at SciVerse ScienceDirect

Animal Behaviour xxx (2012) 1e10

Contents lists available

Animal Behaviour

journal homepage: www.elsevier .com/locate/anbehav

Faking it: deliberately altered voice pitch and vocal attractiveness

Paul J. Fraccaro a, Jillian J. M. O’Connor a, Daniel E. Re b, Benedict C. Jones c, Lisa M. DeBruine c,David R. Feinberg a,*

aDepartment of Psychology, Neuroscience & Behaviour, McMaster University, Hamilton, ON, Canadab School of Psychology, University of St Andrews, St Andrews, U.K.c Institute of Neuroscience and Psychology, University of Glasgow, Glasgow, U.K.

a r t i c l e i n f o

Article history:Received 21 March 2012Initial acceptance 9 May 2012Final acceptance 28 September 2012Available online xxxMS. number: A12-00235R

Keywords:functional honestyhonest signalvocal attractivenessvocal dominancevoice pitch

* Correspondence: D. R. Feinberg, Department ofBehaviour, McMaster, University, Hamilton, ON L8S 4

E-mail address: feinberg@mcmaster.ca (D. R. Feinb

0003-3472/$38.00 � 2012 The Association for the Stuhttp://dx.doi.org/10.1016/j.anbehav.2012.10.016

Please cite this article in press as: Fraccaro,(2012), http://dx.doi.org/10.1016/j.anbehav.2

Previous research has shown that men prefer higher-pitched women’s voices and women prefer lower-pitched men’s voices. However, both men and women can modulate their voice pitch, which can affectothers’ perceptions of the voice. Here we tested whether deliberate pitch changes affect speakers’ vocalattractiveness. Our results suggest that deliberately exaggerating sex-typical voice pitch (i.e. loweringpitch in men and raising pitch in women) does not necessarily increase vocal attractiveness but thatexaggerating sex-atypical voice pitch (i.e. raising pitch in men and lowering pitch in women) maydecrease vocal attractiveness. By contrast with these findings for attractiveness, listeners interpretedlowered-pitch voices as sounding more dominant than habitually pitched voices in same-sex voices,which may aid in avoiding the costs associated with intrasexual competition. These findings suggest thatthe way humans perceive deliberate manipulations of voice pitch can mitigate the potential costs ofusing an alterable cue to assess attractiveness, and that functional honesty may only evolve in domainswhere such honesty would be favourable to perceivers.� 2012 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved.

Fundamental frequency is tied to the rate of vibration of thevocal folds (see Titze 1994 for an overview of vocal production).Here we refer to the perception of fundamental frequency and/orthe resulting harmonic spectrum as ‘pitch’. Previous research hasdemonstrated that, in general, men prefer relatively high-pitchedwomen’s voices (Collins & Missing 2003; Feinberg et al. 2008b;Jones et al. 2008, 2010; Borkowska & Pawlowski 2011; Puts et al.2011). Men’s attractiveness ratings of women’s voices are posi-tively correlated with women’s voice pitch (Collins &Missing 2003;Feinberg et al. 2008b), and men prefer women’s voices manipu-lated to have higher voice pitch to voices manipulated to havelower voice pitch (Feinberg et al. 2008b; Apicella & Feinberg 2009;Jones et al. 2010; O’Connor et al. 2011; Puts et al. 2011). Conversely,women prefer lower-pitched men’s voices (Collins 2000; Feinberget al. 2005; Hodges-Simeon et al. 2010). Women’s attractivenessjudgments of men’s voices are negatively associated with men’svoice pitch (Collins 2000), and women prefer experimentallymanipulated lower-pitched voices to higher-pitched voices(Feinberg et al. 2005; Puts 2005; Vukovic et al. 2008, 2010a, 2011;Hodges-Simeon et al. 2010; Jones et al. 2010; O’Connor et al. 2011,2012).

Psychology, Neuroscience &K1, Canada.erg).

dy of Animal Behaviour. Published

P. J., et al., Faking it: deliber012.10.016

Preferences for exaggerated sex-typical traits may reflect pref-erences for high-quality mates (Thornhill & Gangestad 1999;Feinberg 2008). Masculine characteristics are positively related toindices of long-term health (Rhodes et al. 2003; Thornhill &Gangestad 2006; Gangestad et al. 2010) and physical strength(Fink et al. 2007), and may advertise the robustness of an individ-ual’s immune system (Moore et al. 2011; Rantala et al. 2012).Moreover, among men, exaggerated sex-typical (i.e. masculine)vocal and facial characteristics have positive effects on perceptionsof dominance (Perrett et al. 1998; Feinberg et al. 2006; Puts et al.2006, 2007, 2012; Jones et al. 2010; Watkins et al. 2010). Sincevocal and facial masculinity are positively correlated among men(Saxton et al. 2006, 2009) andmay share a common hormonal basis(Harries et al. 1997; Dabbs & Mallinger 1999; Verdonck et al. 1999;Penton-Voak & Chen 2004; Roney et al. 2006; Evans et al. 2008),women’s preferences for lower-pitched men’s voices may reflecta preference for cues to long-term health and/or dominance (e.g.Puts et al. 2012).

Among women, voice pitch is negatively associated with healthrisk factors (Vukovic et al. 2010a). Women’s voice pitch is alsopositively related to oestrogen levels (Abitbol et al. 1999), which,within the normal range, are positively related to healthy repro-ductive development (Alonso & Rosenfield 2002). Between indi-viduals, ratings of women’s vocal attractiveness are positivelycorrelated with voice pitch (Collins & Missing 2003; Feinberg et al.2008b) and attractive body configurations (Collins & Missing 2003;

by Elsevier Ltd. All rights reserved.

ately altered voice pitch and vocal attractiveness, Animal Behaviour

P. J. Fraccaro et al. / Animal Behaviour xxx (2012) 1e102

Hughes et al. 2004; Vukovic et al. 2010a). For example, women’svoice pitch is negatively related to overall body mass and bodymass index (Vukovic et al. 2010a). In addition, women’s vocalattractiveness is negatively related to waist-to-hip ratio (Hugheset al. 2004), a trait that is positively associated with oestrogenlevels (Jasienska et al. 2004) and many other indices of reproduc-tive health (Singh 2002).

Although habitual pitch (i.e. natural speaking pitch) is a reliableindicator of sex hormone levels (Abitbol et al. 1999; Dabbs &Mallinger 1999), voice pitch is clearly not a fixed trait. Indeed,examples of within-individual variation in voice pitch are ubiqui-tous. During singing, the deliberate manipulation of laryngealmuscles modulates voice pitch to produce specific musical tones(Titze 1994). Voice pitch also changes according to social context.Indeed, context-dependent vocal changes are widespread amongthe animal kingdom and have been observed in anurans (Wagner1989a, b, 1992; Bee & Perrill 1996; Given 1999; Bee et al. 1999,2000; Bee & Bowling 2002; Lardner & bin Lakim 2002; Owen &Gordon 2005), bovids (Frey et al. 2008), cervids (Reby et al.2005), canids (Yin 2002), phascolarctids (i.e. koalas, Charltonet al. 2011), passerine birds (Leavesley & Magrath 2005; Goodale& Kotagama 2006) and primates (Seyfarth et al. 1980; Snowdonet al. 1983; Zuberbühler 2000; reviewed in: Tomasello &Zuberbühler 2002; Seyfarth & Cheney 2003). Among humans,speakers tend to raise their voice pitch when speaking to an infant(Fernald & Kuhl 1987; Trainor & Desjardins 2002), match their pitchto conversation partners who are higher in social status (Gregory &Webster 1996), and speak in a lower- or higher-pitched voice whenspeaking to a less dominant or more dominant listener, respectively(Puts et al. 2006). Similarly, red deer stags, Cervus elaphus, lower theformant frequencies of their roars by extending their vocal tracts tocompetitors that sound larger (Reby et al. 2005). Importantly,human speakers have been found to change their voice pitch whenviewing unattractive or attractive listeners (Hughes et al. 2010;Fraccaro et al. 2011). Fraccaro et al. (2011) asked women to leavevoicemail messages conveying romantic interest to two mendiffering in sexually dimorphic face shape, and found that womenspoke with a relatively higher-pitched voice to the type of face theypreferred. By contrast, Hughes et al. (2010) had men and womenleave voicemail messages about a psychology survey to attractiveand unattractive individuals, and found that both men and womenspoke in a relatively lower-pitched voice to the attractive individ-uals. These seemingly opposing results point to the potentiallyimportant influence of social context on vocal modulations (i.e. ina professional context: Hughes et al. 2010; in a mating context:Fraccaro et al. 2011). This research on human vocal modulations inresponse to social cues complements similar work on vocalmodulations in cervids and bovids. Fallow deer, Dama dama, reddeer, goitered gazelles, Gazella subgutturosa, and Mongoliangazelles, Procrapra gutturosa, lower their laryngeal position toproduce low vocal tract resonance frequencies in response tointruders and during the rut season (Fitch & Reby 2001; McElligottet al. 2006; Frey et al. 2008, 2011). The similarity in frequencyprofiles of vocalizations across languages and species has beensuggested to constitute evidence for a ‘frequency code’ common tomany animals (see Ohala 1983, 1984).

Although research has demonstrated that individuals modulatetheir voice pitch in mate choice (Anolli & Ciceri 2002; Fraccaro et al.2011) and dominance-related contexts (Puts et al. 2006), and thatcomputer-manipulated voice pitch affects attractiveness (e.g.Feinberg et al. 2005, 2008a, b), it is not known whether deliberatemodulations of voice pitch in the absence of social context issufficient to affect attractiveness or dominance judgments in bothsexes. Investigating this issue is potentially important, however; ifdeliberately manipulating voice pitch causes individuals to appear

Please cite this article in press as: Fraccaro, P. J., et al., Faking it: deliber(2012), http://dx.doi.org/10.1016/j.anbehav.2012.10.016

more attractive, it would cast doubt on the extent to which voicepitch is likely to be an honest cue of mate quality. By contrast,because voice pitch can carry information about the emotionalstate of a speaker (Fairbanks & Pronovost 1939; Williams & Stevens1972; Razak et al. 2003), listeners may be inclined to interpret voicepitches associated with dominance as genuine to avoid conflictwith aggressive individuals and/or harm from them. Here, weinvestigated whether deliberate alterations of voice pitch alteredattractiveness (experiment 1) or dominance (experiment 2) judg-ments when compared to a speaker’s habitual voice pitch.

EXPERIMENT 1

First, we tested whether deliberately altered voice pitch influ-ences ratings of vocal attractiveness compared to themodal pitch ofspeakers. Participants made two-alternative forced-choice judg-ments of attractiveness for a speaker’s habitual voice pitch, and hisor her raised or lowered voice pitch. We hypothesized that, if voicepitch is an honest cue of mate quality, voice pitch alterations wouldnot increase the attractiveness of speakers.

Methods

Protocols were approved by the McMaster Research EthicsBoard.

StimuliWe recorded the voices of fourwomen (mean� SE age¼ 19.25�

0.23 years) and four men (mean� SE age¼ 18.25� 0.23 years)speaking the West-Central Canadian English vowel sounds ‘eh’ as inbet, ‘ee’ as in see, ‘ah’ as in father, ‘oh’ as innote and ‘oo’ as in boot. Eachparticipantwas instructed to speak the vowels in threedifferentways:naturally (i.e. in theirhabitualpitch),witha raisedvoicepitchandwitha lowered voice pitch (see Fig. 1). Participants were instructed not toenter the falsetto andpulse registers (i.e. registers aboveandbelowthenormal speaking range, respectively), and to speak the vowels oneafteranother. Thenumberofvoicesused inour study is similar to thoseused in previous studies assessing preferences for raised and loweredvoice pitch (e.g. Feinberg et al. 2006; Fraccaro et al. 2010; Jones et al.2010; Vukovic et al. 2011). Recordings were made using a stand-mounted (via vibrational dampener) Sennheiser MKH70 micro-phone in an anechoic chamber (WhisperRoom SE 2000 Series) usingAdobe OnLocation recording software, in mono, at a sampling rate of48 kHz with 16-bit amplitude quantization. This resulted in 24 voicerecordings, each of which consisted of all five vowel sounds (each ofoureightparticipants spokeonce inhabitual pitch, once in raisedpitchand once in lowered pitch; see Tables 1, 2). Voice pitch wasmeasuredusing the autocorrelation function in Praat (Boersma & Weenink2012). We used a frequency range of 60e100 Hz for male voices and100e600 Hz for female voices (followingFeinberget al. 2005;Vukovicet al. 2010a, b). All raised-pitched voices were higher than habituallypitched voices, and all lowered-pitch voices were lower than habitu-ally pitched voices. Both the raised- and lowered-pitched voices fellwithin the normal pitch range forhuman adults (Childers &Wu1991).Amplitudeswere scaled to a constant presentation level using the rootmean squaredmethod in Praat (Boersma&Weenink 2012). A related-samples Wilcoxon signed-ranks test revealed that the magnitude ofthe unsigned mean of voice pitch change did not significantly differbetween raising and lowering from habitual pitch in women(Z¼ 0.365, N¼ 4, P¼ 0.715) but was marginally significant in men(Z¼ 1.826, N¼ 4, P¼ 0.068), suggesting that men raised their voicepitchmore than they lowered it. Change in duration of recording fromhabitual pitch did not differ between raised and lowered voices(related-samples Wilcoxon signed-ranks test: men: Z¼ 0.365, N¼ 4,P¼ 0.715; women: Z¼ 1.105, N¼ 4, P¼ 0.269). Since formant

ately altered voice pitch and vocal attractiveness, Animal Behaviour

0(2) (3)

5000

Freq

uen

cy (

Hz)

(1)

Time

Figure 1. Narrow-band spectrogram of the vowel ‘ah’ spoken by one female partici-pant in a (1) lowered pitch, (2) habitual pitch and (3) raised pitch. Solid arrow points tothe fundamental frequency; dashed arrow indicates a harmonic (ideally, an integermultiple of the fundamental frequency). In these particular voices, the fundamentalfrequency occurs at (1) 193.99 Hz, (2) 200.43 Hz and (3) 219.52 Hz.

P. J. Fraccaro et al. / Animal Behaviour xxx (2012) 1e10 3

frequency dispersion, a measure of vocal tract resonance and a corre-late of mammalian body size (Fitch 1997; Reby & McComb 2003),affects attractiveness judgments in humans (Collins & Missing 2003;Feinberg et al. 2005; Hodges-Simeon et al. 2010; Pisanski & Rendall2011), we tested for differences in formant dispersion between voi-ces that were raised or lowered (see Tables 1, 2). Habitual pitch in ourstudycloselymatches that reported inprevious studies (Childers&Wu1991; Feinberg et al. 2008b), suggesting that the voiceswe selected asstimuli were a representative sample of population habitual pitch.

We collected additional discrimination data from 68 women(mean � SE age: 21.98 � 0.977 years) and 33 men (mean � SE age:22.27 � 1.31 years) online. Previous studies have shown that col-lecting vocal attractiveness judgments online produces patterns ofresults that are identical to those collected in a laboratory setting(Feinberg et al. 2008a, b). Participants received the followinginstruction: ‘In this experiment you will be asked to choose whichvoice in a pair has the lower pitch. The voices might all sound verysimilar, but they are each subtly different’. Results indicated thatparticipants’ mean discrimination ratings were significantly abovechance for all voice pairs (all t > 4.442, all P < 0.001, allmeans > 70.3%), suggesting that participants could accuratelyidentify the voices with raised and lowered pitch.

Table 1Acoustic variables measured for each female voice recording, given as the difference betwvoice for each speaker

Women’s voices Raised-pitch minus habitually pitched recordings

Speaker A B C DF0 10.09 15.62 19.09 74.53F0�SD 2.82 12.15 17.15 20.36Jitter factor �0.320 �0.180 1.27 �0.26Shimmer factor �0.890 �0.950 �0.130 �0.50F1 10.66 111.38 �55.62 88.78F2 98.14 �172.84 �3.34 �17.52F3 28.86 169.34 28.86 �0.06F4 55.98 �36.28 16.70 �143.38Df 0.331 0.324 �0.141 0.15Pf (z score) 0.331 0.324 �0.141 0.15VTL (cm) �0.355 �0.245 0.287 �0.43HNR (dB) 0.860 0.880 3.00 1.49HNR�SD (dB) 1.05 0.610 0.030 0.44Duration (s) �0.7 �0.84 �0.43 0.96

F0: mean fundamental frequency (i.e. voice pitch); F0 � SD: pitch variability (i.e. variationDf: formant dispersion; Pf: formant position; HNR: harmonics-to-noise ratio; HNR � SD:length; Duration: duration of recordings by sex of voice and recording context. For each plowered-pitch voices were lower than habitually pitched voices. Jitter and shimmer factorin Hz unless otherwise specified.

Please cite this article in press as: Fraccaro, P. J., et al., Faking it: deliber(2012), http://dx.doi.org/10.1016/j.anbehav.2012.10.016

We measured the following variables using Praat acousticphonetics software (Boersma & Weenink 2012; see Tables 1, 2):voice pitch, pitch variability (i.e. the amount of variance in pitchwithin one speaker’s recording), formants 1 through 4, formantdispersion (Fitch 1997), formant position (Puts et al. 2012),apparent vocal tract length, harmonics-to-noise ratio (a measure ofvocal breathiness), jitter and shimmer. We examined multiplemeasures of jitter (a measure of short-term variation in funda-mental frequency) and shimmer (a measure of short-term variationin amplitude) using Praat software (Titze 1994; Feinberg 2004;Hughes et al. 2008; Boersma & Weenink 2012). Since jitter andshimmer are negatively correlated with perceived health andpositively correlated with perceived age (Feinberg 2004), suchattributions may have influenced participant ratings of vocalattractiveness. In addition, one measure of shimmer (11-pointamplitude perturbation quotient) is positively associated withbody symmetry in men (Hughes et al. 2008), so differences inshimmer may cue listeners to a speaker’s mate value (e.g. Møller &Pomiankowski 1993; Little et al. 2007). We note here that whileshimmer measures taken from stand-mounted microphonerecordings are similar to those from head-mounted microphonerecordings, speakers’ head motion can unduly influence shimmermeasurements (Winholtz & Titze 1997). We measured jitter usingfive methods: local, local absolute, relative average perturbation,five-point period perturbation quotient and difference of differ-ences of periods. We then measured shimmer using six methods:local, local dB, 3-, 5- and 11-point amplitude perturbation quotient,and difference of differences of periods. We performed factoranalyses on jitter and shimmer measurements separately, resultingin one jitter factor (eigenvalue: 4.009; KaisereMeyereOlkinsampling adequacy: 0.795; Bartlett’s test of sphericity:c210 ¼ 289:157, P < 0.001) and one shimmer factor (eigenvalue:

5.413; KaisereMeyereOlkin sampling adequacy: 0.798; Bartlett’stest of sphericity: c2

15 ¼ 490:364, P < 0.001). Since vocal prosodymay have differed between recording types, we investigated thestandard deviation of voice pitch within each recording. Thismeasure captures within-subject pitch variation and may beassociated with ratings of vocal attractiveness (see e.g. Hodges-Simeon et al. 2010). In addition, because harmonics-to-noise ratiois negatively associated with perceptions of vocal attractiveness(Bruckert et al. 2010), we investigated potential differences withinstimuli pairs that may have affected listeners’ judgments (seeTables 1, 2).

een each habitually pitched voice and the corresponding raised- and lowered-pitch

Lowered-pitch minus habitually pitched recordings

A B C D�56.93 �16.62 �6.44 �21.7727.22 �6.23 �1.34 �1.27

0 �0.650 �0.190 1.8 0.820�1.32 �0.540 �0.470 �0.240

�15.18 �59.32 �42.42 �3.46�27.30 �81.20 �41.88 �221.84

0 �153.06 �83.04 �34.62 58.6619.54 �104.08 �69.14 �556.70

4 �0.355 �0.611 �0.357 �0.8514 �0.355 �0.611 �0.357 �0.8516 0.348 0.680 0.445 0.940

0.430 1.68 0.990 2.750 0.920 �0.480 0.930 0.650

�0.59 0.06 �0.57 1.86

in fundamental frequency) within recordings; F1eF4: frequencies of formants 1e4;within-recording variability of harmonics-to-noise ratio; VTL: estimated vocal tractarticipant, all raised-pitch voices were higher than habitually pitched voices, and alls were derived from five and six different methods, respectively. All other values are

ately altered voice pitch and vocal attractiveness, Animal Behaviour

Table 2Acoustic variables measured for each male voice recording, given as the difference between each habitually pitched voice and the corresponding raised- and lowered-pitchvoice for each speaker

Men’s voices Raised-pitch minus habitually pitched recordings Lowered-pitch minus habitually pitched recordings

Speaker A B C D A B C DF0 20.3 53.61 40.42 34.39 �13.87 �32.28 �24.92 �14.34F0�SD 4.55 14.1 19.46 6.30 2.24 �14.56 �6.40 �2.36Jitter factor �0.53 0.56 0.21 �1.26 1.09 0.39 �0.020 2.50Shimmer factor �0.43 �0.23 0.41 �1.08 0.58 0.040 0.43 1.49F1 43.88 33.64 55.24 14.22 3.66 �28.34 �45.62 �38.18F2 10.62 �107.02 154.38 41.86 �21.26 �190.34 �85.24 �76.86F3 �77.10 66.82 169.56 123.12 �23.04 �67.04 10.24 �93.74F4 �42.36 166.9 94.26 42.54 15.74 �137.34 �40.96 72.96Df 0.00623 0.180 0.880 0.385 �0.0539 �0.715 �0.350 �0.383Pf (z score) 0.00624 0.180 0.880 0.385 �0.0539 �0.715 �0.350 �0.383VTL (cm) �0.198 �0.325 �1.17 �0.425 0.0319 1.035 0.738 0.582HNR (dB) 0.330 0.770 �1.12 2.55 �2.87 �0.480 �0.120 �4.43HNR�SD (dB) �0.78 1.5 �0.21 �0.030 �1.21 �0.99 0.080 �0.86Duration (s) �0.40 �0.47 �0.96 �0.040 �0.35 0.030 �0.82 �0.66

Description of variables and abbreviations as in Table 1.

P. J. Fraccaro et al. / Animal Behaviour xxx (2012) 1e104

ProcedureWe recruited 104 women (mean � SE age: 18.52 � 0.127 years)

and 110 men (mean � SE age ¼ 18.85 � 0.160 years) from McMas-ter University. Each participant received course credit, a cashreward of $10 Canadian/h, or a combination of course credit andcash for their participation. Each participant completed the two-alternative forced-choice voice preference task. The 24 voicerecordings were grouped into pairs, forming two pairs per indi-vidual (habitual versus raised, habitual versus lowered). In this task,participants listened to 16 pairs of voice recordings (8 male, 8female), consisting of a habitually pitched recording versus eithera raised- or lowered-pitch recording. Pairs of voices consisted ofrecordings from the same speaker. Participants were told: ‘You willhear two women’s (or two men’s) voices. Rate which is moreattractive to you’. The order of presentation of voice pairs and theside of display were fully randomized. Participants could replay thevoices ad libitum.

Male raters

Female raters

1

*

*

*

*

*

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0Lowered

pitchRaisedpitch

Raisedpitch

Male voicesFemale voices

0.9

Prop

orti

on o

f h

abit

ual

−pit

ch r

ecor

din

gs c

hos

en

Loweredpitch

Figure 2. Proportion of habitually pitched voice recordings chosen by male and femaleraters. Proportions of choices significantly different from chance (i.e. 0.5) are markedwith an asterisk. *P < 0.001. Error bars show �1 SE.

Results

We used one-sample t tests to test the relationship between thedirection of voice pitch modulation and preferences for habituallypitched voice recordings (i.e. the proportion of habitually pitchedvoices chosen in each pairing type; see Fig. 2). Women preferredhabitually pitched women’s recordings to both lowered-pitchrecordings (t103 ¼ 9.19, P < 0.001) and raised-pitch recordings(t103 ¼ 5.62, P < 0.001). Female raters also preferred habituallypitched men’s recordings to raised-pitch recordings (t103 ¼ 16.68,P < 0.001), but not when paired with lowered-pitch recordings(t103 ¼ 0.502, P ¼ 0.617).

Male raters preferred habitually pitched women’s recordings tolowered-pitch recordings (t109 ¼ 13.75, P < 0.001) but not to raised-pitch recordings (t109 ¼ 1.11, P ¼ 0.271). Male raters also preferredhabitually pitched men’s recordings to raised-pitch recordings(t109 ¼ 10.22, P < 0.001), but not when habitually pitched men’srecordings were paired with lowered-pitch recordings (t109 ¼ 1.08,P ¼ 0.285).

To examine potential sex differences more formally, we labelledhigh pitch in female voices and low pitch in male voices as ‘sextypical’. We labelled low pitch in female voices and high pitch inmale voices as ‘sex atypical’. To analyse our data relevant to sextypicality (as opposed to raised versus lowered pitch), we con-ducted a repeated measures ANOVA with two within-subjectsfactors, each with two levels: direction of voice pitch modulation(habitual versus sex typical/habitual versus sex atypical) and sex of

Please cite this article in press as: Fraccaro, P. J., et al., Faking it: deliber(2012), http://dx.doi.org/10.1016/j.anbehav.2012.10.016

voice (male/female). Sex of rater was the between-subjects factor.We observed a main effect of the direction of the voice pitchmodulation on preferences (F1,212 ¼ 162.12, P < 0.001) whereparticipants had stronger preferences for habitual pitch whenpaired with sex-atypical voices (i.e. low-pitched women’s voicesand high-pitched men’s voices) than when habitual pitch waspaired with sex-typical voices (i.e. high-pitched women’s voicesand low-pitched men’s voices). This effect occurred in both sexes(women: F1,103 ¼ 83.46, P < 0.001; men: F1,109 ¼ 80.55, P < 0.001).

The repeated measures ANOVA uncovered a significant inter-action between direction of the voice pitch modulation and sex ofvoice (F1,212 ¼ 8.81, P < 0.004), such that the aforementioned maineffect was stronger among men’s voices than among women’svoices. Additionally, we observed a significant interaction amongdirection of the voice pitch modulation, sex of voice and sex of rater

ately altered voice pitch and vocal attractiveness, Animal Behaviour

P. J. Fraccaro et al. / Animal Behaviour xxx (2012) 1e10 5

(F1,212 ¼ 13.31, P < 0.001). When split by sex of rater, the ANOVAshowed a significant interaction between direction of voice pitchmodulation and sex of voice for female raters (F1,103 ¼ 24.35,P < 0.001) but not male raters (F1,109 ¼ 0.213, P ¼ 0.645). Thedifferences between preferences for habitual voice pitch versusmodulated pitch were strongest whenwomen listened to opposite-sex voices. There were no other significant effects or interactions(all F < 1.22, all P > 0.270) in the model.

Our analyses of attractiveness ratings by stimulus pairing wereconducted using participant raters as the unit of analysis. Someresearchers have proposed that using each stimulus as the unit ofanalysis may be more appropriate than using each participant’sratings (e.g. Kroodsma et al. 2001; but see Coss 2009; Schank &Koehlne 2009). Accordingly, we used Wilcoxon signed-ranks testswith a test value of 0.5 (i.e. chance) to determine whether a similarpattern of results emerged to our parametric one-sample t tests,which used individual ratings as the unit of analysis. Judgments ofsex-atypical voice pitch alterations were marginally significantfrom chance, suggesting a similar pattern of results to our t tests.Both men and women chose marginally significantly more habit-ually pitched voices than chance when paired with loweredwomen’s voices or raised men’s voices (all means > 0.5568, allZ ¼ 1.826, all P ¼ 0.068). The P values observed for these testsprobably reflect the relative deficiency in statistical power (N ¼ 4for each comparison) rather than a true inability to reject the nullhypothesis. Conversely, both opposite- and same-sex ratings of sex-typical pitch alterations were highly nonsignificant (allmeans < 0.493, all Z < 0.921, all P > 0.357). These results replicatethose of our by-rater analyses and suggest that those results areunlikely to be produced solely by the chosen unit of analysis (seealso Jones et al. 2010). Additionally, these by-item analyses suggestthat within-category differences in stimulus type are unlikely tohave driven the results of our by-rater analyses, and that our resultsare not likely to have been influenced by pseudoreplication.

To test whether the degree to which individuals raised or low-ered their voices affected responses, we correlated the proportionof habitually pitched voices chosen for each stimulus pair with thefundamental frequency difference (in Hz) between raised-pitchvoices and habitually pitched voices, and between lowered-pitchvoices and habitually pitched voices. Two of our variables werenon-normally distributed (preferences for raised-pitch men’s voi-ces: ShapiroeWilk test: W ¼ 0.727, P ¼ 0.023; degree to whichwomen raised their voices: W ¼ 0.744, P ¼ 0.034); therefore, weused nonparametric tests for these variables. The proportion ofvoice recordings chosen as more attractive correlated with thedegree to which men raised (Spearman rank-order correlation:rS > 0.9, N ¼ 4, P < 0.01), but not lowered (Pearson correlation:r2 ¼ �0.007, P ¼ 0.993), their voices. The degree to which womenraised or lowered their voice was not associated with the propor-tion of voice recordings chosen as more attractive (raised:Spearman rank-order correlation: rS < 0.01, N ¼ 4, P > 0.999; low-ered: Pearson correlation: r2 ¼ �0.425, P ¼ 0.575). These resultssuggest that it is unlikely that the difference in preferences forraised and lowered voices in each sexwas due to the degree of vocalmodulation between stimuli. In addition, these results support thefinding that the magnitude of the unsigned mean of voice pitchchange did not significantly differ between raising and loweringfrom habitual pitch.

Since differences in pitch variability were found betweenrecording types, we conducted correlational analyses to determinewhether the difference in pitch variability between voice pairsaffected attractiveness ratings. As the difference in pitch variabilitybetween lowered-pitch voices and habitually pitched voices wasnon-normally distributed (ShapiroeWilk test: W ¼ 0.757,P ¼ 0.045), we used nonparametric tests for this variable. The

Please cite this article in press as: Fraccaro, P. J., et al., Faking it: deliber(2012), http://dx.doi.org/10.1016/j.anbehav.2012.10.016

degree of pitch variability when men raised their voice pitch fromhabitual pitch correlated positively with female perceptions ofattractiveness (Spearman rank-order correlations: rS > 0.999,P < 0.001). This suggests that as men raised their voice pitch, voicepitch variability increased, and voices with higher voice pitchvariability tended to be chosen as more attractive than voices withlower voice pitch variability. However, all other associationsbetween attractiveness ratings and pitch variability were notsignificant (Spearman rank-order correlations: rS < 0.800, allP > 0.20). This suggests that voice pitch variability is unlikely to bean acoustical parameter that mitigates the effect of sex-typicalvoice pitch alterations on attractiveness judgments here.

Consistent with human vocal physiology (Curry 1937; Hondaet al. 1999), changes in formant dispersion, formant position andestimated vocal tract length were concomitant with the direction ofvoice pitch modulation. Since lowered formant frequencies can beassociated with increased vocal attractiveness among men (e.g.Feinberg et al. 2005) and decreased vocal attractiveness amongwomen (e.g. Puts et al. 2011), the pattern of results observed forattractiveness judgments cannot be explained by apparent elon-gation or contraction of speakers’ vocal tracts because male voiceswith lowered formant frequencies and female voices with raisedformant frequencies were not preferred to controls.

To test whether perceivers could detect the naturalness or vocalcomfort of the recordings, we collected opposite-sex ratings of theapparent naturalness/comfort of speakers. Forty-one women(mean � SE age: 21.11 � 0.401 years) and 25 men (mean � SE age:24.82 � 2.14 years) judged the voices online. Participants wereinstructed as follows: ‘In this experiment you will hear pairs ofwomen’s voices and will be asked to choose which person isspeaking most comfortably/naturally. The voices might all soundvery similar, but they are each subtly different’. The results of ourone-sample t tests (test value: 0.5) showed that women did notchoose the habitually pitched men’s voices as more comfortable/natural sounding more often than chance when paired withlowered-pitch voices (t40 ¼ 0.696, P < 0.491). Women chose men’shabitually pitched voices as sounding more comfortable/naturalthan chance when paired with raised-pitch voices (t40 ¼ 3.130,P ¼ 0.003). Men chose women’s lowered-pitch voices as soundingmore natural/comfortable significantly more than chance whenpaired with habitually pitched women’s voices (t24 ¼ �6.532,P < 0.001). Men chose habitually pitched voices as sounding morenatural/comfortable significantly more than chance when pairedwith raised-pitch women’s voices (t24 ¼ 2.107, P ¼ 0.046). Theseresults suggest that perceptions of vocal comfort or naturalnesscannot explain the pattern of results observed for judgments ofattractiveness, since men’s perceptions of vocal comfort/natural-ness did not correspond with their perceptions of attractiveness.Although men chose women’s lowered-pitch voices as soundingmore natural/comfortable significantly more than chance, they alsochose these voices as significantly less attractive than chance whenpaired with habitually pitched women’s voices. Similarly, womendid not judge men’s habitually pitched voices as sounding moreattractive than men’s lowered-pitch voices, suggesting thatperceptions of comfort/naturalness did not mitigate the gains inattractiveness derived from having an exaggerated sex-typicalvoice pitch during volitional voice pitch alterations.

EXPERIMENT 2

To test whether the effects found in experiment 1 were specificto judgments of vocal attractiveness, we collected dominanceratings using the same experimental design and stimuli as exper-iment 1. Both men and women typically perceive voices of bothsexes with low voice pitch and/or low formant dispersion as

ately altered voice pitch and vocal attractiveness, Animal Behaviour

P. J. Fraccaro et al. / Animal Behaviour xxx (2012) 1e106

sounding more dominant (Puts et al. 2006, 2007; Jones et al. 2010;Watkins et al. 2010; reviewed in Puts et al. 2012). Indeed, manyspecies, including humans, lower their vocal frequencies as puta-tive displays of dominance (Wagner 1989a, b, 1992; Bee & Perrill1996; Given 1999; Bee et al. 1999, 2000; Bee & Bowling 2002;Lardner & bin Lakim 2002; Owen & Gordon 2005; Reby et al. 2005;Frey et al. 2008). This lowering of vocal frequencies, specificallyformant frequencies, functions to exaggerate the size of the callerby elongating the apparent vocal tract of the speaker (Fitch 1999),a correlate of body size (Fitch 1997; Reby & McComb 2003). Whilesuch vocal displays can be energetically costly to callers (Gillooly &Ophir 2010; sensu Chappell et al. 1995; Kilner 2001), these costsmay be negated by the benefits of deterring rivals from threateningoffspring and from pilfering scarce resources, territories and mates.Although voice pitch is generally not a within-sex correlate of bodysize in adult humans (Majewski et al. 1972; Lass & Brown 1978;Künzel 1989; Collins 2000; Rendall et al. 2005; Bruckert et al. 2006;Sell et al. 2010; but see Graddol & Swann 1983; Evans et al. 2006;Puts et al. 2012), women perceive men with lower voice pitch aslarger than those with higher voice pitch (van Dommelen 1993;Collins 2000; González 2003; Feinberg et al. 2005; Smith &Patterson 2005; Smith et al. 2005). Given that humans are sensi-tive to their own individual risks when assessing the dominance ofpotential rivals (Watkins et al. 2010), we expected humans to avoidincurring the potential costs of misjudging vocal dominance byrating the lower of the two voices in each pair as sounding moredominant.

Methods

ProcedureWe tested this hypothesis by collecting dominance ratings of the

stimuli used in experiment 1. As in experiment 1, participantslistened to the same 16 pairs of voice recordings (8 male, 8 female),consisting of a habitually pitched recording versus either a raised-or lowered-pitch recording. Pairs of voices consisted of recordingsfrom the same speaker. Participants were instructed as follows: ‘Inthis experiment you will be asked to choose which voice in a pairsoundsmore dominant. The voices might all sound very similar, butthey are each subtly different’. The order of presentation of voicepairs and the side of display were fully randomized. Participantscould replay the voices ad libitum. We collected ratings by 51women (mean � SE age: 21.01 � 0.378 years) and 24 men (mean -� SE age: 22.75 � 0.657 years) online.

Results

We used one-sample t tests to test the relationship between thedirection of voice pitch modulation and dominance ratings (i.e. theproportion of the lowest-pitched voices chosen for each pairingtype). Women chose the lower-pitched voice of the two women’svoices as more dominant significantly more than chance (i.e. 0.5) inhigh versus habitual (t50 ¼ 2.44, P ¼ 0.019) and low versus habitual(t50 ¼ 2.44, P ¼ 0.019) voice pairings. Similarly, women chose thelower-pitched voice of the two men’s voices as more dominantsignificantly more than chance in high versus habitual (t50 ¼ 10.32,P < 0.001) and low versus habitual (t50 ¼ 6.70, P < 0.001) voicepairings. Men chose the lower-pitched voice of the two men’svoices as more dominant significantly more than chance in bothhigh versus habitual (t23 ¼ 4.05, P ¼ 0.001) and low versus habitual(t23 ¼ 3.42, P ¼ 0.002) voice pairings. When assessing women’svoices, men chose the lower-pitched voice of the two voices asmore dominant significantly more than chance in high versushabitual voice pairings (t23 ¼ 3.16, P ¼ 0.004), but not in low versushabitual voice pairings (t23 ¼ �0.161, P ¼ 0.873).

Please cite this article in press as: Fraccaro, P. J., et al., Faking it: deliber(2012), http://dx.doi.org/10.1016/j.anbehav.2012.10.016

To examine potential sex differences in dominance ratings moreformally, we conducted a repeated measures ANOVA with twowithin-subjects factors, each with two levels: direction of modu-lation (habitual versus high/habitual versus low) and sex of voice(male/female). Sex of rater was the between-subjects factor. OurANOVA reported a marginally significant effect of our between-subjects factor of sex of rater on dominance ratings (F1,73 ¼ 3.98,P ¼ 0.050), so we continued by performing separate ANOVAs formale and female raters. For female raters, we observed a maineffect of direction of modulation on dominance ratings(F1,50 ¼ 14.40, P < 0.001), where participants had stronger domi-nance perceptions of the lower pitch voice in raised versus habitualvoice pitch pairings than in lowered versus habitual voice pitchpairings. Additionally, we observed a significant effect of sex ofvoice on dominance ratings (F1,50 ¼ 13.22, P ¼ 0.001) wherebyfemale raters had stronger dominance perceptions of the lowervoice when listening to men’s voices than when listening towomen’s voices. We also uncovered a significant interactionbetween sex of voice and direction of modulation on dominanceratings (F1,50 ¼ 5.10, P ¼ 0.028) whereby the aforementioned effectof direction of modulation was stronger in women’s voices than inmen’s voices.

For male raters, we observed a main effect of direction ofmodulation on dominance ratings (F1,23 ¼ 5.71, P ¼ 0.025) whereparticipants had stronger dominance perceptions of the lower pitchvoice in raised versus habitual voice pitch pairings than in loweredversus habitual voice pitch pairings. Additionally, we observeda significant effect of sex of voice on dominance ratings (F1,23 ¼ 5.77,P ¼ 0.025) whereby female raters had stronger dominanceperceptions of the lower voice when listening to men’s voices thanwhen listening to women’s voices. Unlike women’s dominanceratings, there was no significant interaction between sex of voiceand direction of modulation on men’s dominance ratings(F1,23 ¼ 0.671, P ¼ 0.421).

We conducted a by-stimulus analysis, performing one-sampleWilcoxon signed-ranks tests for each stimulus type with eachstimulus as the unit of analysis (rather than each listener). Menchose the lower of the two voices marginally significantly morethan chance (i.e. 0.5) when judging men’s raised versus habitual(Z ¼ 1.83, P ¼ 0.068) and low versus habitual (Z ¼ 1.84, P ¼ 0.068)voice pairs, but not when judging women’s raised versus habitual(Z ¼ 1.46, P ¼ 0.144) and lowered versus habitual (Z ¼ 1.84,P ¼ 0.465) voice pairs. Women chose the lower of the two voices asmore dominant marginally significantly more than chance (i.e. 0.5)when judging men’s raised versus habitual (Z ¼ 1.84, P ¼ 0.066)and low versus habitual (Z ¼ ¼ 1.84, P ¼ 0.066) voice pairs, but notwhen judging women’s raised versus habitual (Z ¼ 1.60, P ¼ 0.109)and lowered versus habitual (Z ¼ 1.46, P ¼ 0.144) voice pairs. As inexperiment 1, the P values observed here probably reflect alack of statistical power (N ¼ 4 for each analysis) rather thana true inability to reject the null hypothesis. These by-stimulusanalyses provide concordant evidence that men’s lowered-pitchvoices were perceived as more dominant than their habituallypitched voices. It is noteworthy that the lower voice was chosenas more dominant over 58.1% of the time for all voice pairs(mean� SE¼ 0.694� 0.034) except men’s assessments of women’slow versus habitual voice pairs (mean� SE¼ 0.522� 0.050).

DISCUSSION

Although other studies of vocal attractiveness have investigatedthe effect of interpersonal variation in pitch and experimentallymanipulated voice pitch on attractiveness ratings (reviewedin Feinberg 2008), this is the first study to investigate whetheran intentional voice pitch modulation necessarily affects

ately altered voice pitch and vocal attractiveness, Animal Behaviour

P. J. Fraccaro et al. / Animal Behaviour xxx (2012) 1e10 7

attractiveness ratings. In experiment 1, we found no evidence thatdeliberately altering voice pitch increases speakers’ vocal attrac-tiveness to the opposite sex, although we did find that deliberate,sex-atypical alteration of pitch can decrease speakers’ vocalattractiveness. Of course, absence of a positive effect of deliberatevoice pitch modulation on attractiveness ratings does not precludeits occurrence, but does demonstrate that this preference is notmandatory. In experiment 2, by contrast, we found that perceiversinterpreted lowered voice pitch from same-sex speakers assounding more dominant than a speaker’s habitual voice pitch.Taken together, these results constitute evidence that humans usea relatively low-risk strategy when assessing the mate value anddominance of potential mates and rivals. Specifically, humans tendnot to ascribe higher mate value to volitional sex-typical voice pitchalterations, which may function to aid perceivers in avoiding thecosts associated with making a poor mate choice decision.Conversely, humans tend to ascribe higher dominance to voiceswith lowered pitch than habitually pitched voices, which mayfunction to aid perceivers in avoiding the costs associated withmisjudging the dominance of potentially threatening individuals.However, taken together, our results show that deliberately alteringvoice pitch reliably alters dominance attributions, but does notreliably alter attractiveness judgements, suggesting that voice pitchis a more reliable (i.e. more difficult to fake) cue to others’ attrac-tiveness than it is to others’ dominance. Furthermore, the results ofexperiment 2 suggest that the results of experiment 1 were not dueto a general response bias to naturally modulated voice pitch andthe acoustical changes such alterations entail.

Interestingly, men showed no preference for women’s voiceswith higher pitch when compared to women’s voices with habitualpitch. By contrast, women preferred women’s voices with habitualpitch to the samewomenwith higher-pitched voices. Men typicallyrate both naturally higher-pitched women’s voices (Collins &Missing 2003; Feinberg et al. 2008b) and those electronicallyraised in pitch as relatively attractive (Feinberg et al. 2008b; Joneset al. 2008, 2010; Apicella & Feinberg 2009; Puts et al. 2011).Furthermore, men rated women’s voices that were electronicallyraised in pitch as being produced by women that were more likelyto cheat on their partners than were the same voices electronicallylowered in pitch (O’Connor et al. 2011). Women have also beenfound to rate more feminine-sounding women’s voices (i.e.increased voice pitch and formant dispersion) as more flirtatiousthan relatively masculine-sounding voices (i.e. decreased voicepitch and formant dispersion) (Puts et al. 2011). Thus, it is possiblethat women may rate these deliberately raised-pitched voices asless attractive than habitually pitched voices as a result of theseassociations.

Opposite-sex preferences indicated that although habituallypitched voices were preferred for women’s voices, women showedno significant preference for either habitually pitched voices orlowered-pitch voices. This sex difference in preferences for sex-typical modulations may indicate that lowered-pitch voices maybe as informative as habitually pitched voices in communicatingthe mate quality of a speaker. For example, red deer stags lowertheir laryngeal position during a roar, thus functionally increasingtheir supralaryngeal vocal tract length (the distance between thelarynx and the lips) and producing an acoustic signal with loweredformant dispersion (Fitch & Reby 2001; Reby & McComb 2003).Some researchers (Fitch & Reby 2001; Reby & McComb 2003) havehypothesized that since the sternum imposes an anatomical limiton the extent to which the larynx can be lowered in red deer, themaximally lowered position may still produce an honest acousticsignal of the sender. Indeed, Reby & McComb (2003) found thatminimum formant frequencies produced at the maximum oflaryngeal retractionwere negatively associatedwith bodymass and

Please cite this article in press as: Fraccaro, P. J., et al., Faking it: deliber(2012), http://dx.doi.org/10.1016/j.anbehav.2012.10.016

reproductive success in red deer stags. A similar physiological oranatomical constraint (e.g. vocal fry; the length and thickness of thevocal folds) on the extent to which men can lower their funda-mental frequency may impose honesty, whereas the ability ofwomen to raise their fundamental frequency may be relativelyunconstrained. Future research should address whether sex-typicalmodulations are constrained equally in both sexes.

Some studies have found that humans modulate voice pitchdepending on the attractiveness of the listener (Hughes et al. 2010;Fraccaro et al. 2011), and that certain vocal strategies can securefurther encounters with a desired partner (Anolli & Ciceri 2002).Here we provide some evidence that these modulations may needto be elicited in a particular social context in order to increasea speaker’s attractiveness. Findings from correlational andcomputer manipulation studies indicate reliable human prefer-ences for higher-pitched women’s voices and lower-pitched men’svoices (reviewed in Feinberg 2008). Rather than contradictingthese findings, our results indicate that listeners may perceivedeliberately altered voice pitch differently from differences inhabitual pitch between speakers or from computer-based voicepitch alterations that leave other acoustic characteristics unper-turbed. Furthermore, although voice pitch modulations occur inresponse to social context andmay affect the perceiver inways thatare adaptive (e.g. Anolli & Ciceri 2002), our results suggest thatdeliberate alteration of pitch may not be an adaptive vocal strategywhen attempting to increase perceptions of one’s vocalattractiveness.

The mean pitch difference between stimuli in our study(29.75 Hz) waswell over the just noticeable difference for change invoice pitch reported by Stevens (2000; w3 Hz), Pisanski & Rendall(2011; 6%) and Re et al. (2012; 4.1%). Moreover themagnitude of themean pitch difference in our study is comparable to that used instudies of vocal attractiveness where recordings of voices wereexperimentally manipulated (e.g. Feinberg et al. 2005; Vukovicet al. 2008; Fraccaro et al. 2010; Jones et al. 2010), and abouttwice that of the manipulation used in some studies (e.g. onesemitone; Puts 2005, 2006). Consequently, the null results reportedhere are unlikely to be the result of an inability to discriminatebetween our stimuli. Indeed, the discrimination data from experi-ment 1 confirmed that pitch alterations could be detected. As therewas no significant relationship between attractiveness ratings andthe unsigned mean of voice pitch change when speakers raised orlowered their voice pitch, it is unlikely that differences in saliencecaused the observed sex differences in preferences in experiment 1.Furthermore, the raised- and lowered-pitch voices here werewithin the normal human speaking range (Childers & Wu 1991).Thus, the preferences for habitually pitched voices observed herewere not due to abnormally pitched alternatives (Re et al. 2012).

In our acoustical analysis of stimuli, we found that changes informant dispersion, formant position and estimated vocal tractlength differed significantly between habitually pitched recordingsand recordings with raised or lowered pitch. These differenceswere concomitant with the direction of voice pitch modulation.This is consistent with the current understanding of human phys-iological control of voice pitch, since lowering or raising voice pitchresults in the lowering or raising of laryngeal position, respectively(Curry 1937; Honda et al. 1999). It is well known that laryngealdescent produces lowered formant dispersion, and thus longerestimated vocal tract length (see e.g. Frey et al. 2011). It is similarlyconsistent that the standardized average formant position alsotracked voice pitch modulations. Since lowered formant frequen-cies can be associated with increased vocal attractiveness amongmen (e.g. Feinberg et al. 2005) and decreased vocal attractivenessamong women (e.g. Puts et al. 2011), our results cannot beexplained by apparent elongation or contraction of speakers’ vocal

ately altered voice pitch and vocal attractiveness, Animal Behaviour

P. J. Fraccaro et al. / Animal Behaviour xxx (2012) 1e108

tracts because male voices with lowered formant frequencies andfemale voices with raised formant frequencies were not preferredto controls.

Fairly consistent changes in voice pitch variability (i.e. standarddeviation of fundamental frequency across each vowel sound,averaged across vowels for each speaker) between voice pair typesled us to test whether this variable affected judgments of attrac-tiveness. We found a negative association between the observeddecrease in pitch variability in voice pairs containing lowered-pitchvoices and attractiveness ratings, suggesting that differences inpitch variability similarly did not indicate vocal dishonesty tolisteners. Although men tended to lower their harmonics-to-noiseratio when lowering their voice pitch, this probably had no nega-tive effect on ratings since decreased harmonics-to-noise ratio ispositively associated with vocal attractiveness in morphed voices(Bruckert et al. 2010). Furthermore, although women tended toincrease voice pitch variability when raising their voice pitch(compared with lowering their voice pitch), and increased voicepitch variability is associated with deception in the context of lying(Anolli & Ciceri 1997), we found no association between the degreeof voice pitch variability and attractiveness ratings for raised-pitchversus habitually pitched women’s voices. This suggests that vari-ability in voice pitch cannot explain the pattern of results found forraised-pitch versus habitually pitched women’s voices. We foundno other consistent vocal changes that could alert listeners to vocaldishonesty and produce the pattern of results found here. Identi-fying the potential acoustical differences in vocal modulations eli-cited in mate choice and in cooperative or agonistic interactions, aswell as those in deliberate and/or dishonest vocal modulations is aninteresting avenue for future research in animal communication.

In experiment 2, lower-pitched voices were perceived as moredominant by both sexes, except when men judged women’slowered-pitch versus habitually pitched voices. By rating lowered-pitch voices as sounding more dominant, listeners may bedemonstrating sensitivity to risk in assessing the dominance ofothers (see e.g. Watkins et al. 2010). As voice pitch can carryinformation about the emotional state of a speaker (Fairbanks &Pronovost 1939; Williams & Stevens 1972; Razak et al. 2003), theresults of experiment 2 may suggest that listeners are inclined tointerpret voice pitches associated with dominance as genuine toavoid conflict or harm. Thus, the functional honesty of vocal signalsenforced by perceivers may only arise in domains where perceiverswould benefit from such honesty.

In summary, this is the first study to demonstrate that deliberatemanipulations of voice pitch do not necessarily enhance vocalattractiveness. This study is important because it demonstrates thatpreferences for pitch are not general response biases to stimuli thatare higher or lower in pitch. Indeed, in order for pitch manipula-tions to alter attractiveness ratings, they may need to be producedby computermanipulations (reviewed in Feinberg 2008), or elicitedin a realistic context (e.g. Hughes et al. 2010). As humans are able todeliberately modify their voice pitch (Titze 1994), it may not alwaysbe adaptive for listeners to favour deliberately altered voice pitch.We found no evidence that deliberate, sex-typically altered pitch issufficient to increase vocal attractiveness, but sex-atypical alter-ations may, in fact, decrease attractiveness.

Acknowledgments

This research was supported by grants from the Social Scienceand Humanities Research Council of Canada, Canada Foundation forInnovation (Leaders Opportunity Fund) and an Early Researcheraward from the OntarioMinistry of Research and Innovation to D. R.Feinberg. P. J. Fraccaro, J. J. M. O’Connor are supported by grantsfrom the Social Science and Humanities Research Council of

Please cite this article in press as: Fraccaro, P. J., et al., Faking it: deliber(2012), http://dx.doi.org/10.1016/j.anbehav.2012.10.016

Canada. We thank Cara Tigue and Diana Borak for assistance incollecting judgments of voice stimuli.

References

Abitbol, J., Abitbol, P. & Abitbol, B.1999. Sex hormones and the female voice. Journalof Voice, 13, 424e446, http://dx.doi.org/10.1016/S0892-1997(99)80048-4.

Alonso, L. C. & Rosenfield, R. L. 2002. Oestrogens and puberty. Best Practice &Research Clinical Endocrinology & Metabolism, 16, 13e30, http://dx.doi.org/10.1053/beem.2002.0177.

Anolli, L. & Ciceri, R.1997. The voice of deception: vocal strategies of naive and ableliars. Journal of Nonverbal Behavior, 21, 259e284, http://dx.doi.org/10.1023/A:1024916214403.

Anolli, L. & Ciceri, R. 2002. Analysis of the vocal profiles of male seduction: fromexhibition to self-disclosure. Journal of General Psychology, 129, 149e169, http://dx.doi.org/10.1080/00221300209603135.

Apicella, C. L. & Feinberg, D. R. 2009. Voice pitch alters mate-choice-relevantperception in hunter-gatherers. Proceedings of the Royal Society B, 276, 1077e1082, http://dx.doi.org/10.1098/rspb.2008.1542.

Bee, M. A. & Bowling, A. C. 2002. Socially mediated pitch alteration by territorialmale bullfrogs, Rana catesbeiana. Journal of Herpetology, 36, 140e143, http://dx.doi.org/10.1670/0022-1511(2002)036[0140:SMPABT]2.0.CO;2.

Bee, M. A. & Perrill, S. A. 1996. Responses to conspecific advertisement calls inthe green frog (Rana clamitans) and their role in maleemale communica-tion. Animal Behaviour, 133, 283e301, http://dx.doi.org/10.1163/156853996X00152.

Bee, M. A., Perrill, S. A. & Owen, P. C. 1999. Size assessment in simulated territorialencounters between male green frogs (Rana clamitans). Behavioral Ecology andSociobiology, 45, 177e184, http://dx.doi.org/10.1007/s002650050551.

Bee, M. A., Perrill, S. A. & Owen, P. C. 2000. Male green frogs lower the pitch ofacoustics signals in defense of territories: a possible dishonest signal of size?Behavioral Ecology, 11, 169e177, http://dx.doi.org/10.1093/beheco/11.2.169.

Boersma, P. & Weenink, D. 2012. Praat: Doing Phonetics by Computer, Version5.2.26. http://www.praat.org.

Borkowska, B. & Pawlowski, B. 2011. Female voice frequency in the context ofdominance and attractiveness perception. Animal Behaviour, 82, 55e59, http://dx.doi.org/10.1016/j.anbehav.2011.03.024.

Bruckert, L., Liénard, J.-S., Lacroix, A., Kreutzer, M. & Leboucher, G. 2006. Womenuse voice parameters to assess men’s characteristics. Proceedings of the RoyalSociety B, 273, 83e89, http://dx.doi.org/10.1098/rspb.2005.3265.

Bruckert, L., Bestelmeyer, P., Latinus, M., Rouger, J., Charest, I., Rousselet, G. A.,Kawahara, H. & Belin, P. 2010. Vocal attractiveness increases by averaging.Current Biology, 20, 116e120, http://dx.doi.org/10.1016/j.cub.2009.11.034.

Chappell, M. A., Zuk, M., Kwan, T. H. & Johnsen, T. S. 1995. Energy cost of an avianvocal display: crowing in red junglefowl. Animal Behaviour, 45, 255e257, http://dx.doi.org/10.1016/0003-3472(95)80177-4.

Charlton, B. D., Ellis, W. A. H., McKinnon, A. J., Cowin, G. J., Brumm, J., Nilsson, K.& Fitch, W. T. 2011. Cues to body size in the formant spacing of male koala(Phascolarctos cinereus) bellows: honesty in an exaggerated trait. Journal ofExperimental Biology, 214, 3414e3422, http://dx.doi.org/10.1242/jeb.061358.

Childers, D. & Wu, K. 1991. Gender recognition from speech. Part II: fine analysis.Journal of the Acoustical Society of America, 90, 1841e1857, http://dx.doi.org/10.1121/1.401664.

Collins, S. A. 2000. Men’s voices and women’s choices. Animal Behaviour, 60, 773e780, http://dx.doi.org/10.1006/anbe.2000.1523.

Collins, S. A. & Missing, C. 2003. Vocal and visual attractiveness are related inwomen. Animal Behaviour, 65, 997e1004, http://dx.doi.org/10.1006/anbe.2003.2123.

Coss, R. G. 2009. Pseudoreplication conventions are testable hypotheses. Journal ofComparative Psychology, 123, 444e446, http://dx.doi.org/10.1037/a0016093.

Curry, R.1937. The mechanism of pitch change in the voice. Journal of Physiology, 91,254e258.

Dabbs, J. M. & Mallinger, A. 1999. High testosterone levels predict low voice pitchamong men. Personality and Individual Differences, 27, 801e804, http://dx.doi.org/10.1016/S0191-8869(98)00272-4.

van Dommelen, W. A. 1993. Speaker height and weight identification: a re-evaluation of some old data. Journal of Phonetics, 21, 337e341.

Evans, S., Neave, N. & Wakelin, D. 2006. Relationship between vocal characteristicsand body size and shape in human males: an evolutionary explanation fora deep male voice. Biological Psychology, 72, 160e163, http://dx.doi.org/10.1016/j.biopsycho.2005.09.003.

Evans, S., Neave, N., Wakelin, D. & Hamilton, C. 2008. The relationship betweentestosterone and vocal frequencies in human males. Physiology & Behavior, 93,783e788, http://dx.doi.org/10.1016/j.physbeh.2007.11.033.

Fairbanks, G. & Pronovost, W. 1939. An experimental study of the pitch charac-teristics of the voice during the expression of emotions. Speech Monographs, 6,87e104, http://dx.doi.org/10.1080/03637753909374863.

Feinberg, D. R. 2004. Fundamental frequency perturbation indicates perceivedhealth and age in male and female speakers. Journal of the Acoustical Society ofAmerica, 115, 2609.

Feinberg, D. R. 2008. Are human faces and voices ornaments signaling commonunderlying cues to mate value? Evolutionary Anthropology, 17, 112e118, http://dx.doi.org/10.1002/evan.20166.

ately altered voice pitch and vocal attractiveness, Animal Behaviour

P. J. Fraccaro et al. / Animal Behaviour xxx (2012) 1e10 9

Feinberg, D. R., Jones, B. C., Little, A. C., Burt, D. M. & Perrett, D. I. 2005.Manipulations of fundamental and formant frequencies influence the attrac-tiveness of human male voices. Animal Behaviour, 69, 561e568, http://dx.doi.org/10.1016/j.anbehav.2004.06.012.

Feinberg, D. R., Jones, B. C., Law-Smith, M. J., Moore, F. R., DeBruine, L. M.,Cornwell, R. E., Hillier, S. G. & Perrett, D. I. 2006. Menstrual cycle, traitestrogen level, and masculinity preferences in the human voice. HormonesBehavior, 49, 215e222, http://dx.doi.org/10.1016/j.yhbeh.2005.07.004.

Feinberg, D. R., DeBruine, L. M., Jones, B. C. & Little, A. C. 2008a. Correlatedpreferences for men’s facial and vocal masculinity. Evolution and HumanBehavior, 29, 233e241, http://dx.doi.org/10.1016/j.evolhumbehav.2007.12.008.

Feinberg, D. R., DeBruine, L. M., Jones, B. C. & Perrett, D. I. 2008b. The role offemininity and averageness of voice pitch in aesthetic judgments of women’svoices. Perception, 37, 615e623, http://dx.doi.org/10.1068/p5514.

Fernald, A. & Kuhl, P. 1987. Acoustic determinants of infant preference for moth-erese speech. Infant Behavior and Development, 10, 279e293, http://dx.doi.org/10.1016/0163-6383(87)90017-8.

Fink, B., Neave, N. & Seydel, H. 2007. Male facial appearance signals physicalstrength to women. American Journal of Human Biology, 19, 82e87, http://dx.doi.org/10.1002/ajhb.20583.

Fitch, W. T. 1997. Vocal tract length and formant frequency dispersion correlatewith body size in rhesus macaques. Journal of the Acoustical Society of America,102, 1213e1222, http://dx.doi.org/10.1121/1.421048.

Fitch, W. T. 1999. Acoustic exaggeration of size in birds via tracheal elongation:comparative and theoretical analyses. Journal of Zoology, 248, 31e48, http://dx.doi.org/10.1017/S095283699900504X.

Fitch, W. T. & Reby, D. 2001. The descended larynx is not uniquely human.Proceedings of the Royal Society B, 268, 1669e1675, http://dx.doi.org/10.1098/rspb.2001.1704.

Fraccaro, P. J., Feinberg, D. R., DeBruine, L. M., Little, A. C., Watkins, C. D. &Jones, B. C. 2010. Correlated male preferences for femininity in female faces andvoices. Evolutionary Psychology, 8, 447e461.

Fraccaro, P. J., Jones, B. C., Vukovic, J., Smith, F. G., Watkins, C. D., Feinberg, D. R.,Little, A. C. & DeBruine, L. M. 2011. Experimental evidence that women speakin a higher voice pitch to men they find attractive. Journal of EvolutionaryPsychology, 9, 57e67, http://dx.doi.org/10.1556/JEP.9.2011.33.1.

Frey, R., Gebler, A., Olson, K. A., Odonkhuu, D., Fritsch, G., Batsaikhan, N. &Stuermer, I. W. 2008. Mobile larynx in Mongolian gazelle: retraction of thelarynx during rutting barks in male Mongolian gazelle (Procapra gutturosaPallas, 1777). Journal of Morphology, 269, 1223e1237, http://dx.doi.org/10.1002/jmor.10656.

Frey, R., Volodin, I., Volodina, E., Soldatova, N. V. & Juldaschev, E. T. 2011.Descended and mobile larynx, vocal tract elongation and rutting roars in malegoitered gazelles (Gazella subgutturosa). Journal of Anatomy, 218, 566e585,http://dx.doi.org/10.1111/j.1469-7580.2011.01361.x.

Gangestad, S. W., Merriman, L. A. & Thompson, M. E. 2010. Men’s oxidative stress,fluctuating asymmetry and physical attractiveness. Animal Behaviour, 80, 1005e1013, http://dx.doi.org/10.1016/j.anbehav.2010.09.003.

Gillooly, J. F. & Ophir, A. G. 2010. The energetic basis of acoustic communication.Proceedings of the Royal Society B, 277, 1325e1331, http://dx.doi.org/10.1098/rspb.2009.2134.

Given, M. F. 1999. Frequency alteration of the advertisement call in the carpenterfrog, Rana virgatipes. Herpetologica, 55, 304e317, http://dx.doi.org/10.2307/3893224.

González, J. 2003. Estimations of speaker’s weight and height from speech:a reanalysis of data frommultiple studies by Lass and colleagues. Perceptual andMotor Skills, 96, 297e304, http://dx.doi.org/10.2466/PMS.96.1.297-304.

Goodale, E. & Kotagama, S. W. 2006. Context-dependent vocal mimicry ina passerine bird. Proceedings of the Royal Society B, 273, 875e880, http://dx.doi.org/10.1098/rspb.2005.3392.

Graddol, D. & Swann, J. 1983. Speaking fundamental frequency: some physical andsocial correlates. Language and Speech, 26, 351e366, http://dx.doi.org/10.1177/002383098302600403.

Gregory, S. W. & Webster, S. 1996. A nonverbal signal in voices of interviewpartners effectively predicts communication accommodation and social statusperceptions. Journal of Personality and Social Psychology, 70, 1231e1240, http://dx.doi.org/10.1037/0022-3514.70.6.1231.

Harries, M. L. L., Walker, J. M., Williams, D. M., Hawkins, S. & Hughes, I. A. 1997.Changes in the male voice at puberty. Archives of Disease in Childhood, 77, 445e447, http://dx.doi.org/10.1136/adc.77.5.445.

Hodges-Simeon, C. R., Gaulin, S. J. C. & Puts, D. A. 2010. Different vocal parameterspredict perceptions of dominance and attractiveness. Human Nature, 21, 406e427, http://dx.doi.org/10.1007/s12110-010-9101-5.

Honda, K., Hirai, H., Masaki, S. & Shimada, Y. 1999. Role of vertical larynxmovement and cervical lordosis in F0 control. Language and Speech, 42, 401e411, http://dx.doi.org/10.1177/00238309990420040301.

Hughes, S. M., Dispenza, F. & Gallup, G. G. 2004. Ratings of voice attractivenesspredict sexual behavior and body configuration. Evolution and Human Behavior,25, 295e304, http://dx.doi.org/10.1016/j.evolhumbehav.2004.06.001.

Hughes, S. M., Pastizzo, M. J. & Gallup, G. G. 2008. The sound of symmetryrevisited: subjective and objective analyses of voice. Journal of NonverbalBehavior, 32, 93e108, http://dx.doi.org/10.1007/s10919-007-0042-6.

Hughes, S. M., Farley, S. D. & Rhodes, B. C. 2010. Vocal and physiological changes inresponse to the physical attractiveness of conversational partners. Journal ofNonverbal Behavior, 34, 1e13, http://dx.doi.org/10.1007/s10919-010-0087-9.

Please cite this article in press as: Fraccaro, P. J., et al., Faking it: deliber(2012), http://dx.doi.org/10.1016/j.anbehav.2012.10.016

Jasienska, G., Ziomkiewicz, A., Ellison, P. T., Lipson, S. F. & Thune, I. 2004. Largebreasts and narrow waists indicate high reproductive potential in women.Proceedings of the Royal Society B, 271, 1213e1217, http://dx.doi.org/10.1098/rspb.2004.2712.

Jones, B. C., Feinberg, D. R., DeBruine, L. M., Little, A. C. & Vukovic, J. 2008.Integrating cues of social interest and voice pitch in men’s preferences forwomen’s voices. Biology Letters, 4, 192e194, http://dx.doi.org/10.1098/rsbl.2007.0626.

Jones, B. C., Feinberg, D. R., DeBruine, L. M., Little, A. C. & Vukovic, J. 2010.A domain-specific opposite-sex bias in human preferences for manipulatedvoice pitch. Animal Behaviour, 79, 57e62, http://dx.doi.org/10.1016/j.anbehav.2009.10.003.

Kilner, R. M. 2001. A growth cost of begging in captive canary chicks. Proceedings ofthe National Academy of Sciences, U.S.A., 98, 11394e11398, http://dx.doi.org/10.1073/pnas.191221798.

Kroodsma, D. E., Bryers, B. E., Goodale, E., Johnson, S. & Liu, W. C. 2001. Pseu-doreplication in playback experiments, revisited a decade later. Animal Behav-iour, 61, 1029e1033, http://dx.doi.org/10.1006/anbe.2000.1676.

Künzel, H. J. 1989. How well does average fundamental frequency correlate withspeaker height and weight? Phonetica, 46, 117e125, http://dx.doi.org/10.1159/000261832.

Lardner, B. & bin Lakim, M. 2002. Tree-hole frogs exploit resonance effects. Nature,420, 475, http://dx.doi.org/10.1038/420475a.

Lass, N. J. & Brown, W. S. 1978. Correlational study of speakers’ heights, weights,body surface areas, and speaking fundamental frequency. Journal of theAcoustical Society of America, 63, 1218e1220, http://dx.doi.org/10.1121/1.381808.

Leavesley, A. J. & Magrath, R. D. 2005. Communicating about danger: urgencyalarm calling in a bird. Animal Behaviour, 70, 365e373, http://dx.doi.org/10.1016/j.anbehav.2004.10.017.

Little, A. C., Apicella, C. L. & Marlowe, F. W. 2007. Preferences for symmetry inhuman faces in two cultures: data from the UK and the Hadza, an isolatedgroup of hunter-gatherers. Proceedings of the Royal Society B, 274, 3113e3117,http://dx.doi.org/10.1098/rspb.2007.0895.

McElligott, A., Birrer, M. & Vannoni, E. 2006. Retraction of the mobile descendedlarynx during groaning enables fallow bucks (Dama dama) to lower theirformant frequencies. Journal of Zoology, 270, 340e345, http://dx.doi.org/10.1111/j.1469-7998.2006.00144.x.

Majewski, W., Hollien, H. & Zalewski, J. 1972. Speaking fundamental frequency ofPolish adult males. Phonetica, 25, 119e125, http://dx.doi.org/10.1159/000259375.

Møller, A. P. & Pomiankowski, A. 1993. Fluctuating asymmetry and sexual selec-tion. Genetica, 89, 267e279, http://dx.doi.org/10.1007/BF02424520.

Moore, F. R., Cornwell, R. E., Law Smith, M. J., Al Dujaili, E. A. S., Sharp, M. &Perrett, D. I. 2011. Evidence for the stress-linked immunocompetence handicaphypothesis in human male faces. Proceedings of the Royal Society B, 278, 774e780, http://dx.doi.org/10.1098/rspb.2010.1678.

O’Connor, J. J. M., Re, D. E. & Feinberg, D. R. 2011. Voice pitch influencesperceptions of sexual infidelity. Evolutionary Psychology, 9, 64e78.

O’Connor, J. J. M., Feinberg, D. R., Fraccaro, P. J., Borak, D. J., Tigue, C. C., Re, D. E.,Jones, B. C., Little, A. C. & Tiddeman, B. 2012. Female preferences for malevocal and facial masculinity in videos. Ethology, 114, 321e330, http://dx.doi.org/10.1111/j.1439-0310.2011.02013.x.

Ohala, J. J. 1983. Cross-language use of pitch: an ethological view. Phonetica, 40,1e18.

Ohala, J. J. 1984. An ethological perspective on common cross-language utilizationof F0 of voice. Phonetica, 41, 1e16.

Owen, P. C. & Gordon, N. M. 2005. The effect of perceived intruder proximity andresident body size on the aggressive responses of male green frogs, Rana cla-mitans (Anura: Ranidae). Behavioral Ecology and Sociobiology, 58, 446e455,http://dx.doi.org/10.1007/S00265-005-0961-3.

Penton-Voak, I. S. & Chen, J. Y. 2004. High salivary testosterone is linked tomasculine male facial appearance in humans. Evolution and Human Behavior, 25,229e241, http://dx.doi.org/10.1016/j.evolhumbehav.2004.04.003.

Perrett, D. I., Lee, K. J., Penton-Voak, I., Rowland, D., Yoshikawa, S., Burt, D. M.,Henzi, S. P., Castles, D. L. & Akamatsu, S.1998. Effects of sexual dimorphism onfacial attractiveness. Nature, 394, 884e887, http://dx.doi.org/10.1038/29772.

Pisanski, K. & Rendall, D. 2011. The prioritization of voice fundamental frequencyor formants in listeners’ assessments of speaker size, masculinity, and attrac-tiveness. Journal of the Acoustical Society of America, 129, 2201e2212, http://dx.doi.org/10.1121/1.3552866.

Puts, D. A. 2005. Mating context and menstrual phase affect women’s preferencesfor male voice pitch. Evolution and Human Behavior, 26, 388e397, http://dx.doi.org/10.1016/j.evolhumbehav.2005.03.001.

Puts, D. A., Gaulin, S. J. C. & Verdolini, K. 2006. Dominance and the evolution ofsexual dimorphism in human voice pitch. Evolution and Human Behavior, 27,283e296, http://dx.doi.org/10.1016/j.evolhumbehav.2005.11.003.

Puts, D. A., Hodges, C. R., Cárdenas, R. A. & Gaulin, S. J. C. 2007. Men’s voices asdominance signals: vocal fundamental and formant frequencies influencedominance attributions among men. Evolution and Human Behavior, 28, 340e344, http://dx.doi.org/10.1016/j.evolhumbehav.2007.05.002.

Puts, D. A., Barndt, J. L., Welling, L. L. M., Dawood, K. & Burriss, R. P. 2011.Intrasexual competition among women: vocal femininity affects perceptions ofattractiveness and flirtatiousness. Personality and Individual Differences, 50, 111e115, http://dx.doi.org/10.1016/j.paid.2010.09.011.

ately altered voice pitch and vocal attractiveness, Animal Behaviour

P. J. Fraccaro et al. / Animal Behaviour xxx (2012) 1e1010

Puts, D. A., Apicella, C. L. & Cárdenas, R. A. 2012. Masculine voices signal men’sthreat potential in forager and industrial societies. Proceedings of the RoyalSociety B, 279, 601e609.

Rantala, M. J., Moore, F. R., Skrinda, I., Krama, T., Kivleniece, I., Kecko, S. &Krams, I. 2012. Evidence for the stress-linked immunocompetence handicaphypothesis in humans. Nature Communications, 3, 694, http://dx.doi.org/10.1038/ncomms1696.

Razak, A. A., Abidin, M. I. Z. & Komiya, R. 2003. Emotion pitch variation analysis inMalay and English voice samples. IEEE Conference Proceedings, 1, 108e112,http://dx.doi.org/10.1109/APCC.2003.1274322.

Re, D. E., O’Connor, J. J. M., Bennett, P. J. & Feinberg, D. R. 2012. Preferences forvery low and very high voice pitch in humans. PLoS One, 7, e32719, http://dx.doi.org/10.1371/journal.pone.0032719.

Reby, D. & McComb, K. 2003. Anatomical constraints generate honesty: acousticcues to age and weight in the roars of red deer stags. Animal Behaviour, 65, 519e530, http://dx.doi.org/10.1006/anbe.2003.2078.

Reby, D., McComb, K., Cargnelutti, B., Darwin, C., Fitch, W. & Clutton-Brock, T.2005. Red deer stags use formants as assessment cues during intrasexualagonistic interactions. Proceedings of the Royal Society B, 272, 941e947, http://dx.doi.org/10.1098/rspb.2004.2954.

Rendall, D., Kollias, S., Ney, C. & Lloyd, P. 2005. Pitch (F0) and formant profiles ofhuman vowels and vowel-like baboon grunts: the role of vocalizer body sizeand voice-acoustic allometry. Journal of the Acoustical Society of America, 117,944e955, http://dx.doi.org/10.1121/1.1848011.

Rhodes, G., Chan, J., Zebrowitz, L. A. & Simmons, L. W. 2003. Does sexualdimorphism in human faces signal health? Proceedings of the Royal Society B,Supplement, 270, S93eS95, http://dx.doi.org/10.1098/rsbl.2003.0023.

Roney, J. R., Hanson, K. N., Durante, K. M. & Maestripieri, D. 2006. Reading men’sfaces: women’s mate attractiveness judgments track men’s testosterone andinterest in infants. Proceedings of the Royal Society B, 273, 2169e2175, http://dx.doi.org/10.1098/rspb.2006.3569.

Saxton, T. K., Caryl, P. G. & Roberts, S. C. 2006. Vocal and facial attractivenessjudgments of children, adolescents and adults: the ontogeny of mate choice.Ethology, 112, 1179e1185, http://dx.doi.org/10.1111/j.1439-0310.2006.01278.x.

Saxton, T. K., Burriss, R. P., Murray, A. K., Rowland, H. M. & Roberts, S. C. 2009.Face, body and speech cues independently predict judgments of attractiveness.Journal of Evolutionary Psychology, 7, 23e35, http://dx.doi.org/10.1556/JEP.7.2009.1.4.

Schank, J. C. & Koehlne, T. J. 2009. Pseudoreplication is a pseudoproblem. Journal ofComparative Psychology, 123, 421e433, http://dx.doi.org/10.1037/a0013579.

Sell, A., Bryant, G. A., Cosmides, L., Tooby, J., Sznycer, D., van Rueden, C.,Krauss, A. & Gurven, M. 2010. Adaptations in humans for assessing physicalstrength from the voice. Proceedings of the Royal Society B, 277, 3509e3518,http://dx.doi.org/10.1098/rspb.2010.0769.

Seyfarth, R. M. & Cheney, D. L. 2003. Meaning and emotion in animal vocalizations.Annals of the New York Academy of Sciences, 1000, 32e55, http://dx.doi.org/10.1196/annals.1280.004.

Seyfarth, R. M., Cheney, D. L. & Marler, P. 1980. Vervet monkey alarm calls:semantic communication in a free-ranging primate. Animal Behaviour, 28,1070e1094, http://dx.doi.org/10.1016/S0003-3472(80)80097-2.

Singh, D. 2002. Female mate value at a glance: relationship of waist-to-hip ratio tohealth, fecundity and attractiveness. Neuroendocrinology Letters, Supplement 4,23, 81e91.

Smith, D. R. R. & Patterson, R. D. 2005. The integration of glottal pulse rate andvocal-tract length in judgments of speaker size, sex, and age. Journal of theAcoustical Society of America, 118, 3177e3186, http://dx.doi.org/10.1121/1.2047107.

Smith, D. R. R., Patterson, R. D., Turner, R., Kawahara, H. & Irino, T. 2005. Theprocessing and perception of size information in speech sounds. Journal of theAcoustical Society of America, 117, 305e318, http://dx.doi.org/10.1121/1.1828637.

Snowdon, C. T., Cleveland, J. & French, J. A. 1983. Responses to context- andindividual-specific cues in cotton-top tamarin long calls. Animal Behaviour, 31,92e101, http://dx.doi.org/10.1016/S0003-3472(83)80177-8.

Please cite this article in press as: Fraccaro, P. J., et al., Faking it: deliber(2012), http://dx.doi.org/10.1016/j.anbehav.2012.10.016

Stevens, K. N. 2000. Acoustic Phonetics. Cambridge, Massachusetts: MIT Press.Thornhill, R. & Gangestad, S. W. 1999. Facial attractiveness. Trends in

Cognitive Sciences, 3, 452e460, http://dx.doi.org/10.1016/S1364-6613(99)01403-5.

Thornhill, R. & Gangestad, S. W. 2006. Facial sexual dimorphism, developmentalstability, and susceptibility to disease in men and women. Evolution and HumanBehavior, 27, 131e144, http://dx.doi.org/10.1016/j.evolhumbehav.2005.06.001.

Titze, I. R. 1994. Principles of Voice Production. Englewood Cliffs, New Jersey:Prentice Hall.

Tomasello, M. & Zuberbühler, K. 2002. Primate vocal and gestural communication.In: The Cognitive Animal: Empirical and Theoretical Perspectives on AnimalCognition (Ed. by M. Bekoff, C. Allen & G. M. Burghardt), pp. 293e299. Cam-bridge, Massachusetts: MIT Press.

Trainor, L. J. & Desjardins, R. N. 2002. Pitch characteristics of infant-directedspeech affect infants’ ability to discriminate vowels. Psychonomic Bulletin &Review, 9, 335e340, http://dx.doi.org/10.3758/BF03196290.

Verdonck, A., Gaethofs, M., Carels, C. & de Zegher, F. 1999. Effect of low-dosetestosterone treatment on craniofacial growth in boys with delayed puberty.European Journal of Orthodontics, 21, 137e143, http://dx.doi.org/10.1093/ejo/21.2.137.

Vukovic, J., Feinberg, D. R., Jones, B. C., DeBruine, L. M., Welling, L. L. M.,Little, A. C. & Smith, F. G. 2008. Self-rated attractiveness predicts individualdifferences in women’s preferences for masculine men’s voices. Personality andIndividual Differences, 45, 451e456, http://dx.doi.org/10.1016/j.paid.2008.05.013.

Vukovic, J., Feinberg, D. R., DeBruine, L. M., Smith, F. G. & Jones, B. C. 2010a.Women’s voice pitch is negatively correlated with health risk factors. Journal ofEvolutionary Psychology, 8, 217e225, http://dx.doi.org/10.1556/JEP.8.2010.3.2.

Vukovic, J., Jones, B. C., DeBruine, L. M., Feinberg, D. R., Smith, F. G., Little, A. C.,Welling, L. L. M. & Main, J. 2010b. Women’s own voice pitch predicts theirpreferences for masculinity in men’s voices. Behavioral Ecology, 21, 767e772,http://dx.doi.org/10.1093/beheco/arq051.

Vukovic, J., Jones, B. C., Feinberg, D. R., DeBruine, L. M., Smith, F. G.,Welling, L. L. M. & Little, A. C. 2011. Variation in perceptions of physicaldominance and trustworthiness predicts individual differences in the effect ofrelationship context on women’s preferences for masculine pitch in men’svoices. British Journal of Psychology, 102, 37e48, http://dx.doi.org/10.1348/000712610x498750.

Wagner, W. E. 1989a. Fighting, assessment, and frequency alteration in Blanchard’scricket frog. Behavioral Ecology and Sociobiology, 25, 429e436, http://dx.doi.org/10.1007/BF00300189.

Wagner, W. E. 1989b. Graded aggressive signals in Blanchard’s cricket frog: vocalresponses to opponent proximity and size. Animal Behaviour, 38, 1025e1038,http://dx.doi.org/10.1016/S0003-3472(89)80141-1.

Wagner, W. E. 1992. Deceptive or honest signalling of fighting ability: a test ofalternative hypotheses for the function of changes in call dominant frequencyby male cricket frogs. Animal Behaviour, 44, 449e462, http://dx.doi.org/10.1016/0003-3472(92)90055-E.

Watkins, C. D., Fraccaro, P. J., Smith, F. G., Vukovic, J., Feinberg, D. R.,DeBruine, L. M. & Jones, B. C. 2010. Taller men are less sensitive to cues ofdominance in other men. Behavioral Ecology, 21, 943e947, http://dx.doi.org/10.1093/beheco/arq091.

Williams, C. E. & Stevens, K. N. 1972. Emotions and speech: some acoustic corre-lates. Journal of the Acoustical Society of America, 52, 1238e1250, http://dx.doi.org/10.1121/1.1913238.

Winholtz, W. S. & Titze, I. R. 1997. Miniature head-mounted microphone forvoice perturbation analysis. Journal of Speech, Language & Hearing Research,40, 894e899.

Yin, S. 2002. A new perspective on barking in dogs (Canis familiaris). Journal ofComparative Psychology, 116, 189e193, http://dx.doi.org/10.1037/0735-7036.116.2.189.

Zuberbühler, K. 2000. Referential labelling in Diana monkeys. Animal Behaviour, 59,917e927, http://dx.doi.org/10.1006/anbe.1999.1317.

ately altered voice pitch and vocal attractiveness, Animal Behaviour