (1997). K. Washington. Egan. 166–192. Angus / Personality and Individual Differences 36 (2004) 575–586 Miller.. M. Journal of Sex and Marital Therapy. Buunk.). N. (1982). S. 39–53. Risky sexual behaviour and the five factor model of personality. C. S. S. Journal of Social and Personal Relationships. Reynolds. (2000). Cupach. W. P.. Personality and Individual Differences. Logan. J.. 119–125. Zimmerman..586 V. 21. normative disapproval and extramarital relationships.. V. Prins. W. B. & Clayton. UK: Lawrence Erlbaum Associates. Journal of Research in Personality. J. 23. & Figueredo. C. R. Vazsonyi. Behavioural and Brain Sciences. Rowe. H. Nelson. Rowe. 105–115. (Eds. & VanYperen. The dark side of human relationships. DC. Scheib. R. Leukefeld. B. Evolution. M. & Andreoli-Mathie.