ترجیحات همبسته مردسالاری صورت و آوازی مردان

Previous studies have reported variation in women's preferences for masculinity in men's faces and voices. Women show consistent preferences for vocal masculinity, but highly variable preferences for facial masculinity. Within individuals, men with attractive voices tend to have attractive faces, suggesting common information may be conveyed by these cues. Here we tested whether men and women with particularly strong preferences for male vocal masculinity also have stronger preferences for male facial masculinity. We found that masculinity preferences were positively correlated across modalities. We also investigated potential influences on these relationships between face and voice preferences. Women using oral contraceptives showed weaker facial and vocal masculinity preferences and weaker associations between masculinity preferences across modalities than women not using oral contraceptives. Collectively, these results suggest that men's faces and voices may reveal common information about the masculinity of the sender, and that these multiple quality cues could be used in conjunction by the perceiver in order to determine the overall quality of individuals.

Among humans, face, voice, and body attractiveness are influenced by their degree of masculinity or femininity (DeBruine et al., 2006, Fan et al., 2005, Fan et al., 2004, Feinberg et al., in press, Feinberg et al., 2006, Feinberg et al., 2005, Perrett et al., 1998 and Rhodes et al., 2000). In turn, it has been demonstrated that sex hormones (primarily testosterone, progesterone, and estrogen) are related to the degree of masculinity and femininity displayed by men's and women's faces (Law-Smith et al., 2006, Penton-Voak & Chen, 2004 and Roney et al., 2006), voices (Abitbol et al., 1999, Alonso & Rosenfield, 2002, Brukert et al., 2006, Dabbs & Mallinger, 1999 and Feinberg et al., 2006), and bodies (Jasienska, Ziomkiewicz, Ellison, Lipson, & Thune, 2004). It is likely that males displaying testosterone-dependent traits to a greater degree can afford to produce such traits despite the immunosuppressive effects (Folstad & Karter, 1992 and Thornhill & Gangestad, 1999), antisocial behavior (Archer et al., 1998, Book et al., 2001, Gonzalez-Bono et al., 1999, O'Connor et al., 2004, Rowe et al., 2004, Studer et al., 2005 and Tremblay et al., 1998), and tendency to take risks (Archer, 1999 and Booth et al., 1999) that are thought to be associated with high testosterone levels. Thus, facial and vocal masculinity may be considered cues of costly testosterone levels. Furthermore, men in a natural-fertility population with low voice pitch have higher reproductive success than men with relatively high voice pitch do (Apicella et al., 2007). There is substantial evidence that people who are attractive in one domain (e.g., face, voice, or body) are also attractive in other domains (Collins & Missing, 2003, Feinberg et al., 2005, Hughes et al., 2004, Saxton et al., 2006 and Thornhill & Grammer, 1999). Indeed both men's (Saxton et al., 2006) and women's (Collins & Missing, 2003 and Feinberg et al., 2005) facial attractiveness are positively correlated with the attractiveness of their voices. Both men and women with attractive voices and faces also tend to have attractive body configurations, such as low fluctuating asymmetry in women (Hughes, Harrison, & Gallup, 2002) and a masculine upper-body shape in men (Hughes et al., 2004). The findings described above suggest that humans display multiple cues to the same underlying quality. However, a few key questions regarding the evolution of multiple quality cues in humans remain unresolved. While many studies show that women have consistent preferences for masculine men's voices across studies (Collins, 2000, Feinberg et al., 2006, Feinberg et al., 2004 and Saxton et al., 2006), different studies have yielded preferences in women for masculine (DeBruine et al., 2006 and Johnston et al., 2001), average (Cornwell et al., 2004 and Swaddle & Reierson, 2002), and feminine (Perrett et al., 1998 and Rhodes et al., 2000) men's faces. It has been suggested that differences in the computer graphic methods that have been used in different studies of preferences for masculinity in men's faces may explain these variable findings for women's face preferences (Penton-Voak & Chen, 2004, Rhodes, 2006 and Swaddle & Reierson, 2002). However, studies using the same method to manipulate masculinity in male faces have found different general preferences: DeBruine et al. (2006) reported a general preference for masculinity among women, Cornwell et al. (2004) found that average faces were generally preferred by women to feminized or masculinized versions, and Perrett et al. (1998) reported that women demonstrated strong aversions to masculinity in men's faces. More important, DeBruine et al. (2006) compared the strength of women's preferences for masculine faces using different types of computer graphic methods, finding that women who preferred facial masculinity did so for each type of manipulation. Given that both male vocal and facial masculinity are influenced by testosterone, and masculinity and femininity affect voice and face attractiveness, why are women's preferences for masculinity in the voice consistently above chance, but women's preferences for masculinity in the face vary considerably more from study to study? Studies have revealed a great deal of individual variation in female preferences for both facial and vocal masculinity. Sources of variation in women's preferences for male vocal masculinity that have been identified to date include relationship context (Puts, 2005), menstrual cycle phase (Feinberg et al., 2006 and Puts, 2005), and height (Feinberg, Jones, Little, et al., 2005). Women prefer masculinity more when in the most fertile menstrual cycle phase (Feinberg et al., 2006 and Puts, 2005) and when rating voices as potential short-term partners (Puts, 2005). Taller and heavier women also prefer men with voice characteristics rated as more masculine sounding (Feinberg, Jones, Little, et al., 2005). Similar sources of systematic variation in face preferences have also been found (see Table 1 for an extensive list of studies). Table 1. Potential sources of variation in women's preferences for male facial masculinity Potential source of variation in facial masculinity preference Direction of relationship with facial masculinity preferences Study Being in a committed relationship − Little et al. (2002) Rating faces in a relationship context Short-term + Little et al. (2002) Long-term − Oral contraceptive use Can mask masculinity preferences Little et al. (2002) Self-rated attractiveness + Little et al. (2001) Women's attractiveness as rated by men + Penton-Voak et al. (2003) Waist–hip ratio − Penton-Voak et al. (2003) Menstrual-cycle phase − at nonfertile phases Frost (1994), Johnston et al. (2001), Penton-Voak and Perrett (2000), Penton-Voak et al. (1999) State progesterone level − as progesterone increases Jones, Little, et al. (2005) State testosterone level + as testosterone increases Welling et al. (2007) Second-to-fourth digit ratio + Scarbrough and Johnston (2005) Age + Little et al. (2001) Paternal investment − Penton-Voak et al. (2004) Table options This overlap in sources of individual differences (i.e., menstrual cycle and relationship context) between face and voice is consistent with the hypothesis that preferences for masculinity in men's faces and voices may be concordant, despite variation across studies in women's generalized preferences for male facial masculinity. Indeed, previous studies showing positive associations between the strength of women's preferences for masculinity in men's faces and both putative male pheromones (Cornwell et al., 2004) and the reported masculinity of partnered women's romantic partners (DeBruine et al., 2006) suggest correlated preferences for masculinity in different domains. Nevertheless, while Feinberg, Jones, Law-Smith, et al. (2006) found that women with the lowest average estrogen levels demonstrated the largest cyclic shifts in vocal masculinity preferences, Welling et al. (2007) found that women with the highest average estrogen levels demonstrated the largest cyclic shifts in facial masculinity preferences (see also Johnston et al., 2001, for further evidence that particularly feminine women show larger cyclic shifts in preferences for faces of masculine men). Since any costs of producing multiple ornaments will outweigh the benefits of redundant ornaments, why would men produce more than one cue to testosterone levels? While both facial and vocal masculinity are influenced by testosterone, neither facial nor vocal masculinity is perfectly correlated with testosterone levels. In other words, each cue also has a degree of error (Candolin, 2003 and Møller & Pomiankowski, 1993). Indeed, people can modify their voice pitch (within physiological constraints) and their apparent facial masculinity (e.g., altering brow height; Campbell, Benson, Wallace, Doesbergh, & Coleman, 1999). Thus, it is possible that there remains selection pressure from receivers for senders to produce multiple cues to the same underlying quality in order to (a) more easily detect dishonesty, (b) reduce error in cue perception, or both. Both proximate explanations result in an ultimate effort to evoke a more robust assessment of the sender's overall quality. If multiple quality cues are used by receivers to detect dishonest cues, it may then be an evolutionarily stable strategy (Maynard-Smith, 1976) for senders to produce consistent multiple quality cues. Alternatively, senders' multiple quality cues could merely demonstrate to receivers that they are of such quality that they can spend their resources on more than one ornament if such ornaments themselves are costly. There is evidence of inconsistencies between generalized vocal and facial masculinity preferences and also inconsistencies in the nature of individual differences in the strength of masculinity preferences. Furthermore, although it has been demonstrated numerous times that people are sending multiple quality cues across visual and vocal domains, it is unknown if these cues are used in a consistent manner. To address these issues, we examined the extent to which the strength of men's and women's preferences for male facial masculinity is associated with the strength of their preferences for male vocal masculinity. As studies have shown that hormonal contraception is associated with a disruption of potentially adaptive facial masculinity preferences (Little, Jones, Penton-Voak, Burt, & Perret, 2002) and a disruption of correlations between preferences for male-typical putative pheromones and facial masculinity preferences (Cornwell et al., 2004), we investigated whether women using hormonal contraceptives have similar face and voice masculinity preferences to those not using hormonal contraceptives. As others have found that relationship status (partnered vs. single) affects facial masculinity preferences (Little et al., 2002), we also investigated the role of relationship status on the association between facial and vocal masculinity preferences.