A large literature proposes that preferences for exaggerated sex typicality in human faces (masculinity/femininity) reflect a long evolutionary history of sexual and social selection. This proposal implies that dimorphism was important to judgments of attractiveness and personality in ancestral environments. It is difficult to evaluate, however, because most available data come from large-scale, industrialized, urban populations. Here, we report the results for 12 populations with very diverse levels of economic development. Surprisingly, preferences for exaggerated sex-specific traits are only found in the novel, highly developed environments. Similarly, perceptions that masculine males look aggressive increase strongly with development and, specifically, urbanization. These data challenge the hypothesis that facial dimorphism was an important ancestral signal of heritable mate value. One possibility is that highly developed environments provide novel opportunities to discern relationships between facial traits and behavior by exposing individuals to large numbers of unfamiliar faces, revealing patterns too subtle to detect with smaller samples.Inspired by evidence from nonhuman species indicating that exaggerated sex-typical traits (e.g., large antlers, peacock tails) are often attractive to mates or intimidating to rivals (
1,
2), morphological sex typicality in humans (masculinity in men and femininity in women) has been the focus of considerable research into attractiveness judgments (
3,
4). Facial attractiveness research has been revolutionized by this explanatory framework from the biological sciences, which proposes that attractive human faces honestly signaled mate value within ancestral environments.An influential proposal is that facial femininity is a signal of fertility in human female faces (
4–
9) because, within same-age women, it is associated with estrogens (
10), which, in turn, are related to measures of reproductive health (
11). Like ovarian function, facial femininity declines with age in adulthood (
12,
13). The proposal that fertile women should be attractive to men is seemingly uncontroversial because males who discriminatively mate with fertile females should achieve a straightforward reproductive advantage over those males who do not, with all other factors being equal (
6). Although direct associations between facial femininity and fertility have not been demonstrated, the consensus from Western preferences, and from the limited cross-cultural data available, is that femininity is attractive, as predicted by the fertility hypothesis (
14–
17). In environments where fertility is high and variable, this relationship should be even more apparent.In male faces, masculinity has been variously proposed to signal heritable disease resistance (“good genes” or “immunocompetence”) (
4,
15,
18–
22) and/or perceived as a cue of aggressiveness and, consequently, intrasexual competitiveness (
22,
23). The “honesty” of face shape as an indicator of immunocompetence is proposed to be the result of an immunosuppressive effect of testosterone. Because testosterone influences the growth of sex-typical traits in many species (
24,
25), masculine facial shape is proposed to be a costly, and thus honest, signal of male quality (
22). The hypothesis that cues of heritable health should be attractive to females is widely accepted (
26), although the evidence for a link between heritable health and masculinity in humans is tentative at best (
22).Support for a link between masculinity and aggression is largely indirect, and it consists of an association between testosterone and both aggressive behavior (
27,
28) and face shape (
25), in addition to the fact that honest signaling of dominance is commonly observed in nonhuman species (
3). Masculine faces are perceived as aggressive in those groups (i.e., urban, Western) where the relationship has been tested (
29). Because masculinity may signal both (desirable) immunity and (potentially costly) aggression in humans, some authors have proposed that preferences for masculinity reflect women trading-off benefits of traits putatively associated with health against those traits associated with prosocial behaviors, such as parental investment (
23,
30,
31).Consistent with both of these proposals, data indicate that preferences for masculinity are stronger in circumstances where indirect benefits (heritable quality) can be realized without accompanying direct costs (aggression and low paternal investment). Such circumstances include judging attractiveness in the context of a short-term (vs. a long-term) relationship (
32) and in the follicular phase of the menstrual cycle when conception following intercourse is most likely (
33). Masculinity is also reported to be more strongly preferred in environments with relatively high pathogen burdens (
19,
30) and in environments with higher local homicide rates (
23), which has been interpreted as a response to variation in the benefits of heritable disease resistance (
19) and in the net benefits conferred by aggressive males under varying levels of male–male competition (
23).All of this supporting evidence comes with a very important caveat; although there has been some cross-cultural work in this area (
34), the majority of studies have been conducted in Western, often student, populations characterized by high levels of development and urbanization [Western, educated, industrialized, rich, and democratic; so-called WEIRD participants (
35)]. Research on preferences in other groups is scant and methodologically inconsistent, using Internet-based designs or a limited cross-cultural component (
7,
15–
18). Because there are differences between Western/non-Western and industrial/small-scale societies in many behaviors, including aspects of visual perception and mate choice (
35), this over-representation greatly limits generalizability. Perhaps most importantly, large-scale (post)industrial societies present inhabitants with large numbers of unfamiliar faces and provide venues for the efficient exchange of (visual) social information (e.g., posters, television, Internet); these factors may be instrumental in the acquisition and reinforcement of preferences (
36–
39). It is possible therefore that rather than being a legacy of ancestral selection pressures, preferences for dimorphism emerge in large urban groups as a byproduct of the information-processing strategies used to process large amounts of social information or in response to arbitrary cultural norms.Development also introduces an increased presence of highly differentiated social roles that arise from a greater division of labor, along with opportunities to acquire prestige without strength or aggression. Because partner preferences have been proposed to develop in response to sex-typical social roles (
40,
41), it is possible that increasingly differentiated roles could influence masculinity preferences if desirable social roles not present in less developed groups are associated with facial appearance.We assessed preferences for, and trait attributions made to, faces varying in dimorphism in a cross-cultural sample of 12 groups, including non-Western, nonstudent, and small-scale societies (
n = 962; and
S2). We tested the predictions, derived from the immunocompetence handicapping hypothesis, that (
i) preferences for dimorphism will be stronger in less developed groups and (
ii) masculine faces would be perceived as aggressive in all populations, with perceptions in low-development groups at least as strong as in groups with high development. We estimated social development with the Human Development Index (HDI), which is a composite indicator compiled by the United Nations Development Program. To investigate which aspects of development were associated with variation in perception of our facial stimuli, we took the World Health Organization measures of years lost to disease and United Nations (UN) measures of homicide rates as proxy measures of disease burden and male intrasexual competition, respectively (both log-transformed), and UN measures of levels of urbanization. Using these national statistics almost certainly underestimates disease burden in the small-scale societies in our sample, which is a conservative estimate with regard to our hypotheses.
Table 1.
Summary information for the groups tested
Group | Local region | Country | Subsistence mode | n male | n female | n female after exclusions |
Canadian students | Alberta province | Canada | Market economy | 23 | 60 | 18 |
UK students | Bristol city | United Kingdom | Market economy | 80 | 238 | 134 |
Shanghai students | Shanghai municipality | China | Market economy | 41 | 38 | 38 |
Hangzhou citizens | Zhejiang province | China | Market economy | 43 | 52 | 48 |
Cree Canadians | Alberta province | Canada | Market economy | 26 | 28 | 13 |
Tuvans | Tyva Republic | Russia | Pastoralism, wages | 30 | 30 | 18 |
Kadazan-Dusun | Sabah region | Malaysia | Pastoralism, agriculture | 25 | 26 | 18 |
Fijian villagers | Cakaudrove province | Fiji | Foraging, agriculture, wages | 9 | 10 | 5 |
Shuar | Morona Santiago province | Ecuador | Horticulture, hunting, foraging, recent small-scale agropastoralism | 30 | 31 | 19 |
Miskitu | Región Autónoma del Atlántico Sur | Nicaragua | Horticulture, fishing, hunting | 13 | 17 | 15 |
Tchimba | Kunene region | Namibia | Pastoralism | 35 | 27 | 20 |
Aka | Southwest Central African Republic | Central African Republic | Foraging | 25 | 25 | 11 |
Open in a separate windowParticipants were asked to choose the most attractive face from five sets (representing five different ethnicities, representing considerable phenotypic variation in human faces) of three opposite-sex photographs, with one 60% masculinized [i.e., with the shape differences between male and female faces caricatured by 60% (
4)], one 60% feminized, and one unaltered face in each set (). Participants assessed attractiveness for long-term and short-term relationships. Participants were also asked to choose the most aggressive-looking face, and responses were scored in the same way. Custom randomization tests were used to test for nonrandomness of choice (e.g.,
Fig. S1), and ordinal generalized linear mixed models (GLMMs) were used to test for associations between choices and predictor variables.
Open in a separate windowExamples of stimuli used. A European female composite (
Upper) and an East Asian male composite (
Lower) are shown. Masculinized stimuli (
Left) and feminized stimuli (
Right) are shown.Although the previous literature suggests that familiarity effects of ethnicity can subtly affect dimorphism preferences, this influence is small and inconsistent across cultures and is unlikely to bias results as a result of exposure to ethnic variation in facial appearance (
4,
15).
相似文献