ایجاد صدای مردانه : ارتباط فیزیولوژیکی و بلندگو (ووفر) از رتبه بندی زنان از مردسالاری آوازی مردانه

Abstract Men's voices contain acoustic cues to body size and hormonal status, which have been found to affect women's ratings of speaker size, masculinity and attractiveness. However, the extent to which these voice parameters mediate the relationship between speakers' fitness-related features and listener's judgments of their masculinity has not yet been investigated. We audio-recorded 37 adult heterosexual males performing a range of speech tasks and asked 20 adult heterosexual female listeners to rate speakers' masculinity on the basis of their voices only. We then used a two-level (speaker within listener) path analysis to examine the relationships between the physiological (testosterone, height), acoustic (fundamental frequency or F0, and resonances or ΔF) and perceptual dimensions (listeners' ratings) of speakers' masculinity. Overall, results revealed that male speakers who were taller and had higher salivary testosterone levels also had lower F0 and ΔF, and were in turn rated as more masculine. The relationship between testosterone and perceived masculinity was essentially mediated by F0, while that of height and perceived masculinity was partially mediated by both F0 and ΔF. These observations confirm that women listeners attend to sexually dimorphic voice cues to assess the masculinity of unseen male speakers. In turn, variation in these voice features correlate with speakers' variation in stature and hormonal status, highlighting the interdependence of these physiological, acoustic and perceptual dimensions

Introduction Male masculinity is linked to the expression of sexually selected morphological traits that emerge at sexual maturity (Andersson, 1994) and which are associated with individuals' hormonal and physical quality. For example, sexually dimorphic, masculine facial (i.e. large jaws and pronounced brows) and bodily (i.e. broad shoulders, narrow hips, tallness) traits positively correlate with health status, physical strength and self-reported mating success (Fink et al., 2007, Gallup et al., 2007, Hönekopp et al., 2007, La Batide-Alanore et al., 2003, Prokop and Fedor, 2013, Samson et al., 2000, Smith et al., 2000 and Thornhill and Gangestad, 2006). To the extent that masculinity correlates with underlying fitness, perceiving its variation is crucial when choosing a mate. Indices of masculinity in men's faces and bodies are indeed attractive to women, especially when most fertile during their menstrual cycle (Little et al., 2007, Welling et al., 2007 and Zebrowitz and Rhodes, 2002) and when explicitly asked to judge for short-term mating (Little et al., 2002 and Rhodes et al., 2005). Along with facial and bodily features, the human voice is a sexually dimorphic trait: compared to women, men speak at a lower fundamental frequency (F0 — lower pitch), and lower, more closely spaced formant frequencies (deeper timbre) (Titze, 1994). These differences are at least partly affected by hormonally induced changes occurring during male puberty. Pubertal exposure to androgens causes a 60% increase in men's vocal fold length relative to women, and a corresponding decrease in its inverse acoustic correlate, mean F0 (Harries et al., 1998 and Titze, 1994). Under the influence of androgens, pubertal males also grow 7% taller than women on average (Gaulin and Boster, 1985) and develop a further descended larynx, causing an increase in the lengthening of their vocal tract and thus a permanent drop in its inverse acoustic correlate, formant spacing or ΔF (Fitch and Giedd, 1999 and Vorperian et al., 2009). Because of the relationship between sexually dimorphic acoustic properties and underlying biological dimorphisms such as testosterone levels and body stature, acoustic variations amongst adult males may provide indexical cues of fitness-related features, with lower frequency (more masculine) values signalling greater fitness. For example, according to the immunocompetence handicap hypothesis, testosterone controls the development of sexual markers, while causing immunosuppression (Folstad and Karter, 1992). Thus, cues to testosterone are considered to signal better fitness because only males with strong immune systems can afford to express these costly traits (Folstad and Karter, 1992 and Thornhill and Gangestad, 2006). Testosterone has also has been found to positively correlate with disease resistance (Rantala et al., 2012), perceived masculinity (Penton-Voak and Chen, 2004), dominance (Mazur and Booth, 1998), social status (Eisenegger et al., 2011) and mating success (Peters et al., 2008), though it is also associated with decreased parental investment (Fleming et al., 2002), higher rates of antisocial behaviour (Booth et al., 2006) and infidelity (Booth and Dabbs, 1993). Moreover, in most mammals body size has been shown to play a major role in acquiring mates and resources, as larger males are more likely to win fights (Lindenfors et al., 2007), and are more attractive to females (Charlton et al., 2007, Charlton et al., 2012 and McElligott et al., 2001). In humans, taller men have been found to be healthier (La Batide-Alanore et al., 2003 and Smith et al., 2000), enjoy higher reproductive (Nettle, 2002 and Pawlowski et al., 2000), academic (Hensley, 1993) and socioeconomic (Harper, 2000 and Judge and Cable, 2004) success, and are more attractive to Western women (Mautz et al., 2013, Stulp et al., 2014, Swami et al., 2008 and Yancey and Emerson, 2014) than shorter men, despite possible costs associated with male tallness (i.e. energy allocation trade-off between growth and reproduction: see Pisanski and Feinberg (2013) for a review). In a recent study (Kempe et al., 2013), height has also been found to positively correlate with other indices of masculinity, including greater weight and arm strength (though not with circulating testosterone). Correspondingly, taller men are consistently perceived to be more masculine than shorter men (Bogaert and McCreary, 2011, Jackson and Ervin, 1992, Little et al., 2007 and Melamed, 1992). Alongside body and facial sexually dimorphic traits, acoustic components of the voices have been shown to act as cues to testosterone and height. In particular, men's individual mean F0 has been found to negatively correlate with circulating levels of testosterone (Dabbs and Mallinger, 1999, Evans et al., 2008 and Puts et al., 2012) and higher mating success rates (Apicella et al., 2007 and Hodges-Simeon et al., 2011), while at least one study (Bruckert et al., 2006) has also reported a negative relationship between ΔF and testosterone, though more recent studies have failed to replicate these findings (Evans et al., 2008 and Puts et al., 2012). ΔF also seems to moderately correlate with speakers' body size, and in particular men's height (Evans et al., 2006, Greisbach, 2007 and Rendall et al., 2005; but see Van Dommelen and Moxness, 1995), with taller men speaking with lower ΔF. However, there appears to be no consistent relationship between stature and F0 within sexes: while two studies have reported significant correlations between height and F0 (Graddol and Swann, 1983 and Puts et al., 2012), other studies have failed to identify such a relationship (e.g. Evans et al., 2006, Künzel, 1989, Rendall et al., 2005, Sell et al., 2010 and Van Dommelen and Moxness, 1995). If vocal frequencies signal hormonal (i.e. testosterone levels) and physical (i.e. height) attributes of speakers, attending to such acoustic cues may have important consequences when assessing potential mates. Indeed, psychoacoustic studies (where voice frequencies are artificially manipulated) report that pronounced sexually dimorphic (more masculine) features in men's voices positively affect women's masculinity ratings (Feinberg et al., 2005, Feinberg et al., 2006, Feinberg et al., 2008 and Jones et al., 2010), as also shown for men's faces and bodies (Feinberg et al., 2008, Little et al., 2002 and Little et al., 2007). Moreover, in line with research on facial and bodily traits (Feinberg et al., 2008, Johnston et al., 2001, Little et al., 2002 and Little et al., 2007), preferences for masculine voices are strongest when the benefits of choosing more masculine mates outweigh the costs, such as when women are at the peak of their fertility during their menstrual cycle and when rating men as short-term rather than long-term mating partners (Feinberg et al., 2006 and Puts, 2005). However, the complex relationships between fitness-related, acoustic and perceived dimensions of males' masculinity remain under-investigated. The present study tests the hypothesis that the natural variation in sexually dimorphic voice cues (F0 and ΔF) of male speakers mediates the effects of their fitness-related characteristics (testosterone and height) on masculinity attributions made by women listeners. More specifically, in line with most previous research we expect F0 to mainly mediate between testosterone and perceived masculinity: higher testosterone men are expected to speak with lower F0 and be perceived as more masculine than their lower-testosterone counterparts. We also expect ΔF to mainly mediate between height and perceived masculinity: taller men are expected to speak with lower ΔF and in turn be perceived as more masculine than their shorter counterparts. However, given previous reports of negative correlations between testosterone and ΔF as well as between height and F0, we investigate all possible relationships amongst height, testosterone, F0, ΔF, and perceived masculinity.

Conclusions This study expands on previous investigations of masculinity expression in the human voice, by explicitly exploring the relationships between physiological (body height and testosterone), acoustic (F0 and ΔF) and perceptual dimensions (women listeners' ratings) of men's masculinity. Our results confirm links amongst all three dimensions, suggesting that women extract characteristics from men's voices that are important for mate choice, such as testosterone levels and size of male speakers. Women's ratings may therefore reflect adaptations for identifying high-quality mates, rather than (solely) representing some low-level perceptual bias (e.g. preferring lower voices as dominant, as suggested by Morton (1977) and Ohala (1983)). Future studies could use the same methodology to investigate this assumption by directly investigating the correlations amongst size, testosterone, acoustic variables and vocal attractiveness. While the reported variation in the mediatory effects of F0 and ΔF between the biological and perceptual dimensions is largely in agreement with published data, future studies with larger samples and the inclusion of additional factors at all three levels are also warranted to further clarify these relationships. At the perceptual level, future investigations should take into account listeners' individual differences: e.g. women who are taller and heavier (Feinberg et al., 2005), more fertile (Feinberg et al., 2006) and rate themselves as more attractive (Vukovic et al., 2008) than other women, show comparatively stronger preferences for masculine sounding voices in men. The impact of environmental factors on women's ratings of men's masculinity traits should also be considered: for example, women living in poorer countries with higher levels of infective disease and higher levels of short-term relationships display greater preferences for testosterone-dependent facial traits, and women living in countries with scarce resources exhibit less preference for male tallness (Pisanski and Feinberg, 2013). It is also important to note that, while our study used speech material with relatively neutral content, semantic content (at least when expressing mating interest) may affect the links amongst biological, acoustic and perceptual dimensions: for example, Vukovic et al. (2010) have found that positively valenced men's speech (i.e. were saying ‘I really like you’ as opposed to ‘I don't really like you’), increases women's preferences for masculinised voices. Further studies should investigate whether the correlations we report may be accentuated by the use of speech material with a content highlighting the relevance of masculinity (e.g. dating related). At the biological level, while salivary testosterone is commonly used as a biological marker of masculinity because of its relative temporal stability (Dabbs, 1990a and Sellers et al., 2007), it has also been shown to vary daily and seasonally (Dabbs, 1990b), and in response to different social contexts (i.e. increasing after ‘winning’ — Booth et al., 1989). Replication and extension of the current findings, preferably with repeated testosterone assays to account for testosterone variations, and the inclusion of additional correlates of masculinity (i.e. facial width-to-height ratio (Lefevre et al., 2013), reproductive success (Apicella et al., 2007)), would be desirable to shed light on the extent to which acoustic features cue for fitness-related traits. At the acoustic level, traits other than F0 and ΔF may also cue for fitness-related features and/or be associated with speaker's perceived masculinity and should therefore be included in further analyses: for example, listeners consistently rate more monotonous voices as less feminine than less monotonous voices (Ko et al., 2006). Finally, adults have been found to spontaneously modify sex-dimorphic acoustic cues (F0 and ΔF) in order to vary the expression of gender and related attributes in line with different roles and social (i.e. gender expression, dominance, sexual orientation) contexts (Cartei and Reby, 2012, Cartei et al., 2012, Graddol and Swann, 1983 and Puts et al., 2006), and this variation has a strong effect on the way listeners perceive the personality of speakers (Owen and Hancock, 2010, Puts et al., 2012 and Van Bezooijen, 1995). Because variation in vocal masculinity is likely to have both biological and social sources, future studies should also include social measures of masculinity (i.e. speakers' self-ratings of masculinity) in order to further explore how vocal masculinity relates to speakers' characteristics (both biological and social) and how these are perceived by listeners.