Male Tolerance to Same-Sex Infidelity: A Cross-Cultural Investigation

Abstract

There is accumulating evidence that heterosexual men exhibit tolerance to their partners’ same-sex infidelity. The current study examined such tolerance in the Chinese (N = 949) and the British (N = 305) cultural contexts. Consistent with the predictions derived from an evolutionary framework, across different cultural settings, men exhibited higher tolerance than women to their partners’ same-sex infidelity. In addition, if they had to choose, men were considerably more likely than women to prefer their partners to cheat with an individual of the same than of the opposite sex. Participants were also more tolerant of infidelity involving their short-term than their long-term partners. Moreover, men who preferred same-sex attraction in women were more tolerant to the same-sex infidelity of their female partners than men who did not share these preferences. Finally, men and women who experienced same-sex attractions indicated higher tolerance to infidelity. The implications of these finding for the evolution of same-sex attraction in women were further discussed.

Infidelity is a widespread phenomenon across human cultures (Allen & Baucom, 2006; Tafoya & Spitzberg, 2007). People usually react harshly to their partners’ actual or suspected infidelity, with the termination of the relationship or physical violence against the culprit being common outcomes (Buss, 2000). Still, there has been accumulating evidence suggesting that men do exhibit tolerance to the same-sex infidelity of their female partners (Apostolou, 2018; Compton & Bowman, 2017; Confer & Cloud, 2011; Sagarin, Becker, Guadagno, Nicastle, & Millevoi, 2003; Wiederman & LaMar, 1998). That is to say, if their female partners cheat with other women, men react much less negatively than if their partners cheat with other men. The current research aims to explore such tolerance in different cultural settings and to test specific predictions derived from an evolutionary framework. This framework will be discussed next.

Male Tolerance to Same-Sex Infidelity

Infidelity can compromise people’s reproductive success or fitness considerably, as they risk losing their partners or their partner’s investment to others. It can also result in men raising other people’s children as their own without being aware of it. The potential costs of infidelity would translate into strong selection pressures exercised on people to evolve adaptations that enable them to reduce these costs (Buss, 1996, 2000). Such adaptations include emotions like jealousy, anger, and sadness, which are triggered when clues of infidelity are present, enabling corrective action to be taken (Buss, 2000). For instance, people may terminate a relationship with a mate who is unfaithful in order to avoid experiencing sadness.

Yet infidelity of one’s partner with a same-sex partner is considerably less harmful to the fitness of men than infidelity involving an opposite-sex partner. To begin with, during human evolutionary time, there was no reliable test of paternity, which means that men were facing the risk of being cuckolded—diverting their parental investment to other men’s children who believed that they were their own. If a man’s female partner cheats with another man, doing so increases substantially his risk of being cuckolded. However, if she cheats with another woman, this risk will not increase.

Same-sex infidelity, apart from being less costly, it could potentially be more beneficial for men than for women. A woman engaging in extra-pair relationships may be willing to include other women in the couple’s sex, allowing him in effect direct access to gain sexual access to other women. In this situation, a man not only decreases the risk of being cuckolded, but he increases his reproductive success by gaining access to additional women (Apostolou, Shialos, Khalil, & Paschali, 2017).

In effect, because same-sex infidelity is much less costly and potentially beneficial for men, we would expect that selection forces would adjust the adaptations responsible for generating negative reactions, so that men would react much less negatively to infidelity involving same-sex extra-pair partners than one involving opposite-sex extra-pair partners. On the other hand, in women, there is no reduction in parental uncertainty in the case of same-sex infidelity, as they are always certain that their children are their own. Accordingly, same-sex infidelity is associated with a lower risk of cuckoldry in men but not in women which, in turn, predicts that men would be much more tolerant—that is, they would exhibit less negative reactions to same- than to opposite-sex infidelity. Several studies have produced evidence in support of this prediction.

To begin with, one study asked college students to indicate how upset they would feel if they have found out that their partners had a one-time sexual encounter with an individual of the opposite or of the same sex (Wiederman & LaMar, 1998). The results indicated that, in the case of the same-sex infidelity, men were considerably less upset than women. Similarly, in a different study, men indicated that they would be more likely to terminate an imagined long-term relationship following a partner’s heterosexual affair in comparison to a homosexual affair (Confer & Cloud, 2011). Another study found that men reported significantly less jealousy in relation to their partners’ same-sex than opposite-sex infidelity (Sagarin et al., 2003). Moreover, Apostolou (2016) found that men indicated less anger than women when their long-term partners revealed to them that they had same-sex sexual contacts in the past. Compton and Bowman (2017) employed a sample of 202 undergraduate students in the United States and presented them with a scenario of a heterosexual female individual who had a lesbian best friend. Subsequently, the participants were asked to indicate whether several erotic behaviors between the two, such as oral sex, would be considered as cheating by the heterosexual woman’s male partner. They found that men were less likely than women to perceive these behaviors as cheating.

Nonetheless, not all studies have found higher tolerance in men as compared to women in relation to their partners’ same-sex infidelity. More specifically, Brewer (2014) presented participants with vignettes describing a hypothetical situation where their partner had a heterosexual or homosexual affair and found that male participants did not indicate less distress than female participants in the case of the homosexual affair. Denes, Lannutti, and Bevan (2015) used an online sample of 285 participants in the United States and assessed four emotional responses, namely, anger, hurt, upset, and fear, as well as willingness to terminate a relationship in the scenario involving same-sex and opposite-sex infidelity, respectively. They found that men and women experienced similar negative emotions across the two scenarios, but men were less likely to terminate the relationship in the case of same-sex infidelity. It needs to be said, however, that this study included only 68 men, and consequently, it lacked the necessary power to detect sex differences in emotional responses.

In a more comprehensive research on male tolerance, Apostolou (2018) asked participants to rate five different negative reactions, including anger and terminating the relationship in scenarios involving either same- or opposite-sex infidelity. It was found that men exhibited considerably higher tolerance—that is, they indicated weaker negative reactions—to their partners’ same-sex infidelity. In addition, men and women exhibited higher tolerance to same-sex infidelity involving their long-term than of their short-term partners, with men exhibiting higher tolerance in the latter case. Furthermore, in this study, participants were asked to choose whether they would prefer their partners to cheat with a same- or with an opposite-sex partner. It was found that, about 70% of men but only about 35% of women preferred the former option. The current work aims to advance this line of research by replicating these findings in different samples and testing additional predictions.

Predictions to Be Tested

On the basis of the above theoretical framework, we predict that heterosexual men would exhibit higher tolerance than women to same-sex than to opposite-sex infidelity (Hypothesis 1). Moreover, it has been argued that, as short-term relationships are less committed than long-term ones, people would be more tolerant to infidelity in the former case (Apostolou, 2018). Thus, we predict that men and women would be more tolerant to the same-sex as well as to the opposite-sex infidelity of their short- than of their long-term partners (Hypothesis 2). Also, if choice is possible, men would be more likely than women to prefer their partners to cheat with an individual of the same than one of the opposite sex (Hypothesis 3).

Furthermore, people value similarity in a partner (Buss, 2017; Figueredo, Sefcek, & Jones, 2006; Wang & Apostolou, 2017). Thus, if individuals experience same-sex attractions, they would see the same-sex extra-pair contacts of their partners as a sign of similarity and would likely be more positive about it, resulting in greater tolerance to such infidelity. If this is the case, heterosexual men and women who experience same-sex attractions are expected to be more tolerant to the same-sex infidelity of their opposite-sex partners than men and women who do not experience such attractions. On the other hand, individuals with same-sex attractions are not expected to be more tolerant to their partners’ opposite-sex infidelity than people without such attractions. On this basis, we predict that there would be a significant interaction between sexual attraction and the sex of the extra-pair partner, with people experiencing same-sex attractions exhibiting more tolerance to the same- than to the opposite-sex infidelity (Hypothesis 4).

In addition, men are not constrained by the burden of pregnancy, and they can increase their reproductive success considerably by having sex with different women (Buss, 2017). It has been argued that men could achieve this goal by preferring same-sex attraction and contact in an opposite-sex partner (Apostolou et al., 2017). Men who have female partners who are attracted to other women are likely to enjoy access to additional women, that is, their partners’ female mates. Evidence from several studies indicated that a considerable proportion of men desired as partners women who experienced same-sex attractions and they also liked their female partners to have sex with other women (Apostolou & Christoforou, 2018; Apostolou et al., 2017; Apostolou, Wang, & Jiaqing, 2018). In order to follow such a strategy—that is, increasing access to women by having female partners who have sex with other women—men need not react negatively to same-sex infidelity committed by their partners. In this respect, weak negative reactions constitute one facet of the male strategy of having access to multiple women through having female partners who like women. On this basis, it can be predicted that heterosexual men who prefer same-sex attraction and contact in an opposite-sex partner would exhibit a higher tolerance to same-sex infidelity than men who do not have this preference (Hypothesis 5).

Finally, in the proposed theoretical framework, same-sex infidelity is less costly and in certain cases beneficial, not in a specific place and time, but across different settings. For instance, across different cultures and different time periods, because men do not give birth to their children, they face the issue of paternal uncertainty; therefore, the same-sex infidelity of their partners is less costly for them than the opposite-sex infidelity. On this basis, we predict that a sex difference in tolerance with men exhibiting higher tolerance than women to the same-sex infidelity would be found consistently across different cultures (Hypothesis 6).

Study 1

The purpose of this study was to investigate tolerance to same-sex infidelity in the Chinese cultural context.

Method

Participants

Participants completed an online survey. Given the sensitive nature of the study, in order to get more honest answers, we preferred this method to the paper-and-pencil one. The link of the survey was forwarded to university students and employees in different academic disciplines. Each participant received a small monetary payment after the completion of the survey (i.e., the amount was allocated randomly from 2.0¥ to 10.0¥; $0.32–$3.3).

Overall, 1,021 Chinese men and women took part in the study. With respect to sexual attraction, 79.2% of the participants indicated that they were attracted exclusively to opposite-sex individuals, 13.7% were attracted predominantly to opposite-sex individuals but occasionally to same-sex individuals, 2.9% indicated that they were attracted to both sexes equally, 1.5% indicated that they were predominantly attracted to same-sex but occasionally to opposite-sex individuals, while 2.6% have reported being exclusively attracted to same-sex individuals. Given that the study was designed to assess the preferences of heterosexual people, only the responses of participants who indicated that they were heterosexual were included. Accordingly, data from 949 heterosexual Chinese participants (543 women, 406 men) were included in the analysis. The mean age of women was 24.19 (SD = 4.18), and the mean age of men was 24.90 (SD = 5.48).

Finally, in order to test our hypotheses (see below), we have also employed data from other studies. More specifically, we have employed data from 305 heterosexual British participants (193 women, 112 men; Apostolou et al., 2018) and data from 590 Greek-speaking heterosexual participants (241 men and 349 women; Apostolou et al., 2017).

Materials

The survey was in Chinese and was constructed using the software “sojump,” (Version 2.0.42) which is a widely used tool in China for collecting data online. It consisted of three parts. The first part had four sections. In the first section, participants were given the scenario of finding out that their long-term partner had sexual contact with an opposite-sex individual. In the second section, participants were given the same scenario, with the difference being that their partners had sexual contact with a same-sex individual. In the third section, participants were given the same scenario, with the difference being that they were in a short-term relationship, and their partner have had sex with an opposite-sex individual. In the fourth section, participants were given the scenario of being in a short-term relationship and their partner having had sexual contact with an individual of the same sex.

In each section, participants were asked to rate different reactions, using a 5-point Likert-type scale ranging from 1 = totally disagree to 5 = totally agree. The five reactions, originally used by Apostolou (2018), were “I would get angry,” “I would get jealous,” “I would break up with him or her,” “I would become very sad,” and the “I would consider that he or she has cheated on me.” The order of presentation of each section and of the five reactions in each section was randomized across participants.

In the second part, participants were given the statement “If I had to choose, I would prefer for my partner to cheat with” and were asked to choose one of the two options as a response: “An individual of the same-sex” or “An individual of the opposite sex.” The order of presentation of the two options was randomized across participants. In the third part of the survey, demographic information was collected (i.e., sex, age, marital status, sexual attraction). Note that the order of presentation of the first two parts was randomized across participants.

Finally, if tolerance to same-sex infidelity constitutes one facet of the male preference for same-sex attraction and contact, this male preference should predict then high tolerance to same-sex infidelity. The data presented in the current study were part of a larger study of sexual behavior, which included measurement of preference for same-sex attraction and contact (Apostolou et al., 2018). More specifically, participants were asked about the sexual attraction of their long-term and short-term mates and how frequently they have wanted their long-term and short-term partners to have sexual contact with same-sex individuals. Accordingly, we employed these responses in order to test the prediction that heterosexual men who preferred same-sex attraction and contact in an opposite-sex partner would exhibit a higher tolerance to same-sex infidelity than men who did not have such preferences.

Statistical Analysis

In order to identify significant effects and interactions, we ran a doubly multivariate analysis—a statistical tool which is similar to multivariate analysis of variance (MANOVA), but which allows within-subjects independent variables to be entered. The five negative reactions were entered as the dependent variables, participants’ sex and sexual attraction (exclusively heterosexual/heterosexual with same-sex attractions) were entered as the between-subjects independent factors, and the sex of the extra-pair partner (man, woman) and the seriousness of the relationship (long-term/short-term) were entered as the within-subjects independent factors. Finally, participant’s age was entered as a covariate.

In order to test the prediction that men who preferred same-sex attraction and contact in a partner would also exhibit a higher tolerance toward same-sex infidelity, the following analysis was used: A series of MANOVA tests was applied where the five negative reactions to same-sex infidelity were entered as the dependent variable and the preference for same-sex attraction and contact were entered as the independent variables. For the independent variables, data from Apostolou, Wang, and Jiaqing’s (2018) study were used. More specifically, we ran a test where the dependent variables were the five reactions to the same-sex infidelity of a long-term partner and the independent variable was the participants’ preferences for same-sex attractions in a long-term partner (i.e., a long-term partner’s sexual attraction). The analysis was repeated with the independent variable being the preference for same-sex contact for a long-term partner on this occasion and the dependent variables being the reactions to the same-sex infidelity of a long-term partner. Finally, this analysis was repeated with the preference for the short-term partner scenario.

With respect to the question, “If I had to choose, I would prefer for my partner to cheat on me with,” binary logistic regression was performed where participants’ responses were entered as the dependent variable, and the participants’ sex and sexual attraction were entered as the independent variables. The age was also entered as a covariate.

In order to estimate cultural differences in tolerance to same-sex infidelity, we made statistical comparisons across the Chinese, the British, and Greek samples. The data for the Greek sample came from Apostolou, Shialos, Khalil, and Paschali’s (2017) study. Accordingly, we ran a doubly multivariate analysis, where the five negative reactions to a partner’s same-sex infidelity were entered as the dependent variables, participants’ sex and sexual attraction as well as the sample (Chinese, British, and Greek) were entered as the between-subjects independent factors, and the seriousness of the relationship (long-term/short-term) was entered as the within-subjects independent factor. Finally, participant’s age was entered as a covariate.

In order to estimate cultural difference in responses to the question “If I had to choose, I would prefer for my partner to cheat on me with,” binary logistic regression was performed, where participants’ responses were entered as the dependent variable, and the participants’ sex, sexual attraction, and sample were entered as the independent variables. The age was also entered as a covariate. For the statistical analysis, we have employed the IBM’s SPPS Version 25.

Results

In order to examine the degree of tolerance to same-sex infidelity, we estimated the frequency rates for participants’ responses for each item of the survey. As shown in Table 1, about 33% of men and 22% of women gave scores of “1” or “2,” suggesting low negative reactions to their long-term partners’ same-sex infidelity, with the respective percentages being about 38% and 26% for their short-term partner as a comparison.

Table 1.

The Frequencies of Participants’ Responses on Their Reactions to Their Partners’ Infidelity With an Individual of the Same Sex in Study 1.

Tolerance	Long-Term Relationship						Short-Term Relationship
	Men			Women			Men			Women
	1–2 (%)	3 (%)	4–5 (%)	1–2 (%)	3 (%)	4–5 (%)	1–2 (%)	3 (%)	4–5 (%)	1–2 (%)	3 (%)	4–5 (%)
I would consider that he or she has cheated on me	28.6	15.3	56.1	17.1	7.7	75.1	36.5	18	45.5	20.3	13.4	66.3
I would break up with him or her	32.8	18.7	48.5	14.2	9.4	76.4	36.9	18.7	44.3	17.1	8.5	74.4
I would get angry	26.4	11.3	62.3	16.9	6.6	76.5	31.8	18.2	50	20.6	12.7	66.7
I would get jealous	43.6	14.5	41.9	43.8	13.3	42.9	47	17.2	35.7	46.6	17.5	35.9
I would become very sad	31.8	15	53.2	17.9	11.2	70.9	38.2	18.2	43.6	27.1	16.4	56.5
Mean percentages	32.6	15	52.4	22	9.6	68.4	38.1	18.1	43.8	26.3	13.7	60

Note. The percentages reflect participants scores in the 5-point scale 1 = totally disagree to 5 = totally agree.

To further study male participants’ attitudes toward same-sex infidelity, we classified their responses to five categories: tolerant (i.e., those who gave scores of “1” or “2” for both long-term and short-term partners), intolerant (i.e., those who gave scores of “4” or “5” for both long-term and short-term partners), tolerant-short (i.e., those who gave scores of “1” or “2” for short-term partners, but “3” or more for long-term partners), tolerant-long (i.e., those who gave scores of “1” or “2” for long-term partners but “3” or more for short-term partners), and other (i.e., any other combination which was not included in these categories). As noted in Table 2, nearly 40% of men indicated above average negative reactions to same-sex infidelity involving both their long- and short-term partners. Thus, about 60% of men exhibited moderate or high tolerance to the same-sex infidelity of their partners. The most common strategy was being tolerant to same-sex infidelity involving both long- and short-term partners, with more than one in four men falling into this category.

Table 2.

Male Strategies to Same-Sex Infidelity in Study 1.

Tolerance	Strategies
	Tolerant (%)	Intolerant (%)	Tolerant-Short (%)	Tolerant-Long (%)
I would consider that he or she has cheated on me	23.6	40.6	12.8	4.9
I would break up with him or her	25.4	37.2	11.6	7.4
I would get angry	21.4	47.5	10.3	4.9
I would get jealous	36.7	31.8	10.3	3.4
I would become very sad	25.4	37.5	12.8	6.4
Mean percentages	26.5	38.9	11.6	5.4

Responses to the question “If I had to choose, I would prefer for my partner to cheat on me with” indicated that 76.4% of men would prefer their partner to cheat on them with a same-sex individual and 23.6% with an opposite-sex individual, while 37.4% of women indicated that they would prefer their partner to cheat on them with a same-sex and 62.6% with an opposite-sex individual.

From Table 3, we can see that there was a significant interaction between the sex of the participant and the sex of the extra-pair partner. In particular, as indicated by the means, when we moved from opposite-sex to same-sex infidelity, the negative reactions of men decreased more than the negative reactions of women. Moreover, as we can see from Table 4, there was a significant main effect of the seriousness of the relationship, with participants exhibiting more intense negative reactions to the infidelity of their long-term partners than of their short-term ones. Finally, there was a significant main effect of sexual attraction, with exclusively heterosexual participants indicating more adverse negative reactions than heterosexual participants who also experienced same-sex attractions.

Table 3.

Interaction Effects From Doubly Multivariate Analysis in Participants’ Reactions to Their Partners’ Infidelity in Study 1.

Tolerance	Partner’s Sex				Sex × Partner’s Sex
	Men		Women
	Opposite Sex	Same Sex	Opposite Sex	Same Sex
	Mean (SD)	Mean (SD)	Mean (SD)	Mean (SD)	p Value	${\eta }_{\text{p}}^2$
Total					<.001	.030
I would consider that he or she has cheated on me	3.83 (1.38)	3.33 (1.45)	4.07 (1.31)	3.98 (1.36)	<.001	.016
I would break up with him or her	3.88 (1.38)	3.28 (1.49)	4.12 (1.30)	4.13 (1.32)	<.001	.030
I would get angry	3.97 (1.45)	3.51 (1.48)	4.19 (1.31)	4.00 (1.38)	.006	.008
I would get jealous	3.36 (1.57)	2.87 (1.51)	3.44 (1.54)	2.91 (1.57)	.267	.001
I would become very sad	3.77 (1.42)	3.24 (1.46)	3.99 (1.29)	3.76 (1.40)	.010	.007

Table 4.

Main Effects From Doubly Multivariate Analysis in Participants’ Reactions to Their Partners’ Infidelity With an Individual of the Same Sex in Study 1.

Tolerance	Seriousness of Relationship				Sexual Attraction
	Long Term	Short Term			No Same-Sex Attractions	With Same-Sex Attractions
	Mean (SD)	Mean (SD)	p Value	${\eta }_{\text{p}}^2$	Mean (SD)	Mean (SD)	p Value	${\eta }_{\text{p}}^2$
Total			.001	.024			.053	.012
I would consider that he or she has cheated on me	3.84 (1.52)	3.57 (1.55)	<.001	.014	3.72 (1.46)	3.61 (1.33)	.002	.010
I would break up with him or her	3.82 (1.53)	3.71 (1.56)	.150	.002	3.78 (1.48)	3.68 (1.31)	.028	.005
I would get angry	3.93 (1.52)	3.64 (1.55)	.002	.010	3.81 (1.46)	3.67 (1.32)	.151	.002
I would get jealous	2.99 (1.68)	2.80 (1.62)	.026	.006	2.89 (1.56)	2.90 (1.45)	.367	.001
I would become very sad	3.71 (1.53)	3.36 (1.57)	<.001	.021	3.55 (1.47)	3.45 (1.27)	.196	.002

We proceeded to examine whether the preferences for same-sex attraction and contact predicted tolerance. From Table 5, we can see that for the sexual attraction of a long-term partner, the effect approached but did not pass the significant level, whereas for the remaining comparisons, the effects were significant with moderate to large effect sizes. In all cases, men who exhibited preferences for having as a partner a women who experienced same-sex attractions and had sex with other women indicated higher tolerance to the same-sex infidelity of their partners than men who did not share these preferences.

Table 5.

Preference for Same-Sex Attraction and Contact Predicting Tolerance to Same-Sex Infidelity Using Multivariate Analysis of Variance Analysis in Study 1.

Tolerance	Sexual Attraction (Long Term)		Sexual Attraction (Short Term)		Same-Sex Contact (Long Term)		Same-Sex Contact (Short Term)
	p Value	${\eta }_{\text{p}}^2$	p Value	${\eta }_{\text{p}}^2$	p Value	${\eta }_{\text{p}}^2$	p Value	${\eta }_{\text{p}}^2$
Total	.060	.026	.001	.050	<.001	.057	<.001	.102
I would consider that he or she has cheated on me	.018	.014	<.001	.039	<.001	.040	<.001	.046
I would break up with him or her	.002	.024	<.001	.034	<.001	.055	<.001	.091
I would get angry	.053	.009	.002	.024	.001	.028	<.001	.039
I would get jealous	.581	.001	.330	.002	.170	.005	.523	.001
I would become very sad	.090	.007	.019	.013	.002	.024	<.001	.041

In regard to the question “If I had to choose, I would prefer for my partner to cheat on me with,” binary logistic regression results indicated a significant effect of the participants’ sex, χ²(1, N = 891) = 166.87, p < .001, with men being more likely than women (odds ratio [OR] = 6.9) to answer that they would prefer their opposite-sex partner to cheat with an individual of the same sex than of the opposite sex. In addition, the results indicated a significant effect of the participants’ sexual attraction, χ²(1, N = 891) = 23.72, p < .001, with exclusively heterosexual participants being more likely than heterosexual with same-sex attractions participants (OR = 2.8) to prefer their opposite-sex partner to cheat with an individual of the opposite-sex than with one of the same-sex. Finally, age was not significant, and no significant interactions were found.

Study 2

The purpose of this study was to investigate tolerance to same-sex infidelity in the British cultural context.

Method

Participants

A total of 390 participants completed an online survey. E-mails with the link to the survey embedded were sent to various departments across several universities in the United Kingdom to solicit participation, while the study was also advertised on the subject pool system and on the weekly bulletin of a British University. Participants had to read key information pertaining to the study and then to make an informed decision to partake in it. With respect to sexual attraction, 54.6% of the participants indicated that they were attracted exclusively to opposite-sex individuals, 23.6% that they were attracted predominantly to opposite-sex individuals but occasionally to same-sex individuals, 11.3% indicated that they were attracted to both sexes equally, and 3.8% indicated that they were attracted predominantly to members of the same sex but also occasionally to those of the opposite sex, while 6.7% reported being exclusively attracted to same-sex individuals.

Because the study was designed to assess the preferences of heterosexual people, only the responses of participants who indicated that they were heterosexual were included in the analyses. Thus, data from a total of 305 heterosexual British people who have taken part in this research (193 women, 112 men) were included in the analyses. The mean age of women was 27.15 (SD = 11.01), and the mean age of men was 24.86 (SD = 9.69).

Materials

The content and the structure of the survey were identical to Study 1, but the items were in English instead of Chinese.

Statistical Analysis

The statistical analysis employed in Study 2 was similar to the statistical analysis employed in Study 1.

Results

In order to examine the degree of tolerance for same-sex infidelity, we calculated the frequencies of participants’ responses for each item of the survey. As indicated in Table 6, about 15% of men and 6% of women gave scores of “1” or “2,” suggesting low negative reactions to their long-term partners’ same-sex infidelity, with the respective percentages being about 27% and 15% for their short-term partner.

Table 6.

The Frequencies of Participants’ Responses on Their Reactions to Their Partners’ Infidelity With an Individual of the Same Sex.

Tolerance	Long-Term Relationship						Short-Term Relationship
	Men			Women			Men			Women
	1–2 (%)	3 (%)	4–5 (%)	1–2 (%)	3 (%)	4–5 (%)	1–2 (%)	3 (%)	4–5 (%)	1–2 (%)	3 (%)	4–5 (%)
I would consider that he or she has cheated on me	4.5	20.5	75	1	9.3	90.7	16.1	24.1	59.9	11.4	14	74.6
I would break up with him or her	18.8	31.3	50	3.1	19.7	77.2	23.2	38.4	38.4	9.8	20.7	69.5
I would get angry	13.4	17.9	68.7	2.1	11.4	86.6	25	25	50	8.8	15	76.1
I would get jealous	23.2	17.9	58.9	20.2	15	64.8	33.9	20.5	45.6	24.9	14.5	60.6
I would become very sad	13.4	24.1	62.6	2.1	9.8	90.1	34.8	29.5	35.7	20.7	16.1	63.2
Mean percentages	14.7	22.3	63	5.7	13	81.9	26.6	27.5	45.9	15.1	16.1	68.8

Note. The percentages reflect participants scores in the 5-point scale 1 = totally disagree to 5 = totally agree.

In order to study male participants’ reactions toward same-sex infidelity further, we classified their responses in five categories as in Study 1. In Table 7, we can see that about 58% of men indicated moderate or high tolerance to their partners’ same-sex infidelity. Being tolerant to the same-sex infidelity of both long-term and short-term partners and to be tolerant only to the same-sex infidelity of the short-term partners but intolerant to the infidelity of the long-term partners were common strategies.

Table 7.

Male Strategies to Same-Sex Infidelity in Study 2.

Tolerance	Strategies
	Tolerant (%)	Intolerant (%)	Tolerant-Short (%)	Tolerant-Long (%)
I would consider that he or she has cheated on me	3.6	55.4	12.5	0.9
I would break up with him or her	14.3	32.1	8.9	4.5
I would get angry	11.6	48.2	13.4	1.8
I would get jealous	19.6	41.1	14.3	1.8
I would become very sad	11.6	33.9	23.2	1.8
Mean percentages	12.1	42.1	12.5	2.2

Moreover, 79.3% of men indicated that they would prefer their partner to cheat with a same-sex individual and 20.7% with an opposite-sex individual, when asked to respond to the question “If I had to choose, I would prefer for my partner to cheat on me with.” Finally, 50.5% of women indicated that they would prefer their partner to cheat with a same-sex individual while 49.5% would prefer that to be an opposite-sex individual.

In order to identify significant effects and interactions, we ran a doubly multivariate analysis as in Study 1. From Table 8, we can see that there was a significant interaction between the sex of the participant and the sex of the extra-pair partner. In particular, when we moved from opposite-sex to same-sex infidelity, the negative reactions of men decreased more than the negative reactions of women. In addition, from Table 9, we can see that there was a significant main effect of the seriousness of the relationship, with participants’ exhibiting stronger negative reactions to the infidelity of their long-term partners than of their short-term ones. There was no interaction between the seriousness of the relationship and the sex of the extra-pair partner, indicating that this effect did not depend on the sex of the extra-pair partner. Finally, there was a significant main effect of sexual attraction, with exclusively heterosexual participants indicating that they would be more likely to terminate the relationship, but they would feel less jealous than participants with same-sex attractions.

Table 8.

Interaction Effects From Doubly Multivariate Analysis in Participants’ Reactions to Their Partners’ Infidelity in Study 2.

Tolerance	Partner’s Sex				Sex × Partner’s Sex
	Men		Women
	Opposite Sex	Same Sex	Opposite Sex	Same Sex
	Mean (SD)	Mean (SD)	Mean (SD)	Mean (SD)	p Value	${\eta }_{\text{p}}^2$
Total					<.001	.092
I would consider that he or she has cheated on me	4.20 (0.90)	3.95 (4.37)	4.50 (0.62)	4.37 (0.78)	.226	.005
I would break up with him or her	3.80 (1.05)	3.39 (1.09)	4.13 (0.78)	4.13 (0.85)	<.001	.078
I would get angry	3.91 (1.04)	3.59 (1.06)	4.45 (0.66)	4.28 (0.82)	.033	.015
I would get jealous	3.69 (1.12)	3.32 (1.16)	4.14 (0.88)	3.65 (1.22)	.266	.004
I would become very sad	3.68 (1.08)	3.37 (1.09)	4.23 (0.72)	4.09 (0.85)	.121	.008

Table 9.

Main and Interaction Effects From Doubly Multivariate Analysis in Participants’ Reactions to Their Partners’ Infidelity With an Individual of the Same Sex in Study 2.

Tolerance	Seriousness of Relationship				Sexual Attraction
	Long Term	Short Term			No Same-Sex Attractions	With Same-Sex Attractions
	Mean (SD)	Mean (SD)	p Value	${\eta }_{\text{p}}^2$	Mean (SD)	Mean (SD)	p Value	${\eta }_{\text{p}}^2$
Total			<.001	.120			.005	.054
I would consider that he or she has cheated on me	4.49 (0.83)	3.94 (1.16)	<.001	.060	4.21 (0.89)	4.22 (0.80)	.319	.003
I would break up with him or her	3.98 (1.08)	3.73 (1.17)	.016	.019	3.92 (1.02)	3.72 (0.98)	.018	.019
I would get angry	4.66 (0.99)	3.80 (1.14)	<.001	.064	4.03 (1.02)	4.01 (0.87)	.293	.004
I would get jealous	3.65 (1.28)	3.40 (1.30)	.027	.016	3.35 (1.27)	3.91 (0.93)	.004	.027
I would become very sad	4.21 (1.00)	3.44 (1.25)	<.001	.108	3.77 (1.04))	3.94 (0.93)	.784	.000

Moving on to the prediction that men who preferred same-sex attraction and contact in an opposite-sex partner would exhibit a higher tolerance to same-sex infidelity than men who did not have this preference, we ran a series of MANOVAs as in Study 1. The five reactions were entered as the dependent variables and British participants’ preferences for same-sex attraction and contact from Apostolou et al.’s (2018) study as the independent variables. The results are presented in Table 10, where we can see that, in all cases, the preferences for same-sex attraction and contact predicted tolerance to same-sex infidelity. In particular, men who indicated a preference for female partners who experienced same-sex attractions and had sex with other women exhibited higher tolerance to the same-sex infidelity of their partners than men who did not share these preferences.

Table 10.

Preference for Same-Sex Attraction and Contact Predicting Tolerance to Same-Sex Infidelity Using Multivariate Analysis of Variance in Study 2.

Tolerance	Sexual Attraction (Long Term)		Sexual Attraction (Short Term)		Same-Sex Contact (Long Term)		Same-Sex Contact (Short Term)
	p Value	${\eta }_{\text{p}}^2$	p Value	${\eta }_{\text{p}}^2$	p Value	ηp2	p Value	${\eta }_{\text{p}}^2$
Total	.025	.014	.032	.107	.002	.163	<.001	.191
I would consider that he or she has cheated on me	.270	.011	.095	.025	.001	.100	.007	.064
I would break up with him or her	.002	.084	.016	.052	.001	.097	<.001	.165
I would get angry	.012	.056	.002	.081	.005	.068	<.001	.119
I would get jealous	.752	.001	.088	.026	.616	.002	.129	.021
I would become very sad	.080	.028	.285	.010	.009	.061	.002	.084

In addition, with respect to the question “If I had to choose, I would prefer for my partner to cheat on me with,” binary logistic regression results indicated a significant effect of the participants’ sex, χ²(1, N = 299) = 26, p < .001, with men being more likely than women (OR = 3.9) to answer that they would prefer their opposite-sex partner to cheat with an individual of the same sex than with an individual of the opposite sex. In addition, participants’ sexual attraction was not significant, although it approached the significance level (p = .122). Finally, age was not significant, and no significant interactions were found.

Cross-Cultural Comparisons

Turning to cross-cultural comparisons, from Table 11, we can see that the sample variable was significant, with more negative reactions and thus, less tolerance to be exhibited by the British, followed by the Chinese and the Greeks. Post hoc analysis using Bonferroni indicated that there were no significant differences in any of the reactions between the Chinese and the Greek samples, but there were significant differences between these two samples and the British sample. Furthermore, there was a significant interaction between the sample and the seriousness of the relationship. More specifically, the Chinese participants indicated similar reactions to the infidelity of their short-term and of their long-term partners, but the British and the Greeks indicated more negative reactions to the infidelity of their long-term than of their short-term partners.

Table 11.

Comparisons Between the Chinese, British, and Greek Sample Using Doubly Multivariate Analysis.

Tolerance	Sample				Seriousness × Sample
	Chinese	British	Greek
	Mean (SD)	Mean (SD)	Mean (SD)	p Value	${\eta }_{\text{p}}^2$	p Value	${\eta }_{\text{p}}^2$
Total	3.54 (1.24)	3.89 (0.78)	3.45 (1.03)	<.001	.026	<.001	.018
I would consider that he or she has cheated on me	3.70 (1.44)	4.21 (0.86)	3.66 (1.21)	<.001	.020	<.001	.026
I would break up with him or her	3.77 (1.45)	3.85 (1.01)	3.86 (1.27)	ns	ns	<.001	.009
I would get angry	3.79 (1.44)	3.09 (0.97)	3.59 (1.22)	.005	.006	<.001	.013
I would get jealous	2.89 (1.55)	3.53 (1.20)	2.63 (1.57)	<.001	.026	ns	ns
I would become very sad	3.53 (1.44)	3.82 (1.00)	3.56 (1.19)	.011	.005	<.001	.024

With respect to the question “If I had to choose, I would prefer for my partner to cheat on me with,” binary logistic regression results indicated a significant effect of the participants’ sex, χ²(1, N = 1,733) = 247.60, p < .001, with men being more likely than women (OR = 5.4) to answer that they would prefer their opposite-sex partner to cheat with an individual of the same-sex than with an individual of the opposite sex. In addition, the results indicated a significant effect of the participants’ sexual attraction, χ²(1, N = 1,733) = 24.06, p < .001, with exclusively heterosexual participants being more likely than heterosexuals with same-sex attractions (OR = 2.0) to prefer their opposite-sex partner to cheat with an individual of the opposite-sex than with an individual of the same-sex. The sample was also significant, χ²(2, N = 1,733) = 11.89, p = .003, with the Wald statistic indicating one significant difference (p = .017), namely, that the British participants were more likely than the Chinese (OR = 1.4) to prefer their opposite-sex partner to cheat with an individual of the same-sex than with an individual of the opposite-sex. Age was not a significant predictor in this analysis.

Further investigations revealed a significant interaction between participants’ sex and the sample, χ²(3, N = 1,733) = 253.30, p < .001. To better understand this interaction, we ran the logistic regression separately for men and for women. The results indicated that Greek men were more likely than Chinese men (OR = 1.5) to prefer their partners to cheat with an individual of the opposite sex than with an individual of the same sex. In addition, British women were more likely than Chinese women (OR = 1.6) to indicate that they would prefer their partners to cheat with an individual of the same than with an individual of the opposite sex.

Discussion

Consistent with our original predictions, across different cultural settings, men exhibited higher tolerance than women to their partners’ same-sex infidelity. In addition, given a choice, men were considerably more likely than women to prefer their partners to cheat with an individual of the same than one of the opposite sex. Participants were also more tolerant to the infidelity of their short-term than of their long-term partners. Moreover, men who preferred same-sex attraction and contact in women were more tolerant to their female partners’ same-sex infidelity than men who did not share these preferences. Finally, men and women who experienced same-sex attractions exhibited higher tolerance to their partners’ infidelity.

Our original prediction that there would be a significant interaction between sexual attraction and the sex of the extra-pair partner was not confirmed. What we found instead was that people with same-sex attractions were more tolerant to both same-sex and opposite-sex infidelity than people without such attractions. This finding suggests that a preference for similarity was not the primary reason why people with same-sex attractions demonstrated higher tolerance to same-sex infidelity. One possibility is that same-sex attractions are associated with more unrestricted sociosexuality—willingness to engage in sexual activity outside of a committed relationship (Penke & Asendorpf, 2008; Simpson & Gangestad, 1991)—and more unrestricted sociosexuality is associated with higher tolerance to infidelity. In this scenario, people who experience same-sex attractions have also more unrestricted sociosexuality which renders them more tolerable to infidelity. Future research needs to further investigate the role of sociosexuality to the tolerance to infidelity.

Consistent with our original prediction, the sex difference in tolerance was found in different cultural setting. Yet our analysis indicated that there were also significant cultural differences. Such differences are consistent with the proposed evolutionary framework, as adaptations are not inflexible but responsive to environmental conditions (Buss, Shackelford, Kirkpatrick, & Larsen, 2001). One difference was that the Chinese and the Greeks exhibited higher tolerance to infidelity than the British. A possible reason is that the former two ascribe considerable value to the family (see Wang & Apostolou, 2017) and thus are more likely to tolerate infidelity in order to keep an intimate relationship. On the other hand, British people are more individualistic and thus have a lower threshold of tolerance to their partners’ infidelity. Another cultural difference was that Chinese people indicated more similar rates of tolerance to their short-term and long-term partners’ infidelity than the British and the Greeks. In the Chinese culture, the Confucian ethic is prevalent which mandates the formation of family (Peng, 2014). Thus, Chinese people distinguish less between long-term and short-term partners than other cultures—that is, they tend to see their partners as long-term (Huang, 2018), which can potentially explain the cultural differences we have found in this dimension. Last but not least, the one-child policy interacting with a preference for male children has resulted in a biased sex ratio where there are more Chinese men than women (Deng, 2000). Men in China knowing that women are in short supply they may be more willing to tolerate infidelity than men in other countries where the sex ratio is more balanced. Thus, the unbalanced sex ratio may also contribute in men exhibiting higher tolerance to same-sex infidelity in China than in other countries.

Moving on, if a man cheats with another woman, this is a considerable risk to the existing pair bond. He can leave his current partner, divert resources away from her and her offspring, and commit his parental investment to children outside the relationship. Yet, if a man cheats with another man, the risk is potentially lower, primarily because same-sex contacts do not lead to children. On this basis, we would expect that women would also be more tolerant to their partners’ same- than to the opposite-sex infidelity, which is not what we have found. In addition, if they were forced to choose, more women preferred their partners to cheat with a woman than with a man. One possibility is that male homosexuality in Western societies is often associated with femininity and women tend to prefer masculinized men (Buss, 2017). Male same-sex infidelity then would appear to contradict key mate-value characteristics based on our cultural perceptions of male homosexuality. More research is required if women’s tolerance to same-sex infidelity is to be better understood.

It has been theorized that weak negative selection pressures in the ancestral human societies explained the high prevalence of same-sex attraction among contemporary women (Apostolou, 2016). One source of such weak negative selection pressure was proposed to be high male tolerance to same-sex attraction and contact. In particular, alleles that predispose for same-sex attraction in women are likely to experience negative selection pressures arising from poor relationships with men. Nevertheless, if men exhibit tolerance to same-sex attraction and contact, such selection pressures would be weak, allowing these alleles to exist in higher frequency in the gene pool. In addition, it has been argued that men find same-sex attraction and contact desirable in an opposite-sex partner, which potentially could have been a source of positive selection on this trait (Apostolou et al., 2017). Evidence from the present research indicates that these two hypotheses could be integrated into one.

More specifically, it was found that men who had a preference for same-sex attraction and contact in an opposite-sex partner were more likely to exhibit tolerance to their partners having same-sex contacts than men who did not share such preferences. The effect sizes were not very large, indicating that the overlap in high tolerance and preferences for same-sex attraction and contact was not complete. In other words, some men who preferred same-sex attraction and contact in a female partner were not very tolerant of same-sex infidelity, while some others who were more tolerant of same-sex infidelity did not prefer same-sex attraction and contact in an opposite-sex partner. Nevertheless, for a substantial proportion of men, preference for same-sex attraction and contact in an opposite-sex partner was associated with high tolerance to same-sex infidelity. Assuming that these effects were also present in ancestral human societies, such a preference would likely lead to a higher prevalence of same-sex attraction in women. Future theorizing on same-sex attraction should integrate these findings in order to better understand the evolutionary origin of this trait.

Moreover, we found that women with same-sex attractions were more tolerant of the opposite-sex infidelity of their partners than women without such attractions. This difference can enable us to better understand the male strategy of preferring same-sex attractions in an opposite-sex partner. It is beneficial for men to follow a strategy of having casual sex with different women, but they are constrained by their long-term partner unwillingness to allow them to do so (Buss, 1994, 2017). If women with same-sex attractions are less constraining in this dimension, it would pay for men to desire such women as partners. Another possibility is that same-sex attraction in women has partially evolved to address male preferences for same-sex attraction and contact in partner. But if this male preference has evolved in the first place to enable men to have sexual access to women other than their partners, women who have evolved to address these preferences need also to exhibit higher tolerance to their male partner having sex with other women.

The current work is not without limitations, one being that it was based on self-report data where people may not be accurate in estimating their reactions. For instance, younger individuals may not have the chance to meet potential opposite-sex partners with similar same-sex attractions and, thus, may not be accurate in assessing their possible reactions to same-sex infidelity. Also, some participants may face confusion about their sexual attraction, and they may not be able to accurately classify themselves accordingly to the categories given. Similarly, the definitions of the different categories of sexual attraction may not be in line with the definitions that participants had in mind. For example, heterosexual individuals who occasionally experience same-sex attractions may classify themselves as bisexual instead of heterosexuals with same-sex attractions. Furthermore, because nonrepresentative samples were used, our estimates may not necessarily reflect population estimates.

In conclusion, in the current research, we explored tolerance to same-sex infidelity across different cultural settings as well as how this tolerance relates to the male preference for same-sex attraction and contact in an opposite-sex partner. Our findings have important implications for understanding human sexuality and the evolutionary origins of same-sex attraction in women, but much more research is necessary in order to understand this complex phenomenon.