High (in)fidelity: gender, the Dark Tetrad, and infidelity

Abstract

Infidelity is a common problem in intimate relationships and is often a main contributor to relationship breakdowns. The aim of the current study was to extend the infidelity literature by differentiating between physical, emotional, and malevolent forms of infidelity, and exploring individual differences (i.e., gender and The Dark Tetrad of personality) that are associated with perpetration in these forms. The sample comprised 240 participants aged 18 to 67 years (M_age = 29.41 years, SD = 10.32; 76.8% women). People with higher psychopathy were more likely to engage in all forms of infidelity. High narcissism was predictive of emotional infidelity, and high sadism was predictive of malevolent infidelity. Interestingly, gender was only predictive of malevolent infidelity, with women more likely than men to perpetrate this form of infidelity. Exploratory analyses indicated that many of the associations between Dark Tetrad and infidelity were moderated by gender. Results of the current study may support researchers and therapists in differentiating between different types of infidelity (i.e., emotional, physical, malevolent). Further, these findings may contribute to the development of screening tools for therapists working who provide support for infidelity in intimate relationships.

LAY SUMMARY

We perform of a novel exploration of the Dark Tetrad of personality and perpetration of physical, emotional, and malevolent infidelity. By demonstrating differential patterns of predictors for these forms, we provide support for a dimensional conceptualisation of infidelity. These findings may have important clinical implications for relationship researchers and counsellors.

Keywords:

• Dark tetrad
• narcissism
• psychopathy
• sadism
• machiavellianism
• infidelity

Intimate adult relationships can be one of the most meaningful relationships many individuals experience; however, a common problem in these types of relationships is infidelity (Blow & Hartnett, 2005b; Leeker & Carlozzi, 2014; Mark et al., 2011). Although definitions vary (Blow & Hartnett, 2005a), infidelity commonly refers to any action taken individually by a companion in a committed romantic relationship that encroaches upon the explicit or assumed agreement of exclusive sexual or emotional bonding between the two individuals (Carlson & Sperry, 2010). While infidelity is multifaceted, it is generally characterised by physical and emotional forms (DeSteno & Salovey, 1996; Moller & Vossler, 2015; Wilson et al., 2011). Physical infidelity includes physical and/or sexual acts with someone other than an individual’s primary partner (Leeker & Carlozzi, 2014). Emotional infidelity refers to forming an emotional connection with someone other than one’s current partner (Wilson et al., 2011), and is commonly characterised by deceiving one’s current partner about their feelings towards the other person, sharing important information with the other person exclusively, and/or attending important events with the other person (Guitar et al., 2017). Physical and emotional infidelity can co-occur (e.g., extra-relational romance; Wilson et al., 2011) or occur separately, such as physical infidelity without emotional connection or an emotional connection without sexual behaviour (Guitar et al., 2017).

In the United States, infidelity is relatively common and occurs in approximately 25% of married, heterosexual relationships (Blow & Hartnett, 2005a). Infidelity is often associated with significant negative consequences for individuals (e.g., decreased self-esteem, increased psychological distress; Blow & Hartnett, 2005b; Buunk, 1995; Cano & O’Leary, 2000; Shackelford, 2001; Wilson et al., 2011) and relationships (e.g., relationship breakdown; Garbinsky et al., 2020). Given the relatively common occurrence of infidelity and the individual and relational impacts, there is a need to understand the factors that contribute to acts of infidelity. A number of individual differences have been explored in relation to infidelity perpetration, including demographics (Mark et al., 2011), sexual satisfaction (Buss & Shackelford, 1997), and relationship satisfaction (Atkins et al., 2001). In the current study, we adopt recommendations of previous researchers (Altgelt et al., 2018; Mark et al., 2011) and explore Dark Tetrad personality traits as predictors of perpetrating physical and emotional infidelity.

The Dark Tetrad and infidelity

The Dark Tetrad of personality—comprising narcissism, psychopathy, Machiavellianism, and sadism (Chabrol et al., 2009)—is a collection of four related, but distinct, subclinical personality traits characterised by aggression, exploitation of others, disagreeableness, and deception (Forsyth et al., 2021). The Dark Tetrad traits share small-to-moderate intercorrelations across multiple studies, and are united by a common core (i.e., low honesty/humility and low agreeableness) within the six factor lexical personality model (Međedović & Petrović, 2015). However, the Dark Tetrad traits are also considered characteristically distinct. Narcissism includes a grandiose sense of self-importance, a sense of superiority (Chabrol et al., 2009), low empathy, dominance, a sense of entitlement, and superiority (Caligor et al., 2015). Psychopathy is characterised by antisocial behaviours, thrill-seeking, poor impulse control, low empathy, a callous interpersonal style, grandiosity, and manipulation (Hare, 2003; Hare & Neumann, 2008; Waddell et al., 2020), Machiavellianism is characterised by strategy, cynicism, immorality, and planning (Geis & Levy, 1970; Jones & Paulhus, 2014), and sadism by the desire to harm others, as well as interpersonal control, humilitation, and punishment (Craker & March, 2016; Međedović & Petrović, 2015; Nickisch et al., 2020).

Individuals with higher levels of narcissism, psychopathy, and Machiavellianism are more likely to perpetrate infidelity (see Brewer et al., 2015; Jones & Weiser, 2014; Sevi et al., 2020). This propensity is attributed to key characteristics of each trait (Jones & Weiser, 2014), such as the unrestricted sociosexuality and lower relational commitment associated with narcissism (Foster et al., 2006), the tendency for those with high psychopathy to prioritise short-term mating (Jonason et al., 2019), and the lack of interpersonal affect and high levels of strategic deception associated with Machiavellinaism (Baughman et al., 2014; Jonason et al., 2014). However, past studies have typically operationalised and measured infidelity as a unidimensional construct (see Jones & Weiser, 2014). As the physical and emotional forms of infidelity are considered distinct conditions (Kruger et al., 2015), the failure to distinguish these forms conceptually limits this research.

The lack of differentiation between physical and emotional forms of infidelity may be especially problematic for studies exploring narcissism, psychopathy, and Machiavellianism. Specifically, as those with higher levels of narcissism, psychopathy, and Machiavellianism display increased propensity for risky sexual behaviour (Jonason et al., 2010), it is possible that these individuals are more likely to perpetrate physical, not emotional, infidelity. Those with high psychopathy may also engage in infidelity for reasons (e.g., fast life strategy; Jonason et al., 2010) that differ from those with high narcissism (e.g., low commitment to the primary partner); thus, it is possible that these individuals will exhibit differential propensity to perpetrate physical and emotional infidelity.

So far, the research exploring infidelity and these “dark” traits have largely considered the Dark Triad, including only narcissism, psychopathy, and Machiavellianism (Paulhus & Williams, 2002). More recently, this triad has been reconceptualised as a tetrad with the inclusion of nonclinical sadism (Chabrol et al., 2009). To the best of our knowledge, the role of sadism in predicting infidelity is yet to be explored. Given the conceptual overlap between the Dark Tetrad traits (Chabrol et al., 2009) and the associations between Dark Triad traits and infidelity (Sevi et al., 2020), there is potential for individuals with high sadism to perpetrate more infidelity. In the current study we address this issue by differentiating between physical and emotional infidelity and exploring the Dark Tetrad traits as predictors of perpetration of both forms of infidelity.

Gender and infidelity

In addition to exploring the Dark Tetrad traits we also explore gender as a predictor of physical and emotional infidelity. The relationship between gender and perpetration of infidelity appears inconsistent with some studies reporting men engage in more acts of infidelity than do women (Allen & Baucom, 2004), while others report women engage in more infidelity than do men (Brand et al., 2007). One possible explanation for this discrepancy may be the unidimensional measurement of infidelity. As men, compared to women, are more likely to engage in sexual infidelity (Lalasz & Weigel, 2011), it is possible that failing to differentiate between forms of infidelity obscures the role gender plays in perpetration. Another possibility is that, if women engage in certain forms of infidelity more than men (Brand et al., 2007), but men (on average) score higher on each of the Dark Tetrad traits than do women (Craker & March, 2016; Jonason et al., 2010), these “dark” personality traits may interact with gender to predict infidelity.

Past infidelity research has reported differential patterns of relationships between “dark” personality traits and infidelity for men and women. Compared to men, women high on Machiavellianism reported stronger intentions to commit infidelity (Moore et al., 2020). Previous research exploring infidelity as a unidimensional construct (i.e., not differentiating between physical and emotional forms) reported that, for men, psychopathy was the main correlate of infidelity. However, for women, psychopathy and Machiavellianism were equal correlates of infidelity (Jones & Weiser, 2014). Importantly, researchers exploring “dark” personality traits and infidelity recommend future researchers explore gender as a moderating variable to assess whether gender changes the positive association between dark triad traits and (a) attitudes towards infidelity, and (b) reported infidelity behaviors (Sevi et al., 2020). In the current study, we explore the potential for gender to moderate the relationships between the Dark Tetrad personality traits and physical and emotional infidelity.

Infidelity: a malevolent motivation

Although acts of infidelity are largely attributed to physical (i.e., sexual) and emotional motivations, other motivations to commit infidelity include relationship dissastifaction, neglect, sexual desire, anger, and revenge (Barta & Kiene, 2005). Adopting romantic revenge terminology, infidelity motivated by revenge is characterised as an act of retribution, designed to enact an emotional or reputational cost on one’s primary romantic partner (Brewer et al., 2015; Brewer & Abell, 2015; Rasmussen & Boon, 2014). We term this form of infidelity malevolent infidelity. Although acts of infidelity are commonly motivated by a desire for revenge (Zare, 2011), little is known about the individual differences related to the perpetration of malevolent infidelity. In the current study, we explore, for the first time, the role of the Dark Tetrad traits as predictors of infidelity motivated by anger and committed as a form of revenge. Although researchers have speculated that those with high levels of these traits have different motives for engaging in infidelity (Jones & Weiser, 2014), to date these motives have been largely unexplored. Due to the sense of entitlement associated with narcissism (Caligor et al., 2015), the callousness associated with psychopathy (Hare, 2003; Hare & Neumann, 2008), the manipulation associated with Machiavellianism (Geis & Levy, 1970), and the enjoyment of causing harm associated with sadism (Craker & March, 2016; Nickisch et al., 2020), it is likely that these traits will positively predict propensity to engage in malevolent infidelity.

In addition to the Dark Tetrad traits, we also explore the role of gender in malevolent infidelity. Compared to women, men experience more positive emotions (e.g., pride) when pursuing revenge-related goals (Singer et al., 2006). As such, men may be more inclined to commit infidelity as a way to enact revenge on, and hurt, their partner. Relatedly, if men engage in infidelity for malevolent reasons, and score higher (on average) on each of the Dark Tetrad traits than do women (Craker & March, 2016; Jonason et al., 2010), it is possible that gender will moderate the relationships between the Dark Tetrad traits and malevolent infidelity. Thus, we extend previous infidelity research by exploring (1) the utility of gender and the Dark Tetrad traits to predict propensity to engage in malevolent infidelity, and (2) the potential for gender to moderate the relationships between the Dark Tetrad traits and malevolent infidelity.

Aims and hypotheses

The aims of the current study are to explore (1) the utility of gender and the Dark Tetrad traits to predict past perpetration of physical and emotional infidelity, (2) the potential for gender to moderate the relationships between the Dark Tetrad traits and these forms of infidelity, (3) the utility of gender and the Dark Tetrad traits to predict the propensity to engage in malevolent infidelity, and (4) the potential for gender to moderate the relationships the Dark Tetrad traits and malevolent infidelity. These findings may have important practical, therapeutic implications. For example, intimate relationship therapists may benefit from a more nuanced understanding of why people engage in different forms of infidelity. For example, people with high psychopathy may be inclined to engage in risky behaviour with more immediate benefit (e.g., physical infidelity), whereas people with high sadism may be inclined to engage in malevolent infidelity as they derive enjoyment in emotionally harming their partner. Screening for these traits in couples seeking relationship counselling for infidelity may assist therapists establishing therapeutic interventions and strategies.

Based on previous research, we hypothesised that all Dark Tetrad traits would be positive predictors of past physical and emotional infidelity and a malevolent motivation to commit infidelity. As previous findings regarding gender and infidelity have been inconsistent, gender is entered as an exploratory variable with no predicted direction. Further, given the lack of research on gender, the Dark Tetrad, and infidelity, the interactions between gender and the Dark Tetrad traits were exploratory with no specific hypotheses.

Method

Participants

There were 240 Australian participants aged between 18 and 67 years (M_age= 29.41 years, SD = 10.32¹). Of participants, 75.8% identified as women, 22.9% identified as men, and 1.3% identified their gender as ‘other’. Participants reported their relationship status as either single (30.4%), committed (30.4%), defacto² (15%), married (20%), or divorced (4.2%). For sexual orientation, 84.6% of participants identified as heterosexual, 4.6% identified as homosexual, 6.7% identified as bisexual, and 4.2% identified as ‘other’³.

Measures

Participants completed an online questionnaire which comprised a demographics section (e.g., gender, age, relationship status) and the following measures.

The short dark triad measure (SD3; Jones & Paulhus, 2014)

The SD3 is a 27-item self-report inventory that includes three subscales assessing trait narcissism (e.g., “People see me as a natural leader”), Machiavellianism (e.g., “It’s not wise to tell your secrets”), and psychopathy (e.g., “I like to get revenge on authorities”). Responses are scored on a 5-point Likert scale (1 = Disagree strongly to 5 = Agree strongly). Scores are summed to produce total subscale scores, with high scores indicating high levels of the relevant trait.

The short sadistic impulse scale (SSIS; O’Meara et al., 2011)

The SSIS is a 10-item self-report inventory that assesses trait sadism (e.g., “I would enjoy hurting someone physically, sexually or emotionally”). Each of the items is rated ‘like me’ (scored 1) and ‘unlike me’ (scored 0). Items are summed for a total score, with higher scores indicating higher levels of sadism.

Measures of infidelity

We followed previous methodology (Jones & Weiser, 2014) to assess infidelity by asking participants if they had previously “cheated” (i.e., been unfaithful) in a past relationship. Participants responded yes/no, with 34.2% of participants indicating they had previously cheated in a relationship. Following this, we then assessed physical infidelity via one item (“if you have cheated in the past, were you unfaithful physically/sexually [e.g., intercourse, oral sex, heavy petting/fondling]”), and emotional infidelity via one item (“If you have cheated in the past, were you unfaithful emotionally? [e.g., hugging, buying/receiving gifts, talking on the phone/internet]”). The examples following each question (e.g., intercourse, hugging) were derived from previous research (Wilson et al., 2011). Although previous research (Altgelt et al., 2018; Jones & Weiser, 2014) has typically assessed infidelity perpetration using dichotomous scoring (e.g., 0 = no, 1 = yes), we adopted a Likert scale to enhance response discrimination. Responses were recorded on a 7-point Likert scale (0 = Never to 6 = Very frequently) and these response frequencies can be seen in Table 1.

Table 1. Participant frequency responses to previous physical infidelity and previous emotional infidelity.

	Physical Infidelity			Emotional Infidelity
	Total	Men	Women	Total	Men	Women
Never	70%	61.8%	72%	63.3%	56.4%	65.4%
Very rarely	10%	10.9%	9.9%	9.2%	14.5%	7.7%
Rarely	6.3%	3.6%	7.1%	8.3%	3.6%	9.3%
Occasionally	5.4%	9.1%	4.4%	12.1%	18.2%	10.4%
Somewhat Frequently	5.0%	12.7%	–	2.1%	1.8%	2.2%
Frequently	1.7%	–	2.7%	3.3%	5.5%	2.7%
Very Frequently	1.7%	1.8%	2.2%	1.7%	–	2.2%

To assess malevolent infidelity, we developed two items based on romantic revenge literature (see Rasmussen & Boon, 2014). Participants responded to these two items (“I would use cheating as a form of revenge against my partner” and “I would use cheating as a way of causing my partner pain”) using a 7-point Likert-type scale ranging from 1 = Strongly disagree to 7 = Strongly agree. The two items had a strong positive correlation (r[237] = .76, p < .001) and good internal consistency⁴ (Cronbach’s α = .85) and, as such, were summed to produce a total malevolent infidelity score. The decision to present the malevolent infidelity items as hypothetical, compared to the physical and emotional infidelity incidence items, was intentional. We speculated that given the potential destructive nature of revenge (Sheppard & Boon, 2012), participants may be less inclined to admit to incidence of perpetrating malevolent infidelity. As such, we framed the items as hypothetical to elicit more honest responses.

Procedure

Prior to recruitment, ethical approval was granted from the Federation University Austraia Human Research Ethics Committee (Project No. A15-063). Participants were recruited via advertisements posted on social media (e.g., Facebook, Instagram) from June 2015—August 2015. To participate in the study, potential participants clicked on the URL in the advertisement which directed them to the online questionnaire (hosted by surkeymonkey.com). Upon arriving at the questionnaire, potential participants read an information letter which contained the study aims, the voluntary and anonymous nature of the study, and that participants could withdraw without penalty. Participants were advised that by clicking ‘yes’ to commence the questionnaire, they were providing informed consent to participate. Participants were also provided with support contacts in the event they experienced any emotional discomfort during the questionnaire. Upon completion of the questionnaire, participants were thanked for their participation.

Design and analyses

The current study used a correlational, cross-sectional design. The predictor variables were gender⁵ (men coded 0, women coded 1), narcissism, psychopathy, Machiavellianism, and sadism. There were three criterion variables: Physical infidelity, emotional infidelity, and malevolent infidelity. Hypotheses were tested with three Multiple Regression Analyses. An a priori power test calculated using G*Power (Faul et al., 2007)—with power set at .95, alpha at .05, and an effect size of .15 with 5 predictors—indicated a sample size of 138 was required to achieve adequate power.

Results

Residual scatterplots were used to assess the assumptions of linearity and homoscedasticity, and assumptions were considered met. Normality was assessed via residual scatterplots and histograms, with distributions of psychopathy and sadism positively skewed. However, as the F test is considered generally robust (Keppel & Wickens, 2004) and these distributions are typically skewed, transformations were not performed. One univariate outlier was identified on the Machiavellianism scale, one on the narcissism scale, and one on the psychopathy scale. Upon further inspection, these outliers were removed due to the pattern of responses (e.g., participants answered ‘strongly agree’ to every question). Multivariate outliers were detected via Mahalanobis distance values. Upon screening these outliers they were determined legitimate responses and thus retained for analyses.

Total and gendered descriptive statistics and bivariate correlations can be seen in Table 2. For total correlations, all Dark Tetrad personality traits shared significant, positive correlations with the forms of infidelity. As can be seen in Table 2, correlations between the Dark Tetrad traits and the forms of infidelity differed between men and women (Fisher’s z < .05), indicative of potential moderation. For mean differences, men scored significantly higher than women on narcissism, psychopathy, and sadism, a trend consistent with previous research (Lyons et al., 2022). However, there was no significant gender difference in Machiavellianism.

Table 2. Total and sex differentiated zero-order correlations, descriptive statistics, and t-tests.

	1	2	3	4	5	6	7
Total Correlations
1. Narcissism	–
2. Psychopathy	.31**	–
3. Machiavellianism	.24**	.52**	–
4. Sadism	.30**	.52**	.47**	–
5. Physical infidelity	.18**	.27**	.13*	.18**	–
6. Emotional Infidelity	.22**	.29**	.18*	.18**	.65**
7. Malevolent infidelity	.16*	.35**	.33**	.38**	.23**	.26**	–
Correlations by Sex
1. Narcissism	–	.25**a	.23**a	.22**a	.10a	.15*b	.08b
2. Psychopathy	.44**a	–	.53**a	.49**a	.22**a	.26**a	.30**b
3. Machiavellianism	.31*a	.50**a	–	.45**a	.14a	.20**a	.32**a
4. Sadism	.48**a	.56**a	.50**a	–	.06b	.10b	.31**b
5. Physical infidelity	.32*a	.37**a	.12a	.39**a	–	.64**a	.18*a
6. Emotional Infidelity	.41**a	.36**a	.13a	.39**a	.69**a	–	.22**a
7. Malevolent infidelity	.50**a	.60**a	.40*a	.65**a	.41**a	.44**a	–
Total M (SD)	24.67 (5.25)	19.76 (5.52)	25.03 (5.71)	1.13 (1.79)	1.77 (1.44)	1.97 (1.53)	3.31 (2.51)
Men M (SD)	25.94 (5.55)	21.04 (6.04)	25.19 (6.16)	1.60 (2.28)	2.07 (1.63)	2.11 (1.55)	3.02 (2.41)
Women M (SD)	24.36 (5.12)	19.41(5.33)	24.97 (5.54)	0.98 (1.58)	1.69 (1.38)	1.93 (1.54)	3.38 (2.53)
t-value	1.96*	1.92*	0.25	1.89*	1.57	0.74	−0.93
Hedge’s g	0.30	0.29	0.04	0.05	−0.04	−0.19	−0.44

Note. Correlation coefficients below the diagonal are for men, correlation coefficients above the diagonal are for women; different subscript indicates correlations between sexes differ at Fisher’s z, p < .05; * p < .05; ** p < .01.

Gender, narcissism, psychopathy, Machiavellianism, and sadism explained a statistically significant 7.6% (Adjusted R²) of variance in physical infidelity, R² = .10, F(5, 228) = 4.85, p < .001, with a small-to-medium effect (ƒ² = .11; Cohen, 1992). Further, gender, narcissism, psychopathy, Machiavellianism, and sadism explained a significant 8.6% (Adjusted R²) of variance in emotional infidelity, R² = .11, F(5, 233) = 5.37, p < .001, with a small-to-medium effect (ƒ² = .12). Lastly, gender, narcissism, psychopathy, Machiavellianism, and sadism explained a significant 20.7% (Adjusted R²) of variance in malevolent infidelity, R² = .22, F(5, 231) = 13.05, p < .001, with a medium-to-large effect (ƒ² = .28). Coefficients for all regressions can be seen in Table 3. As can be seen in Table 3, psychopathy was a significant, positive predictor of physical infidelity, psychopathy and naricssim were significant, positive predictors of emotional infidelity, and gender (women), psychopathy, and sadism were significant, positive predictors of malevolent infidelity.

Table 3. Coefficients for gender, narcissism, psychopathy, Machiavellianism, and sadism regressed on physical infidelity, emotional infidelity, and malevolent infidelity.

	B (95% CI)	SE	β	T
Physical Infidelity
Constant	0.26 (-0.99 − 1.52)	0.64
Gender	−0.24 (-0.68 − 0.19)	0.22	−0.07	−1.10
Narcissism	0.03 (-0.01 − 0.06)	0.02	.09	1.39
Psychopathy	0.06 (0.02 − 0.11)	0.02	.24	2.96**
Machiavellianism	−0.01 (-0.05 − 0.03)	0.02	−0.03	−0.45
Sadism	0.04 (-0.09 − 0.17)	0.07	.05	0.59
Emotional Infidelity
Constant	−0.49 (-1.81 − 0.84)	0.67
Gender	−0.01 (-0.47 − 0.45)	0.23	−0.01	−0.05
Narcissism	0.04 (0.00 − 0.08)	0.02	.15	2.19*
Psychopathy	0.07 (0.02 − 0.11)	0.02	.24	3.01**
Machiavellianism	0.01 (0.04 − 0.04)	0.02	.01	0.10
Sadism	0.01 (-0.13– 0.14)	0.07	.01	0.04
Malevolent Infidelity
Constant	−1.68 (-4.02 − 0.55)	1.19
Gender	0.73 (0.03 − 1.43)	0.35	.12	2.05*
Narcissism	0.01 (-0.5 − 0.07)	0.03	.02	0.32
Psychopathy	0.08 (0.02 − 0.15)	0.03	.18	2.42*
Machiavellianism	0.06 (-0.01 − 0.12)	0.03	.13	1.79
Sadism	0.38 (0.17 − 0.58)	0.10	.26	3.62***

Note. *p < .05; **p < .01; Gender is coded as 0 = men, 1 = women.

To test interactions, a series of PROCESS moderation analyses were run with gender as the moderator, the Dark Tetrad traits as the predictors, and physical infidelity, emotional infidelity, and malevolent infidelity as the criterions. There were interactions between gender and sadism for physical infidelity (β = −0.22, SE = .11, p = .035), and gender and narcissism for malevolent infidelity (β = −0.18, SE = .07, p = .011). No other interactions reached statistical significance. Although the interaction between gender and narcissism for physical infidelity did not reach statistical significance (β = −0.07, SE = .04, p = .094), there was a significant conditional effect which we tentatively report. The conditional effects⁶ for the moderation analyses can be seen in Table 4 and are depicted visually in Figure 1. As shown in Figure 1, for high levels of sadism, men were more likely than women to perpetrate physical infidelity. At low levels of narcissism, men were less inclined to engage in malevolent infidelity than women. Lastly, at high levels of narcissism, men were more likely than women to perpetrate physical infidelity—though the interaction coefficient did not reach statistical significance.

Figure 1. Conditional effects for the moderator (sex), predictors (narcissism and sadism), and criterions (physical and malevolent forms of infidelity); low = -1 SD; average = M; high = +1SD.

Table 4. Conditional effects for the moderator (sex), predictors (narcissism and sadism), and criterions (physical and malevolent forms of infidelity).

	Effect (95% CI)	SE	t
Physical Infidelity
Sadism
Men	0.28 (0.11; 0.45)	0.08	3.32**
Women	0.05 (-0.08; 0.18)	0.07	0.79
Narcissism
Men	0.10 (0.03; 0.17)	0.04	2.73**
Women	0.03 (-0.01; 0.07)	0.02	1.32
Malevolent Infidelity
Narcissism
Men	0.22 (0.10; 0.34)	0.06	3.60**
Women	0.04 (-0.03; 0.11)	0.04	1.03

Note. *p < .05; **p < .01; Gender is coded as 0 = men, 1 = women.

Discussion

In this study, we explored (1) the utility of gender and the Dark Tetrad traits to predict perpetration of physical and emotional infidelity, (2) the potential for gender to moderate the relationships between the Dark Tetrad traits and these forms of infidelity, (3) the utility of gender and the Dark Tetrad traits to predict the propensity to engage in malevolent infidelity, and (4) the potential for gender to moderate the relationships between the Dark Tetrad traits and malevolent infidelity. We hypothesised that all Dark Tetrad traits would be positive predictors of physical infidelity, emotional infidelity, and malevolent infidelity, and results provided partial support for the hypothesis. High psychopathy predicted physical infidelity and emotional infidelity, high narcissism explained unique variance in emotional infidelity, and high psychopathy and sadism predicted malevolent infidelity. Machiavellianism did not predict perpetration of physical or emotional forms of infidelity over narcissism, sadism, and psychopathy. Gender and the interactions between gender and the Dark Tetrad traits were exploratory with no specific hypotheses. When accounting for the variance explained by the Dark Tetrad traits, women (compared to men) showed greater propensity for malevolent infidelity. Lastly, gender moderated the relationships between sadism, narcissim, and physical infidelity, and narcissism and malevolent infidelity. We discuss results in terms of individual predictors followed by moderations.

Gender, the Dark Tetrad, and infidelity

After accounting for the variance explained by Dark Tetrad traits, the lack of utility for gender to predict either physical or emotional forms of infidelity contributes to the extant literature, providing further evidence that gendered-trends of infidelity are disappearing. However, compared to men, women showed greater propensity to engage in malevolent infidelity. Interestingly, this result provides the first empirical evidence of a trend identified by IllicitEncounters.com⁷ (Naftulin, 2018), where women were more likely to cheat as a form of revenge upon an unfaithful partner. Future research could explore whether this propensity translates to perpetration, and if women’s malevolent infidelity is a response to an unfaithful mate or is motivated by a more general intent to cause psychological harm to one’s partner (Brewer et al., 2015; Brewer & Abell, 2015; Rasmussen & Boon, 2014).

Of the Dark Tetrad traits, only psychopathy was a predictor of both physical and emotional infidelity. The unique relationship between psychopathy and infidelity is consistent with previous research (Jones & Weiser, 2014; Sevi et al., 2020). The tendency for psychopathy to predict all forms of infidelity may be attributed to the callous, thrill-seeking, and impulsive nature of these individuals (Jones & Weiser, 2014). Previous research has also found both narcissism and Machiavellianism to be useful predictors of infidelity (Jones & Weiser, 2014), whereas in the current study narcissism was predictive of only emotional infidelity, and Machiavellianism did not emerge as a significant predictor of either form. The utility for narcissism to predict emotional infidelity could be attributed to the narcissists’ low relationship satisfaction (Casale et al., 2020); a potential product of decreased attention and admiration over time (Gewirtz-Meydan & Finzi-Dottan, 2018). Given that low relationship satisfaction increases the chance of emotional infidelity (Blow & Hartnett, 2005a), and relationship satisfaction has been found to fully mediate the relationship between narcissism and intentions to be unfaithful (Altınok & Kılıç, 2020), logical connections can be drawn such that individuals with high levels of narcissism may be more inclined to be emotionally unfaithful. This result warrants further investigation.

Machiavellianism did not predict perpetration of physical or emotional forms of infidelity, a finding that contradicts previous results (Brewer & Abell, 2015). As Machiavellianism had no effect on either men’s or women’s past acts of infidelity, it appears that infidelity may be better explained by characteristics associated with psychopathy (e.g., thrill-seeking, poor-impulse control; Hare, 2003; Hare & Neumann, 2008) and narcissism (e.g., entitlement, self-importance, grandiosity; Caligor et al., 2015; Chabrol et al., 2009) than the strategic, planned, delayed-gratification nature of Machiavellianism (Geis & Levy, 1970). The utility for psychopathy, not Machiavellianism, to predict both forms of infidelity may also be attributed to the high degree of shared variance between the two traits (Miller et al., 2017). Another possibility could be due to sample characteristics; specifically, previous studies report men score higher on Machiavellianism than women (Lyons et al., 2022), whereas we found no gender difference. We compared our obtained descriptive statistics with previous research (Lyons et al., 2022) and found that Machiavellianism scores were significantly lower in the current sample⁸, particularly for men. It is possible that the low predictive utility for Machiavellianism is attributed to low levels of the trait in our sample.

Importantly, the differential pattern of results regarding psychopathy, narcissism, Machiavellianism, and physical and emotional forms of infidelity support the dimensional conceptualisation and operationalisation of infidelity. We speculate that these results help address inconsistencies in previous research exploring these personality traits and infidelity. Specifically, Sevi et al. (2020), who explored the Dark Triad and infidelity, noted the disparity between their own obtained results (i.e., only trait psychopathy was a significant predictor of infidelity) and the results of Brewer et al. (2015) who studied the Dark Triad and infidelity and found trait narcissism predicted infidelity. Given that, in the current study, trait narcissism was found to predict emotional, but not physical, infidelity, we suggest that previous discrepancies in findings might stem from the tendency to operationalise infidelity as a unidimensional construct.

Sadism also showed no unique utility in predicting physical and emotional infidelity, suggesting that the Dark Tetrad model offers no advantage over the Dark Triad model when exploring infidelity perpetration. However, the Dark Tetrad may offer distinct advantage over the Dark Triad when exploring infidelity motivations, as individuals with high sadism showed greater propensity to engage in malevolent infidelity. Further to sadism, those with high psychopathy were also more inclined to engage in malevolent infidelity. Conceptually, the finding that only sadism and psychopathy predict malevolent infidelity intentions may appear at odds with the Dark Tetrad model, as all traits have been associated with cruelty, malevolence, poor emotional bonds, and a lack of concern with morality (Baughman et al., 2014; Buckels et al., 2013; Crysel et al., 2013). It is not particularly surprising that the two traits best characterised by callousness and enjoyment of harming others (i.e., psychopathy and sadism, respectively) predict motivations to engage in infidelity as a form of harmful revenge. However, it is curious that trait narcissism and Machiavellianism were not significant predictors. Both narcissism and Machiavellianism were significantly associated with malevolent infidelity in bivariate correlations, but were not significant predictors in the regression model. The most parsimonious explanation for this appears to be the shared variance of dark personality traits, a significant portion of which is explained by callousness (Book et al., 2015; 2016). In the presence of psychopathy and sadism, Machiavellianism and narcissism did not add predictive value to the model beyond that explained by their shared variance with psychopathy and sadism. A similar case could be made for the results of the physical (and to a lesser extent emotional) infidelity models.

Gender as a moderator of the Dark Tetrad and infidelity

Moderation analyses showed that at high levels of sadism, men engaged in more physical infidelity than women. Although speculative, one possibility for this result could be attributed to men’s own reaction to physical, not emotional, infidelity. Specifically, as physical infidelity is more distressing for men than women (Kruger et al., 2015), men who enjoy harming others (i.e., high sadism) may be more inclined to be physically unfaithful, as they believe this will cause their partner greater distress. Future research might consider exploring this possibility with a serial mediation analysis, where the motivation to perpetrate infidelity is entered as a mediator of the relationship between sadism and physical infidelity.

Moderation analyses also revealed that for low levels of narcissism, men had lower malevolent infidelity intentions compared to women. Again, although interpretation is speculative, one possibility for this result could be attributed to dimensions of narcissism (i.e., grandiose and vulnerable; Miller et al., 2011). Women, compared to men, score higher on vulnerable narcissism (Green et al., 2020), and vulnerable narcissism is a correlate of aggression (Ryan et al., 2008) and abuse (March et al., 2021) in intimate relationships. Future research exploring gender as a moderator of narcissism and infidelity should seek to operationalise and measure narcissism as dimensional (i.e., grandiose and vulnerable; Miller et al., 2011), not unidimensional.

Limitations and future directions

There are several limitations of the current study that warrant consideration. Although we established psychopathy as a predictor of physical and emotional infidelity, and the motivation to engage in malevolent infidelity, our study is limited by not measuring the dimensions of psychopathy that could explain these results. Previous research exploring the Dark Triad and infidelity (Brewer et al., 2015) differentiated between, and assessed, primary psychopathy (e.g., callous, manipulative, deceptive) and secondary psychopathy (e.g., antisocial, impulsive, neurotic). By not differentiating between these forms of psychopathy, we are unable to determine if these different forms of infidelity are best attributed to callousness and manipulation, or antisociality and neuroticism. We recommend future researchers exploring psychopathy and infidelity conceptualise and operationalise psychopathy as dimensional. It is also worth noting that the Dark Tetrad traits account for only relatively small portions of variance in emotional and physical infidelity. As such, future research exploring infidelity could consider other factors not investigated here, including sociosexual orientation (Ali & Chamorro-Premuzic, 2010), sexual satisfaction (Buss & Shackelford, 1997), and relationship satisfaction (Atkins et al., 2001).

The current sample may be somewhat limited in generalisability, given the smaller sample size (N = 240) and predominantly female sample (75.8%). Nevertheless, we suggest that despite these sample limitations, the current study provides novel insight regarding personality and infidelity, demonstrating for the first time that trait narcissism may only relate to emotional forms of infidelity. Further, the exploration of the Dark Tetrad traits and motivations to engage in infidelity is novel, contributing to the extant literature on infidelity and providing interesting directions for future infidelity research.

Implications and conclusions

The current study was, to the best of our knowledge, the first to assess the efficacy of gender and the Dark Tetrad as predictors of physical and emotional infidelity, and the intention to perpetrate infidelity as a means of revenge and causing one’s partner harm (i.e., malevolent infidelity). These results have important therapeutic implications for practice with couples seeking help to overcome partner infidelity, as therapists may benefit from an increased awareness of the motivations for, and contributors to, extradyadic relationships. For example, engaging in physical and emotional infidelity might best be explained by thrill-seeking and impulsivity (i.e., psychopathy), with emotional infidelity also explained by superiority and egoism (i.e., narcissism). Lastly, thrill-seeking and impulsivity (i.e., psychopathy) and enjoying harming others (i.e., sadism) might best explain intentions to be unfaithful as a means of revenge and causing a partner harm. Screening for these factors in couples seeking relationship counselling around issues of fidelity may assist therapists in deciding on the goals of any intervention. Importantly, results of the current study add to the extant body of infidelity research by demonstrating differential patterns of predictors of physical and emotional infidelity, emphasising the need for future research to differentiate these forms.

Ethical statement

The study was approved by the relevant institution Human Research Ethics Committee. We certify that the study was performed in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards.

Informed consent

Informed consent was obtained from all individual participants included in the study.

Disclosure statement

No potential conflict of interest was reported by the authors.

Data availability statement

The data are not publicly available due to ethical approval restrictions.

Notes

1 There was no gender difference in age, t(211) = -0.62, p = .539.

2 A defacto relationship status is when couples are not married but are in a relationship, and living (or cohabiting) together.

3 There was no gender difference in relationship status, χ²(4) = 6.08, p = .194.

4 We interpret the coefficient with caution, mindful of the reliability limitations for only two items.

5 Participants who identified their gender as man or woman were included in analyses.

6 Conditional effects are one SD below the mean, the mean, and one SD above the mean.

7 An online dating platform for married people seeking extradyadic relations

8 Independent measures t-tests: Total group, t(432) = 2.74, p = .006; Men t(128) = 3.09, p = .003; Women, t(328) = 1.21, p = .227