Skip to main content
NCBI home page
Behavioral Ecology logoLink to Behavioral Ecology
. 2014 May 12;25(4):984–988. doi: 10.1093/beheco/aru081

Vocal fundamental and formant frequencies are honest signals of threat potential in peripubertal males

Carolyn R Hodges-Simeon a,✉, Michael Gurven a, David A Puts b, Steven JC Gaulin a
PMCID: PMC4095947  PMID: 25024638

Lay Summary

Men with lower voices are perceived as more physically dominant; however, it is unclear why, and whether this perception is accurate. We examine whether vocal frequencies are more strongly related to measures of threat potential (size or strength) among adolescents. Results show that vocal frequencies are more informative about strength than about size. These findings suggest that listeners ascribe greater dominance to low male voices because low voices are reliably associated with strength.

Key words: costly signaling, formants, fundamental frequency, pitch, voice.

Abstract

Fundamental and formant frequencies influence perceived pitch and are sexually dimorphic in humans. The information content of these acoustic parameters can illuminate the forces of sexual selection shaping vocal sex differences as well as the mechanisms that ensure signal reliability. We use multiple regression to examine the relationships between somatic (height, adiposity, and strength) and acoustic (fundamental frequency [F 0], formant position [P f], and fundamental frequency variation [F 0-SD]) characteristics in a sample of peripubertal Bolivian Tsimane. Results indicate that among males—but not females—strength is the strongest predictor of F 0 and P f and that F 0 and P f are independent predictors of strength when height and adiposity are controlled. These findings suggest that listeners may attend to vocal frequencies because they signal honest, nonredundant information about male strength and threat potential, which are strongly related to physical maturity and which cannot be ascertained from visual or other indicators of height or adiposity alone.

INTRODUCTION

Two acoustic attributes—fundamental frequency (F 0) and formant structure (formant dispersion and formant position; Fitch 1997; Puts et al. 2012)—appear particularly salient in mammalian vocal communication. Low F 0 and low, narrow formant structure contribute to the auditory perception of a deeper-sounding voice. The length and mass of the vocal folds are developmentally responsive to testosterone (Pedersen et al. 1986) and jointly determine F 0 (Titze 1994), which is the primary determinant of perceived pitch. The dimensions of the supralaryngeal vocal tract (including the pharynx, nasal cavities, and mouth) determine the formants (Fitch 1997), which are energy peaks in the harmonic spectrum (Fitch 1997). Lower and more closely spaced formants give a deeper, more “resonant” character to the voice—analogous to the fuller and deeper resonance of a tuba versus a trumpet. Both anatomical clusters, and hence F 0 and formant structure, are sexually dimorphic in humans (Gonzalez 2004; Rendall et al. 2005; Puts et al. 2012). Recent research suggests that variation in F 0 (F 0-SD) might also be sexually dimorphic (Puts et al. 2012) and sexually selected (Hodges-Simeon et al. 2010, 2011), with dominant males speaking in a monotone manner with less variable F 0 (Puts et al. 2012).

Researchers studying nonhuman primate vocalizations (Rendall et al. 2005; Ey et al. 2007) and the human voice (Pedersen et al. 1986; Titze 1994; Fitch 1997; Evans et al. 2006; Sell et al. 2010; Puts et al. 2012) have speculated about the information content of these masculine vocal parameters. Such auditory features may convey honest information about a signaler if either the costs of production are high (“handicaps”) (Zahavi 1975) or the signal is constrained by other aspects of the phenotype (“indices”) (Fitch 1997; Fitch and Giedd 1999). When vocalizations honestly convey fitness-relevant information, conspecific listeners benefit by attending to them (Searcy and Nowicki 2005). Thus, to clarify how selection has shaped sexually dimorphic vocal features—both anatomical and acoustical—one must determine the information content and honesty of the associated signals (Fitch 1997; Rendall et al. 2005; Evans et al. 2006; Ey et al. 2007; Sell et al. 2010; Puts et al. 2012). Research on pandas (Charleton et al. 2010, 2011), red deer (Reby and McComb 2003; Reby et al. 2005), elephant seals (Sanvito et al. 2007), and primates (Rendall et al. 2005; Ey et al. 2007), among other species, suggests that conspecifics attend to fundamental and formant frequencies in courtship and competitive interactions as honest indicators of fitness-relevant characteristics such as age, sex, body size, or hormonal status.

As a basis for an index-based mechanism of signal reliability among humans, initial approaches to the information content of F 0 and formant structure suggested a possible allometry between overall body size and vocal tract dimensions. Such a relationship would make vocal signals partly redundant with visual observation of height and/or weight. Although inconclusive, previous research suggests the allometry assumption may be incorrect: Although several studies have found associations between height (Graddol and Swann 1983; Puts et al. 2012) or weight (Evans et al. 2006) and F 0, more have reported no significant association between measures of overall body size and this vocal parameter (e.g., Collins 2000; Gonzalez 2004; Rendall et al. 2005). Research on the relationship between body size and formant structure is similarly inconclusive: Several studies have reported positive associations (Gonzalez 2004; Evans et al. 2006; Puts et al. 2012), but others have not (Collins 2000; Rendall et al. 2005; Sell et al. 2010). Recently, Puts et al. (2012) proposed a new calculation of formant structure, formant position, which may improve on previous measures and thus clarify the relationship between format structure and body size.

The focus on body size in the human and nonhuman literatures arises from an assumed relationship between size and threat potential among males. Because greater height and weight are often associated with success in physical contests (e.g., Sanvito et al. 2007), competitors can adaptively base fight-or-flight decisions on assessments of body size (Maynard Smith and Parker 1976). Although larger size has advantages in combat (Reilly and Secher 1990; Katic et al. 2005), Sell et al. (2010) suggest that body size often usefully reveals fighting ability because it is reliably associated with muscular strength. In humans, upper-body strength predicts a history of fighting better than height or weight; in addition, perceptions of fighting ability from photographs track strength better than they track height or weight (Sell et al. 2009). For these reasons, vocal parameters may have evolved to signal threat potential via their association with fighting ability, which is more strongly related to strength (Sell et al. 2009; Puts et al. 2012) than to body size. Puts et al. (2012) examined US and Hadza adult males, finding an association between formants and strength in the US sample only, and between F 0 and strength among the Hadza only. Sell et al. (2010) found relationships between perceptions of strength and both acoustic dimensions, but no associations with actual strength.

Here, we attempt to identify the information content of sexually dimorphic human vocal parameters by examining relationships among somatic variables (height, adiposity, and upper-body strength) and 3 sexually dimorphic acoustic characteristics (F 0, formant position, and F 0-SD) in a sample of male and female peripubertal Tsimane. Most studies target adult men (e.g., Evans et al. 2006; Sell et al. 2010; Puts et al. 2012); but puberty heralds a dramatic change in both acoustic and somatic characteristics (Hodges-Simeon et al. 2013). If male vocal parameters honestly signal physical formidability, then vocal and anatomical characteristics should change in parallel with measures of physical formidability—such as height, weight, and strength—during and after puberty. This period also offers a greater range in vocal frequencies, sizes and strengths than can be found in adult samples, more accurately reflecting the range of these parameters found in natural human groups. Because physical formidability may be more dependent on strength than on mere size, we expect low male voice to better predict strength than it predicts height or weight.

METHODS

Participants

Seventy-nine male and 76 female Tsimane (age 8–23; M = 13.6, SD = 3.3) participated. The Tsimane are forager-horticulturalists living in central Bolivia (Gurven et al. 2007). Research was approved by the Human Subjects Committee at the University of California, Santa Barbara. All procedures were conducted in accordance with ethical standards, including assent by all participants and informed consent by their parents.

Ages were estimated using participants’ stated age and birth date, which were verified against the Tsimane Health and Life History census. When the 2 ages were in conflict, the census age was used (see Gurven et al. 2007; Hodges-Simeon et al. 2013 for detailed age-assignment methods). Because the onset and length of pubertal development can vary widely across individuals and populations (Bogin 1999), a liberal age range was used to capture all potential peripubertal somatic and vocal variation. Analyses are performed on the full sample as well as a more limited age range (11–19; see Results).

Anthropometric measures

Body and strength measurements were collected by CHS using standard anthropometric protocols (Frisancho 1990). Right-side measurements were recorded in duplicate and averaged for analyses. Grip strength was measured to the nearest 0.5 psi using a Baseline bulb pneumonic hand dynamometer. Flexed-biceps size was recorded to the nearest 0.2cm using an anthropometric tape measure. Triceps and subscapular skinfolds were measured to the nearest 0.2mm using a Harpenden caliper. Skinfolds were used to calculate adiposity using the Slaughter formula for peripubertal individuals (Slaughter et al. 1988); arm fat area (another measure of adiposity; Frisancho 1990) was substituted for Slaughter adiposity into all models without any appreciable change in the results. We used adiposity instead of weight in our analyses due to the high correlation (r = 0.95) and multicollinearity (variation inflation factor [VIF] > 10) between weight and height. A composite strength measure was created by averaging standardized grip strength and flexed-bicep circumference; this composite has been used to index upper-body strength in previous voice research (Sell et al. 2009; Puts et al. 2012).

Voice measurement and analysis

Participants were shown 5 photographs of common objects in Tsimane life (cat, dog, plantains, soccer ball, and frog) and asked to name each object in sequence. The spoken sample was analyzed as one single voice sample. These words were selected because they terminate in 5 different vowel sounds (míshi: “ee”; açhuj: “oo”; pe’re: “ā”; perota: “ah”; and ococo: “oh”). Because each vowel sound has a different formant profile, the full voice clip for each participant represents a normal range of formant values. A Sony PCM-M10 digital audio recorder with a sampling rate of 44100 Hz and 16-bit quantization recorded voice samples in mono. A headset-mounted Audio-Technica lavalier microphone was placed 5cm from the lips for each participant. Recordings were saved as high-quality uncompressed linear PCM.wav files.

For each participant’s recording, mean F 0 (and standard deviation; F 0-SD) and formant structure (P f) were measured in Hertz (Hz) using Praat voice analysis software (Version 5.1.37; Boersma and Weenik 2010). Formants were obtained using methods described by Puts et al. (2012); however, formant ceilings were adjusted for different developmental groups (see below; Boersma and Weenik 2010). Participants completed the Pubertal Development Scale (Petersen et al. 1988), a self-report scale that has been shown to be a reliable approximation of the Tanner stages (Brooks-Gunn et al. 1987; Petersen et al. 1988; Coleman and Coleman 2002). The Tanner stages are 5 stages of reproductive development based on genital and pubic hair maturation and are widely used to mark pubertal development (Brooks-Gunn et al. 1987; Petersen et al. 1988; Coleman and Coleman 2002). Males and females in stages 1 and 2 (mean age = 10.9±1.6) were measured using a formant ceiling of 7000 Hz. Females in stages 3, 4, and 5 (mean age = 15.4±2.7) and males in stage 3 (mean age = 15±1.5) were measured using a formant ceiling of 6000 Hz. And males in stages 4 and 5 (mean age = 18.2±2.3) were measured with a formant ceiling of 5000 Hz. F1 through F4 were measured at each glottal pulse, targeting voiced speech only, and then averaged for analyses (mean number of glottal pulses per file = 527±273). In order to calculate P f, the first 4 formants were standardized using between-sex means and standard deviations (Puts et al. 2012).

RESULTS

Although males were the primary targets of our analyses, we conducted parallel analyses on females in order to provide a comparative baseline. A log transformation adequately corrected for nonnormality in the distributions of height, strength, age, F 0, P f, and F 0-SD. Height, adiposity, and strength were entered together into separate multiple regression models to predict F 0, P f, F 0-SD (Table 1, Models 1, 2, 3). For males only, strength was a significant predictor of F 0 (β = −0.62, P < 0.01); height (β = −0.25, ns) and adiposity (β = 0.04, ns) were both nonsignificant. Strength was the only significant predictor of male P f (β = −0.70, P < 0.01), and height was the only significant predictor of female P f (β = −0.53, P < 0.01). Both adiposity (β = −0.47, P < 0.01) and height (β = 0.63, P < 0.05) were significant predictors of F 0-SD for females. Effect sizes were large. Height, adiposity, and strength explained 71% of the variance in male F 0 and 68% of the variance in male P f. These 3 variables explained 21% of the variance in female F 0 and 62% of the variance in female P f.

Table 1.

Multiple regression models

Males Females
Standardized beta coefficients (t statistic) Zero-order correlations r Partial correlations Standardized beta coefficients (t statistic) Zero-order correlations r Partial correlations
1. Predicting F 0 a
 Height –0.25 (–1.28) –0.82*** –0.15 –0.02 (–0.07) –0.42*** –0.01
 Adiposity 0.04 (0.50) –0.34** 0.06 0.04 (0.50) –0.41** –0.10
 Strength –0.62 (–3.19**) –0.84*** –0.34** –0.32 (–1.15) –0.45*** –0.13
2. Predicting P f b
 Height –0.17 (–0.87) –0.79*** –0.10 –0.53 (–3.06**) –0.78*** –0.34**
 Adiposity 0.08 (1.17) –0.29** 0.13 –0.11 (–0.88) –0.64*** –0.10
 Strength –0.70 (–3.41**) –0.82*** –0.36** –0.19 (–0.99) –0.75*** –0.12
3. Predicting F 0-SDc
 Height –0.62 (–1.81†) –0.18 –0.20† 0.63 (2.47*) 0.13 0.27*
 Adiposity –0.08 (–0.62) –0.12 –0.07 –0.47 (–2.64**) –0.15 –0.30**
 Strength 0.48 (1.38) –0.12 0.16 –0.17 (–0.61) 0.03 –0.07
4. Predicting strengthd
 Height 0.61 (8.15***) 0.87*** 0.70*** 0.56 (6.31***) 0.91*** 0.60***
 Adiposity 0.09 (2.56*) 0.33*** 0.29* 0.20 (2.90**) 0.80*** 0.33**
 F 0 –0.17 (–2.23**) –0.84*** –0.30** –0.05 (–1.07) –0.45*** –0.13
  P f –0.11 (–2.02†) –0.82*** –0.21† –0.02 (–0.26) –0.75*** –0.03
  F 0-SD 0.10 (2.80*) –0.12 0.31** –0.02 (–0.45) 0.03 –0.05
 Age 0.11 (2.00) 0.87*** 0.18 0.21 (2.80**) 0.81*** 0.32**
Open in a new tab

Males: a F(3,79) = 62.00***; R 2 = 0.71. b F(3,79) = 54.60***; R 2 = 0.68. c F(3,79) = 1.72; R 2 = 0.06. d F(6,79) = 156.85***; R 2 = 0.93. Females: a F(3,76) = 6.47**; R 2 = 0.21. b F(3,76) = 39.24***; R 2 = 0.62. c F(3,76) = 4.17**; R 2 = 0.15. d F(6,76) = 81.90***; R 2 = 0.88.

† P < 0.08, *P < 0.05, **P < 0.01, ***P < 0.001.

Having established that strength is associated with F 0 and P f among males, we explored whether vocal characteristics explain unique variance in strength that is not explained by height and adiposity. Thus, F 0, P f, F 0-SD, age, height, and adiposity were used to predict strength. In this model, F 0 (β = −0.17, P < 0.01), F 0-SD (β = 0.10, P < 0.05), height (β = 0.61, P < 0.001), and adiposity (β = 0.09, P < 0.05) were statistically significant (Table 1, Model 4). P f approached significance (β = −0.11, P = 0.07). Among females, none of the vocal parameters explained any additional variance in strength beyond the effects of height (β = 0.56, P < 0.001) and adiposity (β = 0.20, P < 0.01). This model explained 93% of the variance in male strength and 88% of the variance in female strength.

Because the boundaries of pubertal development vary across populations, all analyses (Models 1 through 4) were rerun with a narrower age range: 11–19 (N = 63 for both males and females). Results remained largely unchanged (i.e., only small changes in beta and P values). In only Model 4, the P value for one predictor, F 0-SD, changed from P < 0.05 to P = 0.07.

DISCUSSION

For peripubertal males, but not females, aspects of voice pitch provide additional information about strength that is not revealed by height or adiposity. Controlling for these 2 size variables as well as age, males with lower voices (as indicated by low F 0 and P f) were significantly stronger than males with higher voices. This study is the first to show that both F 0 and P f signal additive, nonredundant information to the listener about male strength that cannot be gained from information about height or adiposity. These findings also suggest that listeners can make accurate assessments of strength, and hence probably physical dominance, when they attend to low F 0 or P f (cf., Puts et al. 2007; Hodges-Simeon et al. 2010) in peripubertal males.

The signal content of fundamental and formant frequencies has been of continuing interest (Graddol and Swann 1983; Fitch and Giedd 1999; Collins 2000; Sanvito et al. 2007; Sell et al. 2009). Across many mammals (e.g., Sanvito et al. 2007), vocalizations mediate access to resources and mates (Maynard Smith and Parker 1976; Searcy and Nowicki 2005). Theory suggests that vocalizations should be ignored unless they carry honest information about the vocalizer’s phenotype—information useful to competitors or potential mates. Our findings suggest that both F 0 and P f signal strength, providing additive information about potential competitors’ formidability.

Previous research on the value of fundamental and formant frequencies for intrasexual competition among humans has been mixed (Graddol and Swann 1983; Collins 2000; Gonzalez 2004; Rendall et al. 2005; Evans et al. 2006; Puts et al. 2012). We improve on past research in 2 ways. First, with a few exceptions (Sell et al. 2010; Puts et al. 2012), past studies measure height and weight, but not strength. By examining multiple measures of threat potential, it is possible to identify which features best predict vocal frequencies and to examine the unique contributions of different acoustic parameters to strength measures. Second, we target peripubertal ages rather than adulthood. This age group provides valuable information on this topic, as tremendous change in both vocal and somatic features occurs during puberty (Hodges-Simeon et al. 2013). In addition, because mating competition intensifies during this period, attention to signals of dominance should be particularly high.

Contrary to predictions, F 0-SD was positively associated with strength when other vocal parameters were controlled. That is, males with less monotone, more variable F 0 were stronger than those with less variable F 0. This finding is in contrast to conclusions made in previous research on the value of F 0 variation for dominance signaling (Hodges-Simeon et al. 2010, 2011). That is, F 0 variation may leak information on autonomic activation (i.e., nervousness and arousal) and therefore may only be useful during ecologically relevant social interactions (e.g., Hodges-Simeon et al. 2010, 2011). Clearly, more research on F 0 variation is needed.

Signaling systems are vulnerable to cheaters and will be evolutionarily stable only if constraints ensure reliability (Grafen 1990; Searcy and Nowicki 2005). Fitch and Giedd (1999) argue that formant structure is an index of height; that is, allometry with linear dimensions assures honesty. If such developmental constraints dominated, formant structure would scale with linear growth and this vocal feature would be a redundant indicator of height. Although P f and F 0 are correlated with height in both sexes, the results presented here show that when both strength and body size (i.e., height and adiposity) are entered simultaneously into a multiple regression, strength is the only significant predictor of P f and of F 0. These results support the hypothesis that vocal frequencies signal physical formidability, which is more strongly related to strength than to body size.

A potential concern is whether these findings generalize to other populations. The Tsimane live in an energetically demanding, highly infectious environment, and it is possible that our results may not extend to other populations. However, because the Tsimane ecology more closely aligns with the environment in which humans evolved, these data may provide a closer representation of the selective forces on male vocal anatomy than data from industrialized countries (Henrich et al. 2010). This is the first study to examine the relationships among peripubertal strength, body size, and sexually dimorphic vocal parameters in a nonindustrialized population. By targeting interpopulation variation in somatic and acoustic characteristics, future studies will reveal the extent to which ecology influences the information content of vocal parameters.

A further consideration is the extent to which these results generalize beyond puberty. Many studies have examined the relationship between body size (height and weight) and vocal characteristics among adult men, with mixed results (Graddol and Swann 1983; Collins 2000; Gonzalez 2004; Rendall et al. 2005; Evans et al. 2006; Puts et al. 2012). A question of continuing interest is why adult listeners assign larger size (i.e., greater height and weight) to lower-sounding vocalizations if the empirical relationship between them is so weak. Rendall et al. (2007) propose 3 hypotheses to address this question: 1) listeners overgeneralize from age- and sex-based classifications, 2) listeners overgeneralize from broader environmental sound-size associations, or 3) the voice is closely aligned with other dimensions of a speaker and that these associations intrude on our perceptions of size. For instance, F 0 and P f appear to be linked with testosterone (Pedersen et al. 1986; Dabbs and Mallinger 1999), which itself may affect traits like strength. Several studies have examined the association between strength and vocal parameters; however, like height and weight, the results have been inconclusive (Sell et al. 2010; Puts et al. 2012).

The results of the present study support Rendall’s first and third hypotheses. In addition, these findings suggest an additional explanation: Listeners may attend to low-voiced speakers because it provides a record of energetic condition and formidability during pubertal development. In a previous study using these data, modeling the developmental trajectory of vocal characteristics showed that individuals in better condition have an earlier voice change and a faster rate of change than those in poorer condition. This precocious developmental trajectory may lead to a lower voice in adulthood (Hodges-Simeon et al. 2013), suggesting that adult vocal characteristics may reveal pubertal condition. This may be useful information to potential mates and competitors if it co-occurs with characteristics such as heritable quality or coalitional strength. These are questions for future research.

In summary, our findings indicate that both F 0 and P f contribute nonredundant, additive information about physical strength, beyond what is revealed by height or adiposity, in peripubertal males. These results suggest that humans attend to these auditory characteristics because, on average, they indicate something about male formidability that cannot be ascertained from visual inspection of height or adiposity. Because contest competition is far less intense in women than in men (Daly and Wilson 1983; Puts 2010), this information is less useful to females; accordingly, F 0 and P f contribute no additional information about female strength. The relationship between strength and vocal parameters suggests that sexual selection may have targeted F 0 and P f in the evolution of men’s voices.

FUNDING

This work was supported by a Wenner-Gren dissertation fieldwork grant. The Tsimane Health and Life History project is supported by grants from National Science Foundation (BCS-0136761) and National Institutes of Health/National Institute on Aging (R01AG024119-01, R56AG024119-06, R01AG024119-07).

Acknowledgments

The authors extend many thanks to the Tsimane participants and their families, as well as the Tsimane Health and Life History researchers and staff.

REFERENCES

  1. Bogin BA. 1999. Patterns of human growth. Cambridge (UK): Cambridge University Press [Google Scholar]
  2. Boersma P, Weenick D. 2010. Praat: doing phonetics by computer, Version 5.1.37 [cited 2010 June 23]. Available from http://wwwpraat.org/
  3. Brooks-Gunn J, Warren MP, Rosso J, Gargiulo J. 1987. Validity of self-report measures of girls’ pubertal status. Child Dev. 58:829–841 [PubMed] [Google Scholar]
  4. Charleton BD, Keating JL, Kersey D, Rengui Li, Huang Y, Swaisgood RR. 2011. Vocal cues to male androgen levels in giant pandas. Bio Letters. 7:71–74 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Charleton BD, Zhihe Z, Snyder RJ. 2010. Giant pandas perceive and attend to formant frequency variation in male bleats. Anim Behav. 79:1221–1227 [Google Scholar]
  6. Coleman I, Coleman J. 2002. The measurement of puberty: a review. J Adolesc. 25:535–550 [DOI] [PubMed] [Google Scholar]
  7. Collins SA. 2000. Men’s voices and women’s choices. Anim Behav. 60:773–780 [DOI] [PubMed] [Google Scholar]
  8. Dabbs JM, Mallinger A. 1999. High testosterone levels predict low voice pitch among men. Pers Indiv Differ. 27:801–804 [Google Scholar]
  9. Daly M, Wilson MI. 1983. Sex, evolution and behavior, 2nd ed. Boston (MA): Willard Grant Press. [Google Scholar]
  10. Evans S, Neave N, Wakelin D. 2006. Relationships between vocal characteristics and body size and shape in human males: an evolutionary explanation for a deep male voice. Biol Psychol. 72:160–163 [DOI] [PubMed] [Google Scholar]
  11. Ey E, Pfefferle D, Fischer J. 2007. Do age- and sex-related variations reliably reflect body size in non-human primate vocalizations? A review. Primates. 48:253–267 [DOI] [PubMed] [Google Scholar]
  12. Fitch WT. 1997. Vocal tract length and formant frequency dispersion correlate with body size in rhesus macaques. J Acoust Soc Am. 102:1213–1222 [DOI] [PubMed] [Google Scholar]
  13. Fitch WT, Giedd J. 1999. Morphology and development of the human vocal tract: a study using magnetic resonance imaging. J Acoust Soc Am. 106:1511. [DOI] [PubMed] [Google Scholar]
  14. Frisancho AR. 1990. Anthropometric standard for the assessment of growth and nutritional status. Ann Arbor (MI): University of Michigan Press [Google Scholar]
  15. Gonzalez J. 2004. Formant frequencies and body size of speaker: a weak relationship in adult humans. J Phonetics. 32:277–287 [Google Scholar]
  16. Graddol D, Swann J. 1983. Speaking fundamental frequency: some physical and social correlates. Lang Speech. 26:351–366 [DOI] [PubMed] [Google Scholar]
  17. Grafen A. 1990. Biological signals as handicaps. J Theor Biol. 144:517–546 [DOI] [PubMed] [Google Scholar]
  18. Gurven M, Kaplan H, Supa AZ. 2007. Mortality experience of Tsimane Amerindians of Bolivia: regional variation and temporal trends. Am J Hum Biol. 19:376–398 [DOI] [PubMed] [Google Scholar]
  19. Henrich J, Heine SJ, Norenzayan A. 2010. The weirdest people in the world? Behav Brain Sci. 33:1–75 [DOI] [PubMed] [Google Scholar]
  20. Hodges-Simeon CR, Gaulin SJC, Puts DA. 2010. Different vocal parameters predict perceptions of dominance and attractiveness. Hum Nat. 21:406–427 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Hodges-Simeon CR, Gaulin SJC, Puts DA. 2011. Voice correlates of mating success in men: examining “contests” versus “mate choice” modes of sexual selection. Arch Sex Behav. 40:551–557 [DOI] [PubMed] [Google Scholar]
  22. Hodges-Simeon CR, Gurven M, Cárdenas RA, Gaulin SJC. 2013. Voice change as a new measure of male pubertal timing: a study among Bolivian adolescents. Ann Hum Biol. 40:209–219 [DOI] [PubMed] [Google Scholar]
  23. Katic R, Blazevic S, Krstulovic S, Mulic R. 2005. Morphological structures of elite karateka and their impact on technical and fighting efficiency. Coll Antropol. 29:79–84 [PubMed] [Google Scholar]
  24. Maynard Smith J, Parker GA. 1976. The logic of asymmetric contests. Anim Behav. 24:159–175 [Google Scholar]
  25. Pedersen MF, Moller S, Krabbe S, Bennett P. 1986. Fundamental voice frequency measured by electroglottography during continuous speech. A new exact secondary sex characteristic in boys in puberty. Int J Pediatr Otorhinolaryngol. 11:21–27 [DOI] [PubMed] [Google Scholar]
  26. Petersen AC, Crocket I, Richards M, Boxer A. 1988. A self-report measure of pubertal status--reliability, validity, and initial norms. J Youth Adolesc. 17:117–133 [DOI] [PubMed] [Google Scholar]
  27. Puts DA. 2010. Beauty and the beast: mechanisms of sexual selection in humans. Evol Hum Behav. 31:157–175 [Google Scholar]
  28. Puts DA, Apicella CL, Cárdenas RA. 2012. Masculine voices signal men’s threat potential in forager and industrial societies. Proc R Soc B Biol Sci. 279:601–609 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Puts DA, Hodges CR, Cárdenas RA, Gaulin SJC. 2007. Men’s voices as dominance signals: vocal fundamental and formant frequencies influence dominance attributions among men. Evol Hum Behav. 28:340–344 [Google Scholar]
  30. Reby D, McComb K. 2003. Anatomical constraints generate honesty: acoustic cues to age and weight in the roars of red deer stags. Anim Behav. 65:519–530 [Google Scholar]
  31. Reby D, McComb K, Cargnelutti B, Darwin C, Fitch WT, Clutton-Brock T. 2005. Red deer stags use formants as assessment cues during intrasexual agonistic interactions. Proc R Soc B Biol Sci. 272:941–947 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Reilly T, Secher N. 1990. Physiology of sports: an overview. In: Reilly T, Secher N, Snell P, Williams C. editors. Physiology of sports. London (UK): E & FN Spon. [Google Scholar]
  33. Rendall D, Kollias S, Ney C, Lloyd P. 2005. Pitch (F) and formant profiles of human vowels and vowel-like baboon grunts: the role of vocalizer body size and voice-acoustic allometry. J Acoust Soc Am. 117:944–955 [DOI] [PubMed] [Google Scholar]
  34. Rendall D, Vokey JR, Nemeth C. 2007. Lifting the curtain on the Wizard of Oz: biased voice-based impressions of speaker size. J Exp Psych. 33:1208–1219 [DOI] [PubMed] [Google Scholar]
  35. Sanvito S, Galimberti F, Miller E. 2007. Vocal signalling of male southern elephant seals is honest but imprecise. Anim Behav. 73:287–299 [Google Scholar]
  36. Searcy WA, Nowicki S. 2005. The evolution of animal communication: reliability and honesty in signaling systems. Princeton (NJ): Princeton University Press [Google Scholar]
  37. Sell A, Bryant GA, Cosmides L, Tooby J, Sznycer D, von Rueden C, Krauss A, Gurven M. 2010. Adaptations in humans for assessing physical strength from the voice. Proc R Soc B Biol Sci. 277:3509–3518 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Sell A, Cosmides L, Tooby J, Sznycer D, Rueden CV, Gurven M. 2009. Human adaptations for the visual assessment of strength and fighting ability from the body and face. Proc R Soc B Biol Sci. 276:575–584 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Slaughter MH, Lohman TG, Boileau RA, Horswill CA, Stillman RJ, Van Loan MD, Bemben DA. 1988. Skinfold equations for estimation of body fatness in children and youth. Hum Biol. 60:709–723 [PubMed] [Google Scholar]
  40. Titze IR. 1994. Principles of voice production. Englewood Cliffs (NJ): Prentice Hall [Google Scholar]
  41. Zahavi A. 1975. Mate selection: a selection for a handicap. J Theor Biol. 53:205. [DOI] [PubMed] [Google Scholar]

Articles from Behavioral Ecology are provided here courtesy of Oxford University Press

RESOURCES