Do more attractive women show stronger preferences for male facial masculinity?

Abstract

Researchers have suggested that more attractive women will show stronger preferences for masculine men because such women are better placed to offset the potential costs of choosing a masculine mate. However, evidence for correlations between measures of women's own attractiveness and preferences for masculine men is mixed. Moreover, the samples used to test this hypothesis are typically relatively small. Consequently, we conducted two large-scale studies that investigated possible associations between women's preferences for facial masculinity and their own attractiveness as assessed from third-party ratings of their facial attractiveness (Study 1, N = 454, laboratory study) and self-rated attractiveness (Study 2, N = 8972, online study). Own attractiveness was positively correlated with preferences for masculine men in Study 2 (self-rated attractiveness), but not Study 1 (thirdparty ratings of facial attractiveness). This pattern of results is consistent with the proposal that women's beliefs about their own attractiveness, rather than their physical condition per se, underpins attractiveness-contingent masculinity preferences.

An exploratory, cross-cultural study on perception of putative cyclical changes in facial fertility cues

“…Aside from interpopulation variation, individual differences in the person expressing the preference also can account for variation in facial preferences. A multitude of characteristics have been found to influence human facial preferences, including age (Marcinkowska et al 2017), sexual openness (Marcinkowska et al 2020), relationship status and resource availability (Lyons et al 2016), self-rated attractiveness and health 19 , contraception use ( 20 however see 21 for null effect of hormonal contraception on sexual dimorphism preference), and family constitution, including parenthood status 22 or similarity of judged face to self, sibling or parent 23,24 . This vast array of confounding variables putatively leads to discrepant results, depending on how the study is being conducted, and how many of possibly confounding variables have been included in the model.…”