Original ArticleVocal modulation during courtship increases proceptivity even in naive listeners
Introduction
The human voice is remarkably variable. Aside from communication through verbal content, paralinguistic elements of the voice during speech enable individual recognition and assessment of the speaker's physical characteristics such as sex (Puts, Apicella, & Cárdenas, 2012), body size (Feinberg, Jones, Little, Burt and Perrett, 2005, Xu et al., 2013), physical strength (Sell et al., 2010), femininity (Feinberg, Jones, DeBruine, et al., 2005, Feinberg, 2008), attractiveness (Feinberg, Jones, DeBruine, et al., 2005, Feinberg, Jones, Little, Burt and Perrett, 2005, Xu et al., 2013), conception risk (Pipitone & Gallup, 2008), and sexual maturity (Mulac & Giles, 1996). In humans, perceived attractiveness and mate quality can be manipulated by artificially lowering the pitch of male voices or artificially increasing it in female voices, commensurate with sex-typical vocal properties (Collins, 2000, Feinberg, Jones, Little, Burt and Perrett, 2005). In fact, there is evidence for increased reproductive success in traditional societies for both low-pitched males (Apicella, Feinberg, & Marlowe, 2007), and high-pitched females (Atkinson et al., 2012).
In addition, vocal parameters can be intentionally varied, for example altering the intensity (loudness), rhythm and pitch. The classic example of such intentional modulation is infant directed speech (IDS) (Ferguson, 1977, Falk, 2005), in which adults alter vocal characteristics such as pitch, cadence and intonation contours when speaking to infants. Infants prefer these altered signals over adult-directed speech (Fernald & Kuhl, 1987), and it has been suggested that IDS aids human acquisition of vocal language (Burnham, Kitamura, & Vollmer-Conna, 2002) and might underpin the origins of musicality (Dissanayake, 2000, Trehub, 2003). In human and animal social interactions, modulations of the intensity of speech or vocalisations are often associated with hostility (Collias, 1960, Kudo, 1987) and dominance (Ohala, 1982, Tusing and Dillard, 2000), and changes in intensity contribute to emotional expression (Baker, 2001). Regarding pitch modulations, men lower their voices during competitive interactions when they perceive themselves as physically dominant (Puts, Gaulin, & Verdolini, 2006), and while women have been found to increase voice pitch when directing speech towards attractive faces (Fraccaro et al., 2011), both men and women have also been found to lower their voice pitch when speaking attractive targets of the opposite sex (Hughes, Farley, & Rhodes, 2010). This suggests that, while more evidence for specific types of modulation is needed (e.g. in the case women responding to attractive opposite-sex stimuli), modulations do actually occur. Similar subtle modulation in voices might be expected in courtship contexts. In fact, there is evidence of vocal differences between speech directed towards romantic partners and same-sex friends which can be detected by listeners (Farley, Hughes, & LaFayette, 2013), and intentional voice manipulations make female voices, but not male voices, sound more attractive (Hughes, Mogilski, & Harrison, 2013; see also Fraccaro et al., 2013). Indeed, such modulations occur in other species including frogs (Ryan, 1980), koalas, Phascolarctos cinereus (Charlton, Ellis, Brumm, Nilsson, & Fitch, 2012), fallow deer, Dama dama (Charlton & Reby, 2011), red deer, Cervus elaphus (Reby et al., 2005, Reby et al., 2010), and birds. For example, in the zebra finch, Taeniopygia guttata, males sing more rapidly to females than when they sing alone, producing syllables with lower spectral variability (Kao & Brainard, 2006).
Studies aiming to measure the effects that acoustic parameters have on human communication are hampered by the confounding influence of verbal content. To address this issue, many studies record voices enunciating vowel sounds or speaking standard sentences, or measure responses to voices with artificially manipulated vocal parameters (e.g. Feinberg, Jones, Little, Burt and Perrett, 2005, Puts et al., 2007). These methodologies have provided important insights into the role that vocal parameters play in human communication. Similarly, to study vocal modulation, and unlike research on animals or IDS (where infants understand little or none of the semantic content), it is necessary to control the confounding influence that verbal content may play. Some studies have used scripted speech (e.g. Hughes et al., 2010, Fraccaro et al., 2011), therefore eliminating prosodic variation in vocal acoustic parameters. Although challenging, testing free, unscripted speech is ideal, as standard sentences may not accurately reflect the levels of natural vocal variation; standardised sentences likely limit the kind of spontaneous paralinguistic variation found in normal free speech, as well as the nuance and range of paralinguistic modulation known at least to occur in IDS, which is characterised by an extreme range of pitches, typically starting from a high pitch and containing many glissandos. Finally, while some studies have successfully tested natural vocal variation during speech (e.g. Hodges-Simeon et al., 2010, Hodges-Simeon et al., 2011), apparent paralinguistic modulation in one language may be underpinned by specific parameters of that language (e.g. rhythm, intonation, and use of specific phonemes). Here we circumvented these issues (i.e. the confounding influence of verbal content, using unscripted speech, and the potential effects of one language in paralinguistic modulation) by adopting a cross-language design involving two model languages.
Based on evolutionary theory and the current knowledge of human voices, we hypothesized that males and females would modulate their acoustic parameters (study 1), depending on the sex and attractiveness of the target, to affect the way in which they would be perceived. Because speech intensity is associated with hostility and dominance (Collias, 1960, Ohala, 1982, Kudo, 1987) we expected participants to speak with increased intensity in responses to same-sex targets, in comparison to opposite-sex targets. Furthermore, because emotional expressiveness is attractive (Sprecher, 1989), and changes in intensity improve emotional expression (Baker, 2001), we expected participants to speak with increased variability in intensity when responding to opposite-sex targets, and especially when those targets were attractive. In addition, based on the body of knowledge produced by studies testing perception of manipulated pitch, we predicted that women, and especially men, would emphasise sex-specific vocal characteristics when responding to attractive individuals of the opposite sex (i.e. lowering F0 in men, and increasing it in females), and that both sexes would increase F0 variability, in order to sound more attractive to those attractive targets. Additionally, we predicted that these modulations would be detectable by naive listeners (study 2), and that speakers would sound more attractive when speaking to attractive versus unattractive targets.
Section snippets
Study 1
First, we tested the possibility that individuals might alter vocal parameters in speech directed at potential romantic partners or competitors depending on the attractiveness of the listener. Recorded voice samples from speakers of two different languages were used to avoid the possibility that apparent paralinguistic modulation in one language might be reinforced by specific parameters of that language (e.g. rhythm, intonation, and use of specific phonemes).
Study 2
Findings from study 1 indicate that paralinguistic parameters vary depending on the attractiveness of the target, but did not test the perception of this modulation. For it to be functionally relevant and have an effect on mate choice, it must be perceptually detectable and influence proceptivity towards the speaker. Study 2 aimed to investigate whether this is indeed the case.
General discussion
Although previous results suggest that voice pitch plays a role in human courtship (Puts et al., 2006, Fraccaro et al., 2011), our cross-language experimental design provides new insights into the specific nature and mechanisms of paralinguistic modulation involved in courtship. While the two languages (English and Czech) are both European, they lie on separate branches of the Indo-European family with several millennia of largely independent development (Gray, Atkinson, & Greenhill, 2011) and
Supplementary Materials
The following are the Supplementary data to this article.
Acknowledgments
We are grateful to Lisa M. DeBruine and two anonymous reviewers for valuable comments on the manuscript. We thank V.M. Mileva and J. Kreisinger for their constructive comments, A. Murray and K. Potyszová for their help in collecting data, and all our participants. J.D.L. is funded by the Colombian Administrative Department of Science, Technology and Innovation (COLCIENCIAS). J.H. is supported by the Czech Science Foundation grant (P407/10/1303) and J.H. and K.K. by the Charles University
References (57)
- et al.
Female koalas prefer bellows in which lower formants indicate larger males
Animal Behaviour
(2012) Men’s voices and women's choices
Animal Behaviour
(2000)- et al.
Correlated preferences for men’s facial and vocal masculinity
Evolution and Human Behavior
(2008) - et al.
The voice and face of woman: one ornament that signals quality?
Evolution and Human Behavior
(2005) - et al.
Manipulations of fundamental and formant frequencies influence the attractiveness of human male voices
Animal Behaviour
(2005) - et al.
Acoustic determinants of infant preference for motherese speech
Infant Behavior and Development
(1987) The biology and evolution of music: a comparative perspective
Cognition
(2006)- et al.
Faking it: deliberately altered voice pitch and vocal attractiveness
Animal Behaviour
(2013) - et al.
Verifying the primacy of voice fundamental frequency in social status accommodation
Language and Communication
(2001) - et al.
What is specific to music processing? Insights from congenital amusia
Trends in Cognitive Sciences
(2003)