Men were more accurate than women in detecting attraction when they were not interested in their partner compared to when they were interested

The Role of Emotion Projection, Sexual Desire, and Self-Rated Attractiveness in the Sexual Overperception Bias. Iliana Samara, Tom S. Roth & Mariska E. Kret. Archives of Sexual Behavior, Aug 13, 2021. https://rd.springer.com/article/10.1007/s10508-021-02017-5

Abstract: A consistent finding in the literature is that men overperceive sexual interest in women (i.e., sexual overperception bias). Several potential mechanisms have been proposed for this bias, including projecting one’s own interest onto a given partner, sexual desire, and self-rated attractiveness. Here, we examined the influence of these factors in attraction detection accuracy during speed-dates. Sixty-seven participants (34 women) split in four groups went on a total of 10 speed-dates with all opposite-sex members of their group, resulting in 277 dates. The results showed that attraction detection accuracy was reliably predicted by projection of own interest in combination with participant sex. Specifically, men were more accurate than women in detecting attraction when they were not interested in their partner compared to when they were interested. These results are discussed in the wider context of arousal influencing detection of partner attraction.

Discussion

The present study explored the effects of sex, own interest, sexual desire, and self-rated attractiveness in the overperception bias using a naturalistic speed-dating paradigm. Overall, we found that men were more willing to go out with their partner as compared to women. Importantly, our findings illustrate that projection of own interest influences attraction detection, particularly in men. Specifically, men were more accurate in detecting attraction if they were not interested in their partner compared to when they were. Furthermore, when men were interested in their partner, they overperceived interest more than women. However, there was no difference between sexes when participants were not interested in their partner. Women were approximately 50% accurate in detecting attraction, independent of whether they were interested in their partner or not. Sexual desire and self-rated attractiveness did not influence accuracy in detecting attraction. In the section below, we discuss these results in more detail.

First, we found that men were more likely to indicate that they were interested in going out with their partner again compared to women. This is in line with previous literature across different countries and target samples (i.e., university students and general population) showing a consistent pattern in terms of reduced male selectivity (e.g., Asendorpf et al., 2011; Fisman et al., 2006; Kurzban & Weeden, 2005; Lenton & Francesconi, 2010; McClure et al., 2010; Overbeek et al., 2013; Todd et al., 2007). An explanation could be that men wanted to maximize the number of dates that they could get, consistent with EMT (Haselton & Buss, 2000) which suggests that missing a dating opportunity could be more costly for men than for women. Also, the low likelihood of women indicating that they would like to meet their partner again supports previous findings showing that women are typically choosier than men (Todd et al., 2007; Trivers, 1972). In conclusion, we show that men were more likely than women to decide that they would like to go on another date with their partner supporting the notion that men are slightly less picky regarding dating.

It might be argued that the increased tendency of men to respond positively after a date can be explained by the fact that only men had to rotate between partners in our study. This effect was described by Finkel and Eastwick (2009), who showed that the reduced selectivity is nullified when female participants also rotate between partners. However, a recent meta-analysis showed that the female choosiness effect is robust across studies, and that the rotation effect did not moderate female choosiness (Fletcher et al., 2014), nor has been replicated (e.g., Overbeek et al., 2013). It is therefore unlikely that the partner-rotation effect can explain our findings. Nonetheless, future research should examine whether the sex-rotation-setup modulates the relationship between sex and the sexual overperception bias.

Interestingly, we found that men were more accurate when they were not interested in their partner compared to when they were, whereas women were approximately at 50% independent of their interest in their partner. An explanation for this interaction between sex and projection of own interest might be because of a link between choice biases and physiological arousal. Previous research has shown that men can detect changes in genital arousal that indicate sexual arousal within five minutes, and importantly, the correlation between genital arousal and subjective sexual arousal is reliable for men, but not for women (Kukkonen et al., 2007; see also Dekker & Everaerd, 1988). Physiological arousal influences our affective state, which can in turn bias our decisions (Damasio, 1996; see also Storbeck & Clore, 2008). For example, men that were shown sexually arousing stimuli were more likely to indicate that attractive women were sexually aroused than not (Maner et al., 2005) and sexually aroused participants are more likely to engage in risky sexual practices (Ariely & Lowenstein, 2006; Skakoon-Sparling & Cramer, 2021; Skakoon-Sparling et al., 2015). Thus, our findings might suggest that in situations where men were not interested in their partner, this biasing emotional state was not present, thus allowing them to accurately detect that their partner is not interested in them. Indeed, previous research has suggested that cues signaling disinterest might be easier to detect than cues signaling interest, especially in zero-order acquaintance settings (Hall et al., 2015). Given that the concordance between bodily and subjective arousal is not as robust in women, it is not surprising that women were not necessarily biased as much as men in terms of detecting attraction. In conclusion, our findings extend previous evidence showing that accuracy does not only depend on sex or projecting one’s own emotion on a partner, but accuracy is in fact dependent on an interplay between these two factors.

The estimation model complemented the results of the accuracy models. Interestingly, we found that both men and women were likely to overperceive attraction when they were interested in their partner compared to when they were not. Crucially, when men were interested in a partner, they overperceived interest more than women, which likely explains the decreased accuracy exhibited in men. These findings are partially consistent with EMT (Haselton & Buss, 2000). EMT predicts that men would be more likely to overperceive attraction than women. However, our findings highlight that perhaps the effect of being attracted to a given partner should be incorporated as an additional parameter in EMT (Lee et al., 2020), because if men are not interested, they are in fact very likely to be accurate regarding attraction. Thus, our findings support and further extend the EMT framework by showing that the addition of interest in a given partner might be crucial in predicting overperception.

Curiously, we found no effect of sexual desire on attraction detection accuracy. Our results are inconsistent with previous findings (Lee et al., 2020; Perilloux et al., 2012). One reason for this discrepancy could be that previous studies focused on short-term mating strategies, whereas we examined overall sexual desire. It is well known that sociosexuality—the inclination to form short-term relationships (Kinsey et al., 1948)—differs between men and women (Clark & Hatfield, 1989). Importantly, given that sexual desire and sociosexuality are highly correlated (O’Connor et al., 2014), we expected to observe similar findings as Lee et al. (2020). However, in our dataset we found no difference in sexual desire between sexes, whereas in Lee et al. (2020) sociosexuality was significantly higher for men than women (see also Roth et al., 2021). Either due to the differences in instruments or the differences in sample characteristics, we did not find an effect of sexual desire on attraction detection accuracy. Future research should investigate the effect of sexual desire and its association with sociosexuality and sex on attraction detection accuracy.

In addition, we found no effect of self-rated attractiveness on accuracy, in contrast with previous research (Lee et al., 2020; Perilloux et al., 2012). A potential explanation for this finding could be that in the present study, we examined physical attractiveness exclusively. We could therefore only speculate that our sample was similar to previous research in terms of other factors that can constitute attractiveness (e.g., personality). Nevertheless, previous research has shown that personality has negligible effects on both men and women’s desirability (Kurzban & Weeden, 2005). Furthermore, self-rated attractiveness has been found to play a role in overperception together with short-term mating styles (Howell et al., 2012; see also Lee et al., 2020; Perilloux et al., 2012). However, in our sample, most participants indicated they were searching for a long-term relationship. Thus, this pronounced long-term relationship focus might have prevented the interplay between self-attractiveness and mating strategy to emerge.

One crucial point that cannot be disentangled in the context of the present study is whether women and men interpreted the question regarding the wish to go on another date with their partner similarly. Specifically, in previous studies, participants were asked to indicate how sexually interested they were in their partner (Lee et al., 2020; Perilloux et al., 2012). However, in the present study, participants were asked to indicate whether they would like to go on another date with their partner (see also Asendorpf et al., 2011; Overbeek et al., 2013; Todd et al., 2007 for similar setups). It could be argued that this question led female participants to respond to the perceived question of “Are you romantically interested in your partner?” and male participants to respond to the question of “Are you sexually interested in your partner?” Even though this cannot be tested in the present study, it is quite likely that the response pattern would have remained the same. Previous research has shown that romantic interest and sexual interest follow the same sex differences, where women are choosier than men (Fletcher et al., 2014). Crucially, asking about the wish to go on another date rather than sexual interest is a strength of the current study, as it increases its ecological validity, given that it resembles real-life situations more closely (e.g., online dating sites; see Kurzban & Weeden, 2005).

It should be noted that in the present study, we examined only heterosexual participants; therefore, our findings cannot be directly generalizable to non-heterosexual populations. Furthermore, our sample consisted predominantly of university students. University students offer a prime target sample for sexuality research given the greater interaction frequency with opposite-sex partners and the increased necessity to infer sexual interest (Perilloux et al., 2012) and are commonly the primary target for such studies (e.g., Lee et al., 2020). Importantly, most participants in our study were interested in a committed relationship (only 2 participants were not), which limited our ability to investigate whether different mating strategies might influence attraction detection accuracy (e.g., Lee et al., 2020; Perilloux et al., 2012). Crucially, a limitation that stems from the use of a speed-dating setup is that we cannot assess whether the personality characteristics and social skills of our sample are representative of a wider population (Finkel & Eastwick, 2008). Future research should investigate more heterogeneous samples in terms of educational background and age.

The current study shed light on several factors that underlie the sexual overperception bias. Given that this bias is linked to the likelihood of assault (Abbey et al., 1998), the study’s findings are crucial in elucidating and reducing miscommunication between the sexes in dating contexts (Perilloux et al., 2012). Crucially, we showed that sex and projection of own interest are intertwined and should not be seen as competing, but rather as complementary explanations. Importantly, our findings cast doubt on previous research suggesting that one’s own interest, sexual desire, and self-rated attractiveness might fully explain the sexual overperception bias (Lee et al., 2020; see also Roth et al., 2021). Therefore, our results not only support the EMT framework, but further suggest that the incorporation of sex differences in projection of own interest might be a useful addition to the EMT framework.