Original Article The low male voice is a costly signal of phenotypic quality among Bolivian adolescents Carolyn R. Hodges-Simeon a, ⁎, Michael Gurven b , Steven J.C. Gaulin b a Department of Anthropology, Boston University, 232 Bay State Road, MA 02215 b Department of Anthropology, University of California, Santa Barbara, CA 93106 abstract article info Article history: Initial receipt 14 July 2014 Final revision received 22 January 2015 Available online xxxx Keywords: Voice pitch Costly signaling Handicap Immunocompetence Fundamental frequency Testosterone The human voice is one of the most conspicuous and dimorphic human secondary sexual characteristics; males' low fundamental and formant frequencies barely overlap with females'. Researchers often assert that low male voices are costly signals of phenotypic quality; however, no evidence currently exists linking low voices with in- dicators of quality (i.e., health or physical condition) during the ages where the larynx develops to adult propor- tions. In the present study, we examine the relationships between condition, testosterone, and vocal parameters in 91 Bolivian peri-pubertal adolescent males. Condition is operationalized as immune function (based on secre- tory IgA) and energetic reserves (BMI-for-age residuals from Tsimane-specific growth curves, and body fat per- centage), and “masculine” vocal parameters is operationalized as having low fundamental frequency, narrow formant position, and low fundamental-frequency variation. We target peri-pubertal individuals to capture var- iation in vocal parameters during the canalization period for vocal fold and vocal tract growth. Results indicate that males in better energetic condition have higher testosterone levels and lower voices, controlling for age. Fur- ther, testosterone mediates the relationship between condition and a lower voice (i.e., lower fundamental and formant frequencies). We suggest that testosterone plays a key mediating role in the causal pathway linking phenotypic condition to a “masculine” voice. Our results provide support for a costly-signal model of low men's voices. © 2015 Elsevier Inc. All rights reserved. 1. Introduction Secondary sexual characteristics facilitate male competitive and courtship interactions in a wide variety of vertebrates (Andersson, 1994). Among humans, one of the most conspicuous and dimorphic sec- ondary sexual characteristics is the voice. Males have substantially lower fundamental and formant frequencies than women (Baken, 1987; Puts, 2005; Titze, 1994) due to testosterone-driven vocal fold en- largement and vocal tract lengthening during male puberty (Fitch & Giedd, 1999; Kahane, 1978; Harries, Hawkins, Hacking, & Hughes, 1998; Vuorenkoski, Lenko, Tjernlund, & Vuorenkoski, 1978). Lower male voices are perceived as belonging to men who are older (Collins, 2000; Feinberg, Jones, Little, Burt, & Perrett, 2005), larger (Collins, 2000; Feinberg et al., 2005), more physically and socially dominant (Puts, Gaulin, & Verdolini, 2006; Puts, Hodges, Cárdenas, & Gaulin, 2007; Sell et al., 2010; Wolff & Puts, 2010), and more attractive to women (Collins, 2000; Feinberg et al., 2005; Puts, 2005). These percep- tions carry important reproductive, social, economic, and political con- sequences (Apicella, Feinberg, & Marlowe, 2007; Hodges-Simeon, Gaulin, & Puts, 2010; Klofstad, Anderson, & Peters, 2012; Tigue, Borak, O’Connor, Schandl, & Feinberg, 2012). According to Hamilton and Zuk (1982), observers attend to conspic- uous sexually dimorphic traits like the voice because they provide im- portant information about the condition or quality of the speaker. In this model, quality- or condition-dependent characteristics provide conspecifics with honest signals of the bearer's genetic resistance to parasites due to the costs involved in building and maintaining the dis- play trait (Zahavi, 1975). Folstad and Karter (1992) later proposed a proximate mechanism for this model; according to their ‘immunocom- petence handicap hypothesis’ only males who are highly immunocom- petent can tolerate the high testosterone (T) levels required to fully express their conspicuous characters. Experimental manipulation of T in animal models shows that high levels of T down-regulate humoral and cell mediated immune function (Duffy, Bentley, Drazen, & Ball, 2000) further supporting the idea that T imposes an immunosuppres- sant burden that only can be sustained by high quality individuals. Amendments to this view highlight the role of T as immune- modulatory rather than -suppressive (Braude, Tang-Martinez, & Taylor, 1999; Da Silva, 1999; McDade, 2003), upregulating investment in sec- ondary sex characteristics like the voice only in response to surplus im- mune and energetic capacity (Bribiescas, 2001; Ellison, 2001; Muehlenbein & Bribiescas, 2005). In this way, T—and the signal traits developmentally influenced by T—can serve as costly indicators of quality-dependent condition to potential mates and competitors. Among adults (Dabbs & Malinger, 1999; Evans, Neave, Wakelin, & Hamilton, 2008; Hamdan et al., 2012; Puts, Apicella, & Cárdenas, Evolution and Human Behavior xxx (2015) xxx–xxx ⁎ Corresponding author. Department of Anthropology, Boston University, 232 Bay State Road, Boston, MA 02215. Tel.: +1 627 358 0227. E-mail address: crhodges@bu.edu (C.R. Hodges-Simeon). http://dx.doi.org/10.1016/j.evolhumbehav.2015.01.002 1090-5138/© 2015 Elsevier Inc. All rights reserved. Contents lists available at ScienceDirect Evolution and Human Behavior journal homepage: www.ehbonline.org Please cite this article as: Hodges-Simeon, C.R., et al., The low male voice is a costly signal of phenotypic quality among Bolivian adolescents, Evo- lution and Human Behavior (2015), http://dx.doi.org/10.1016/j.evolhumbehav.2015.01.002