Original Article
Genetic analysis of human extrapair mating: heritability, between-sex correlation, and receptor genes for vasopressin and oxytocin

https://doi.org/10.1016/j.evolhumbehav.2014.10.001Get rights and content

Abstract

As in other socially monogamous species, pair-bonded humans commonly engage in sex with a partner other than their primary mate. For men, extrapair mating is straightforwardly explained from an adaptive perspective in terms of the reproductive benefits of multiple mates. For women, whose reproductive output is limited by their reproductive biology rather than by their number of mates, the adaptive benefits of extrapair mating are less obvious. Dominant adaptive explanations focus on women obtaining genetic benefits for their offspring by mating with high-quality extrapair partners. Non-adaptive explanations have rarely been considered in humans, but recent findings in birds suggest that females' predisposition to extrapair mating may result from indirect selection, via direct selection on males and a between-sex genetic correlation. To examine the plausibility of this non-adaptive explanation of extrapair mating in women, we used data on recent extrapair mating in 7,378 Finnish twins and their siblings. Genetic modelling showed within-sex broad-sense heritability—i.e. the percentage of variation in extrapair mating due to genetic variation—of 62% in men and 40% in women. There was no between-sex correlation in extrapair mating, making indirect selection unlikely. Based on previous animal and human findings, we also tested for association of the arginine vasopressin receptor 1A gene (AVPR1A) and oxytocin receptor gene (OXTR) with extrapair mating. We found gene-based association for AVPR1A in women but not in men, and OXTR showed no significant association in either sex. Overall, these findings confirm genetic underpinnings of extrapair mating in humans, but do not suggest that women's predisposition to extrapair mating is due to selection on men.

Introduction

In most socially monogamous species (e.g. many birds and some mammals), both male and female members of a pair commonly seek copulations with other individuals (Barash and Lipton, 2001, Griffith et al., 2002, Reichard, 1995). Males have a low minimal investment to reproduce (i.e. one copulation), so males mating outside the pair can increase their reproductive output; any genes predisposing males to seek extrapair mates would be adaptive (in the absence of strong countervailing selective pressures). However, females' reproductive potential is constrained by their biological capacity to reproduce, so females do not necessarily increase their reproductive potential by extrapair mating—in addition, females may also incur direct costs from extrapair copulations, such as disease transmission and withdrawal of paternal investment into offspring of uncertain paternity (Albrecht et al., 2006, Arnqvist and Kirkpatrick, 2005). As such, it is not clear why females in socially monogamous species have evolved such that they mate outside the pair (Forstmeier, Nakagawa, Griffith, & Kempenaers, 2014).

There have been proposed a number of adaptive explanations for female extrapair mating, along with challenges to the traditional theoretical and empirical basis for the expectation of sex-differentiation in adaptation for extrapair mating (Gowaty, 2013, Gowaty et al., 2012). The dominant explanation of female extrapair mating has been that it can be adaptive if females are able to obtain extrapair mates of higher genetic quality than their social mates, thereby increasing the genetic quality of their offspring and increasing their number of grandoffspring (Jennions and Petrie, 2000, Neff and Pitcher, 2005). However, reviews of the empirical evidence in socially monogamous birds suggest that the genetic benefits to offspring of extrapair matings are generally very weak or nonexistent, and are likely to be outweighed by direct costs (Akçay and Roughgarden, 2007, Arnqvist and Kirkpatrick, 2005). While there was debate as to the correct interpretation of these results (Eliassen and Kokko, 2008, Griffith, 2007), several more recent studies directly testing for such indirect benefits in birds suggest that offspring of extrapair matings actually have lower lifetime fitness and genetic value than offspring of within-pair matings (Hsu et al., 2014, Reid and Sardell, 2012, Sardell et al., 2012; though see Gerlach, McGlothlin, Parker, & Ketterson, 2012), which poses a major challenge to this as a general adaptive explanation of female extrapair mating. As such, alternative explanations need to be considered.

One such alternative (nonadaptive) explanation is the between-sex genetic correlation hypothesis, which is that genetic variants predisposing males to male extrapair mating (and hence putatively selected for) might also predispose females to extrapair mating (Arnqvist and Kirkpatrick, 2005, Forstmeier et al., 2011, Forstmeier et al., 2014). That is, female extrapair mating behaviour is maintained as a byproduct of selection for this behaviour in males. A recent finding of genetic correlations between measures of male and female extrapair mating behaviour in zebra finches (Forstmeier et al., 2011) is consistent with this hypothesis. While this finding does not in itself invalidate adaptive hypotheses in this or other species, it does warrant the consideration of between-sex genetic correlation as a plausible alternative to adaptive explanations of female extrapair mating.

These findings have important implications for evolutionary research into human mating; socially monogamous partnerships are the most common form of marriage even among forager societies in which other arrangements (e.g. polygyny, polyandry, promiscuity) are also common (Marlowe, 2003). As in other species, extrapair copulation is common in humans across cultures (Greiling and Buss, 2000, Marlowe, 2000), and nonpaternity rates are non-zero in all societies that have been studied (Anderson, 2006) and are quite high (9% and 17%) in the two small-scale natural-fertility (i.e. similar to ancestral) populations in which this has been carefully investigated (Neel and Weiss, 1975, Scelza, 2011)—this rate is comparable to an estimated average rate of extrapair paternity among bird species (11%; Griffith et al., 2002).

The dominant evolutionary theories of human mating strategies (e.g. sexual strategies theory Buss & Schmitt, 1993; strategic pluralism Gangestad & Simpson, 2000, dual mating strategies Fisher, 1992) regard both men and women as having evolved distinct psychological mechanisms adapted for both long-term and short-term (including extrapair) mating strategies. Pillsworth and Haselton (2006) specifically propose that women are endowed with suites of adaptations that function to form a social partnership with a man she judges to be a reliable investing partner while surreptitiously seeking good genes (for her offspring) from another man through extrapair sexual encounters. While there is indirect evidence from a variety of sources consistent with this hypothesis (reviewed in Gangestad, 2006, Pillsworth and Haselton, 2006), there is no direct evidence to this effect (e.g. there is no evidence that offspring of extrapair matings are fitter than offspring of within-pair matings). Given this and the aforementioned recent findings in socially monogamous birds, which suggest that extrapair offspring are less fit than within-pair offspring (Hsu et al., 2014, Reid and Sardell, 2012, Sardell et al., 2012) and that there is substantial cross-sex correlation in extrapair mating behaviours (Forstmeier et al., 2011), it is worthwhile investigating the plausibility of the between-sex genetic correlation as an alternative explanation for female extrapair mating in humans. Previously, this alternative explanation has barely been considered.

There is evidence from studies of identical and nonidentical twins that sociosexuality (i.e. orientation towards short- or long-term mating strategy) is heritable in both men and women. Bailey, Kirk, Zhu, Dunne, and Martin (2000) estimated that genetic factors account for 26% and 43% of the variance in men and women, respectively, although it should be noted that the male genetic variance did not reach statistical significance. Furthermore, there was a significant between-sex correlation, consistent with the between-sex genetic correlation hypothesis. However, the sociosexuality score was made up of a variety of measures, most of which did not pertain to extrapair mating per se (i.e. copulating with others while in a pair–bond relationship). There has been one twin study specifically on extrapair mating, but only in women (Cherkas, Oelsner, Mak, Valdes, & Spector, 2004); in that study, 41% of the variance in female infidelity was estimated to be accounted for by genetic factors. It remains unknown as to what extent genetic factors influence men's extrapair mating behaviour and whether they are the same genetic factors as influence on women's extrapair mating behaviour. This knowledge is crucial in weighing the relative merits of adaptionist and genetic-constraint explanations of female extrapair mating in humans.

Here we conduct two studies investigating potential genetic influences on male and female extrapair mating, and whether the same genetic factors influence the behaviour in both sexes. Study 1 uses the classical twin design to estimate the proportion of variation in extrapair mating that can be attributed to genetic differences in general, while study 2 tests variation in two specific genes (oxytocin and vasopressin receptor genes) for association with extrapair mating.

Section snippets

Study 1

In study 1 we used data from 7,378 twins and siblings who are in long-term relationships to estimate within-sex heritability and test for a between-sex correlation in recent extrapair copulation in order to assess the plausibility of the between-sex genetic correlation explanation of female extrapair mating in humans.

Participants

The full Finnish community-based twin-sibling sample consisted of 13,092 individuals aged from 18 to 49 (M = 29.2, SD = 7.3) from 7,737 families (see Johansson et al., 2013); for analysis we used the subset of individuals who had been in a relationship for at least the last year (see Measures for details), which consisted of 7,378 individuals aged from 18 to 49 (M = 29.8, SD = 6.4). Families with only one participating member who was in a relationship were retained because those data help stabilise the

Preliminary testing

Of the individuals who had been in a relationship for at least the last year, 9.8% of men and 6.4% of women reported two or more sexual partners in the last year, indicating extrapair mating. Age effects on extrapair mating were not significant.

Corresponding DZ twin and sibling correlations were equated (i.e. male DZ/sibling pairs; female DZ/sibling pairs; opposite-sex DZ/sibling pairs) without loss of model-fit (χ23 = 3.28, p = .35), consistent with the equal genetic similarity of DZ and sibling

Study 2

In study 2 we tested whether some of the genetic variation in extrapair mating identified in study 1 might be due to specific genes that have previously been associated with pair-bonding behaviour. Arginine vasopressin and oxytocin are hormones found in most mammals. A substantial body of work on monogamous and non-monogamous species of voles implicates these hormones and their receptor genes—arginine vasopressin receptor 1A gene (AVPR1A) and oxytocin receptor gene (OXTR)—in the striking

DNA extraction and genotyping

From saliva samples, 12 SNPs were genotyped in the OXTR gene, and 7 SNPs in the AVPR1A gene. In addition, two microsatellites in the promoter region of AVPR1A (RS1, which is a (GATA)14 tetranucleotide repeat; and RS3, a complex (CT)4-TT-(CT)8-(GT)24 repeat, both upstream from the transcription start site) were genotyped. Saliva samples were collected using the Oragene™ DNA (DNA Genotek, Inc.) self-collection kits that were posted to the participants and returned by mail. The participants were

Descriptive statistics

Genotype data were available for a subset of individuals (n = 2483–2527, the exact sample size varying between different loci due to individual occurrences of genotyping error) from the second data collection of the GSA sample (Johansson et al., 2013). The allele frequencies and genotype distributions for the SNPs can be seen in Table 3 (data for men and women presented together). On average, SNP data were available for 946 men (range = 933–953) and 1564 women (range = 1550–1579).

A standard test for

Discussion

There are several novel findings from these two studies. First, we found significant genetic influences accounting for around half the variation in extrapair mating in both sexes, confirming biological underpinnings to the behaviour. Second, we found a near-zero cross-sex correlation in extrapair mating—that is, 697 brother–sister pairs showed no similarity in likelihood of having extrapair mates. A near-zero cross-sex correlation means that extrapair mating in females is unlikely to be

Acknowledgments

We thank Stuart Macgregor for assistance with the gene-based tests. BPZ is supported by a Discovery Early Career Research Award from the Australian Research Council. PJ was funded by a research grant (no. 138291) from the Academy of Finland.

References (49)

  • K.G. Anderson

    How well does paternity confidence match actual paternity? Evidence from worldwide nonpaternity rates

    Current Anthropology

    (2006)
  • G. Arnqvist et al.

    The evolution of infidelity in socially monogamous passerines: The strength of direct and indirect selection on extrapair copulation behavior in females

    American Naturalist

    (2005)
  • J.M. Bailey et al.

    Do individual differences in sociosexuality represent genetic or environmentally contingent strategies? Evidence from the Australian twin registry

    Journal of Personality and Social Psychology

    (2000)
  • M.J. Bakermans-Kranenburg et al.

    A sociability gene? Meta-analysis of oxytocin receptor genotype effects in humans

    Psychiatric Genetics

    (2014)
  • D.P. Barash et al.

    The myth of monogamy: Fidelity and infidelity in animals and people

    (2001)
  • F.J. Bosker et al.

    Poor replication of candidate genes for major depressive disorder using genome-wide association data

    Molecular Psychiatry

    (2011)
  • D.M. Buss et al.

    Sexual strategies theory - An evolutionary perspective on human mating

    Psychological Review

    (1993)
  • L.F. Cherkas et al.

    Genetic influences on female infidelity and number of sexual partners in humans: A linkage and association study of the role of the vasopressin receptor gene (AVPR1A)

    Twin Research

    (2004)
  • L.E. Duncan et al.

    A critical review of the first 10 years of candidate gene-by-environment interaction research in psychiatry

    American Journal of Psychiatry

    (2011)
  • S. Eliassen et al.

    Current analyses do not resolve whether extra-pair paternity is male or female driven

    Behavioral Ecology and Sociobiology

    (2008)
  • H.E. Fisher

    Anatomy of love: The natural history of monogamy, adultery, and divorce

    (1992)
  • W. Forstmeier et al.

    Female extrapair mating behavior can evolve via indirect selection on males

    Proceedings of the National Academy of Sciences of the United States of America

    (2011)
  • S.W. Gangestad

    Evidence for adaptations for female extra-pair mating in humans: Thoughts on current status and future directions

  • S.W. Gangestad et al.

    The evolution of human mating: Trade-offs and strategic pluralism

    Behavioral and Brain Sciences

    (2000)
  • Cited by (22)

    • Disgust sensitivity predicts sociosexuality across cultures

      2022, Evolution and Human Behavior
      Citation Excerpt :

      A second possible theoretical framework is the “good genes” model. In environments with higher pathogen risk, women may benefit by pursuing the males with the highest apparent immunocompetence (Larmuseau et al., 2019; Little, DeBruine, & Jones, 2011; Little, Jones, Penton-Voak, Burt, & Perrett, 2002; Penton-Voak & Perrett, 2000; Zietsch, Westberg, Santtila, & Jern, 2015). That is, highly infectious environments necessitate greater deployment of a mixed-mating strategy.

    • A critique of life history approaches to human trait covariation

      2020, Evolution and Human Behavior
      Citation Excerpt :

      First, a large proportion of variation in life history and related traits is attributable to genetic variation among individuals, and little is attributable to variation in the shared environment (i.e. the developmental home environment shared within twin pairs, including socioeconomic status, parenting style, father absence, risky upbringing). Heritability estimates from twin studies range from around 20 to 60% for age at menarche, age at first birth, age of menopause, number of children, number of sexual partners, sociosexuality, extra-pair mating, and the Big Five personality traits (Bailey, Kirk, Zhu, Dunne, & Martin, 2000; Jang, Livesley, & Vemon, 1996; Kirk et al., 2001; Zietsch et al., 2008; Zietsch, Kuja-Halkola, Walum, & Verweij, 2014; Zietsch, Westberg, Santtila, & Jern, 2015). Indeed, a factor designed to capture fast-slow life history ‘strategy’ has itself been estimated at 65% heritability (Figueredo et al., 2004).

    • The AVPR1A Gene and Its Single Nucleotide Polymorphism rs10877969: A Literature Review of Associations with Health Conditions and Pain

      2018, Pain Management Nursing
      Citation Excerpt :

      Vasopressin and oxytocin are nonapeptides located in the pituitary, consisting of nine amino acids arranged in a cyclic structure and differing at only at the three and eight positions. Arginine and phenylanine in vasopressin are replaced by leucine and isoleucine in oxytocin (Baribeau & Anagnostou, 2015; Ebstein et al., 2009; Jern et al., 2012; Moons et al., 2014; Zietsch et al., 2015). The structure of AVP is very similar to that of oxytocin, and oxytocin appears to interact with AVPR1A.

    • Monogamy and Nonmonogamy: Evolutionary Considerations and Treatment Challenges

      2016, Sexual Medicine Reviews
      Citation Excerpt :

      Research on male prairie voles suggests that variations in the V1aR gene predicts sexual fidelity.59 Similar research suggests that variation in the vasopressin receptor gene AVPR1A correlates with pair bonding in human males60 and females.61 Genes in the dopamine system play a role in human infidelity and promiscuity in both men and women.62

    View all citing articles on Scopus
    View full text