Pregnant women may strategically employ sexual behavior to secure or reinforce investment in their future offspring, perhaps by reinforcing a partner's perception of paternity

Women’s sexual strategies in pregnancy. Jaclyn Magginetti, Elizabeth G. Pillsworth. Evolution and Human Behavior, Volume 41, Issue 1, January 2020, Pages 76-86, https://doi.org/10.1016/j.evolhumbehav.2019.10.001

Abstract: Humans exhibit an unusual pattern of sexual behavior compared to other mammalian females. Women's extended sexuality has been hypothesized to be related to a variety of possible benefits, especially non-genetic reproductive benefits, such as securing male investment via reinforced pairbonds or paternity confusion. But sexual behavior also comes at a cost, particularly for pregnant women, in terms of energetic costs, potential disease, and possible harm to the fetus. We hypothesize, therefore, that sexual behavior in pregnant women should reflect adaptive strategies and that pregnant women will be particularly strategic about their sexual behavior in order to maximize potential benefits while minimizing potential costs. One hundred twelve pregnant women completed a survey of their partners' qualities and their sexual desires toward their primary partners and men other than their primary partners. Results showed that women's perceptions of relationship threat positively predicted sexual desire for primary partners, while their perceptions of their partner's investing qualities negatively predicted sexual desire for extra-pair mates. These qualities, as well as cues to partner's genetic quality and gestation age, also interacted in ways that suggest that pregnant women's sexual desires are sensitive to cues of future investment and relationship stability.

4. Discussion

The results from this study provide much-needed insight into the strategic use of sexual behavior in pregnant women. The data suggest that sexual behavior in pregnancy reflects a strategic tool designed to benefit women's reproductive success through a variety of means. In contrast to previous literature that has reported a general decline in sexual desire across pregnancy, and in contrast to some studies that have documented apparent increases in desire specifically during the second trimester, we found no evidence of any direct effect of gestational age on in-pair sexual desire, either in a linear or curvilinear fashion. This suggests that variation in sexual desire during pregnancy is neither a simple hormonal response, nor likely to be a physiological byproduct of pregnancy.

Based on existing primate literature, one specific function that post-conception copulation may serve is to confuse paternity, which may be beneficial either in the context of potential risks posed by non-fathers to the offspring or if non-fathers may be able to provide additional investment or resources to the child. While misattributed paternity is only likely to occur if multiple mating takes place relatively early in pregnancy, cultures that exhibit beliefs in partible paternity provide an example of how sexual behavior throughout pregnancy can impact the paternal investment and resources available to a woman and her child (e.g., Walker et al., 2010). In our data, we saw no evidence that extra-pair sexual attraction was directly related to gestation age, but we did find that other factors that were related to women's experience of extra-pair sexual attraction tended to have greater effects earlier in pregnancy compared to later.

Many lines of evidence have shown that even the suggestion of possible infidelity can provoke extremely costly responses in male partners, from relationship desertion to homicide (Buss & Shackelford, 1997; Buunk & Massar, 2019; Jewkes, 2002; Wilson & Daly, 1993). Therefore, we expect that the potential benefits of pursuing any sort of extra-pair strategy should only exceed the likely costs if the benefits provided by the primary partner are low. Our data showed a pattern consistent with this prediction, though statistically weak, in which those women who rated their partners as less satisfactory in terms of their investing qualities were more likely to also report experiencing extra-pair sexual desire or flirtation, particularly early in pregnancy. It is important to note that our composite variable of extra-pair attraction included measures of both desire and potentially observable behavior, specifically, flirting with extra-pair men. Even subtle forms of overt behavior are likely to carry risks of partner retribution or more general social sanctions, and thus we find in many studies of human sexual strategies that measures of desire are more useful in illuminating the underlying sexual psychology that ultimately direct sexual behavior than are measures of the behaviors themselves (e.g. Brtnicka, Weiss, & Zverina, 2009; Haselton & Gangestad, 2006; Larson, Pillsworth, & Haselton, 2012; Pillsworth & Haselton, 2006; Roney & Simmons, 2013). In this study, too, we found that the predicted patterns were more evident when excluding overt behaviors and looking only at the unobservable factor of desire.

In addition to the predicted effect of partner investing qualities on pregnant women's extra-pair desires, we also observed a surprising effect of partner's physical attractiveness. Studies of women's extra-pair sexual desires across the ovulatory cycle have consistently shown that fertile women are more likely to experience extra-pair sexual desire if their primary partners are low in physical attractiveness (e.g. Pillsworth & Haselton, 2006). We observed the opposite effect among pregnant women; it was those with more physically attractive partners who experienced greater extra-pair sexual desire. This effect, like that of partner investing quality, was most evident in early pregnancy, when paternity confusion is most likely to occur. Previous research has suggested that men with greater cues to genetic quality, such as physical attractiveness, are, on average, less likely to be reliable investing long-term partners (Aitken, Lyons, & Jonason, 2013; Fisher, 2003). Perhaps those women who have conceived with “cads” are more likely to be seeking “dads” early in pregnancy. Although this is a striking pattern, and theoretically consistent with our overall argument, we caution against overinterpreting this finding. While statistically significant, this pattern is less robust than the others observed in our data, and could be susceptible to undue influence by outliers in the data. We suggest that this is an area that warrants further research.

We also predicted that an extra-pair strategy, whether with the motive of confusing paternity, extracting resources from more than one male, or mate-switching, should be predicted by perceived threats to the current relationship. We did not, however, observe any evidence that perceived threats predicted extra-pair attraction in our participants. Given that relationship maintenance, discussed below, and extra-pair tactics are ultimately opposing strategies, it seems likely that any attempt to pursue both simultaneously could easily result in failure at both. Our participant sample was heavily biased toward women in stable, committed relationships who expressed very high satisfaction with their partners, particularly in terms of their partners' investing qualities. A relationship maintenance strategy is likely, on average, to be more beneficial than a extra-pair strategy in this context, and may account for our failure to find any effect of perceived relationship threats on extra-pair attraction among this sample of pregnant women. Future research would benefit from examining strategic variation within a wider range of relationship contexts to uncover the contextual features that will predict one strategy over the other.

The two final strategies that have been proposed to explain post-conception copulations in the primate literature are relationship maintenance (in the context of “friendships,” “consortships,” or pairbonds) and intrasexual competition. Both of these strategies focus on engaging the primary partner (often, but not always, the father) in sexual interactions. However, the cues that should evince the behavior are likely to differ based on the strategy. If the aim is to secure future resources from a cooperative male partner, then pregnant women's in-pair sexual desire should best be predicted by both the value of that investment and cues that such investment may be at risk. If, on the other hand, the behavior is primarily intended to disrupt the reproduction of other women, then it should be less sensitive to features related to the male partner's likely investment and more sensitive to cues of female competitors' reproductive status. Because we did not have information in our data regarding the reproductive status, including fertility, of female competitors, we can not completely disambiguate these two strategies, but in general the data are consistent with a relationship maintenance strategy.

In our measure of perceived threats to the relationship, we distinguished between “perceived poaching attempts” (perceptions that other women were interested in or trying to “steal” one's partner) and “perceived cheating” (perceptions that the primary partner was, in fact, engaging in extra-pair sexual or romantic behavior). We found perceived relationship threat was the single biggest predictor of pregnant women's in-pair sexual desires, but only in the context of perceived poaching attempts. Perceived partner cheating had no effect on any outcome variables. As discussed above, however, our sample was heavily biased toward women in stable and satisfying relationships, the vast majority of whom reported no suspicions of partner cheating at all. We suggest that in a more representative sample, we would likely find distinct patterns based on the different types of threat, with perceived cheating being more likely to evoke a mate-switching strategy, resulting in increased extra-pair attraction. This should especially be the case if the primary partner is perceived as a less reliable investor. It is also possible that the increase in in-pair sexual desire in the context of potential mate poaching might be better explained as an effect of mate copying (Place, Todd, Penke, & Asendorpf, 2010) than of relationship maintenance, and that women simply find their partners more attractive when other women also find them desirable. In our data, however, perceptions of poaching threat and partner cheating had similarly negative relationships with women's satisfaction with their partner's investing qualities and physical attractiveness, making that explanation less likely. This is another area of research that deserves more attention going forward.

In our data, neither independent partner qualities, such as partner's physical attractiveness or investing qualities, nor dyadic qualities, such as sexual compatibility, had any statistically significant direct effects on pregnant women's experience of in-pair sexual desire. These findings suggest that it is the quality or stability of the relationship itself, rather than the specific qualities of the partner, that is most important to the pregnant woman in predicting her experience of sexual desire for her partner. It is particularly informative that, controlling for sexual compatibility, a partner's physical attractiveness was unrelated to women's sexual desire for her partner. This strongly suggests that sexual desire, in pregnant women as in fertile women, follows a strategic logic and is not simply the result of being faced with an attractive partner.

While there were no direct effects of partner qualities on women's in-pair sexual desire, we did find, as predicted, that the partner's qualities interacted with perceived threats to the relationship to predict women's in-pair sexual desire. A partner's willingness and ability to invest in a woman and her children will define, in part, the cost of losing the relationship. And in fact, we did observe a trend in the data suggesting that the effect of perceived mate poaching attempts on in-pair sexual desire was stronger among those women who rated their partners as better investors compared to those who rated their partners as poorer investors (see Fig. 3). This trend did not reach statistical significance, but we were, again, limited by the small range of variation in women's ratings of their partners' investing qualities. It is relevant to note here that factor analysis indicated that among our sample of pregnant women, parenting qualities (e.g., “good parenting,” “ability to be kind and understanding,” and “desire for children”) and resource-related qualities (e.g., “financial resources,” “social status,” and “ambition”) all loaded onto a single factor that we interpreted as investing qualities. In previous studies using a similar measure with non-pregnant women, these two types of investment (parenting and resources) have appeared to be separable to participants (e.g. Roberts, Craig Roberts, & Little, 2008). This further indicates that the stability of the relationship, and the cooperative parenting benefits that can be obtained therein, are likely to be of particular importance to pregnant women.

Finally, we also observed an unpredicted interaction between perceived poaching attempts and gestational age on women's in-pair sexual desires. As illustrated in Fig. 4, this pattern demonstrates that the effect of perceived threats on pregnant women's sexual desire is more pronounced among women who are earlier in their pregnancies, relative to those who are further along. While not specifically predicted, this pattern is consistent with the general hypothesis that pregnant women may strategically employ sexual behavior to secure or reinforce investment in their future offspring, perhaps by reinforcing a partner's perception of paternity. A promising direction for future research would be to investigate whether actual likelihood of misattributed paternity or even a partner's unfounded skepticism of paternity could influence the practice and timing of sexual behavior in response to perceived threats to the stability of the relationship.

The current study has some limitations, chief among them the relatively small and largely homogenous sample population. Despite recruiting online and receiving responses from all over the U.S., the sample was largely Caucasian, middle-class, and more-or-less happily married. To more fully investigate the strategic patterns of women's sexuality in pregnancy, it will be necessary to have a more diverse sample in terms of relationship status, resource access, and social support. Another limitation was that our sample lacked women very early in pregnancy. 95% of our sample were at least a month into their pregnancies, and while we think it is unlikely that this negatively affected the ability to test our hypotheses, future studies would benefit from recruiting women trying to conceive and those in the post-partum period in order to test predictions from the time of conception through the first weeks of parenthood.