Sexual dimorphism and quality of voice across the menstrual cycle

Pavela Banai, Irena. (2017). Sexual dimorphism and quality of voice across the menstrual cycle. PhD Thesis. Filozofski fakultet u Zagrebu, Department of Psychology.
(Poslijediplomski doktorski studij psihologije) [mentor Šimić, Nataša].

Preview

PDF (Croatian) Download (3MB) | Preview

Abstract

Previous studies showed various changes in women's behavior across the menstrual cycle. During their fertile period, women reported an increased preference for male masculine traits, extra-pair sexual interests and fantasies, tendency to flirt with an attractive man, to wear more revealing clothes and attractive make-up. Beside changes in behavior, studies showed changes in some physical characteristics, with female face, odour, and gait being more attractive in the fertile period. It is assumed that these changes represent an adaptation with the purpose of finding an adequate partner and enhancing the genetic quality of the offspring. It is proposed that these changes are regulated by the fluctuations of sex hormone levels across the cycle. Receptors for sex hormones have been found on the vocal folds, thus suggesting a link between hormone levels and vocal fold function. Increased estrogen levels result in thickening of the laryngeal mucosa and increased mucus production, while increased progesterone levels result in drying of the laryngeal mucosa. These physiological changes might cause changes in voice production, as well as in voice quality and perceived attractiveness. Indeed, previous studies showed that attractiveness ratings of women's voices varied across the menstrual cycle, with higher ratings for voice recordings made in fertile phase. It was also found that women with feminine faces have more feminine and attractive voices. Moreover, studies showed that women tend to change their voice when interacting with an attractive man. This implies that voice is a valuable source of biologically important information, such as women's fertility status and reproductive value. However, attempts to identify specific acoustic characteristics that convey that information produced mixed results. There are several methodological aspects to consider in explaining inconsistent findings. Most of the researchers focused on investigating changes of sexually dimorphic vocal features, such as fundamental frequency and its variability, while less focus has been placed on assessing features that signal vocal quality, such as perturbation measures (jitter and shimmer). Furthermore, the methods used to identify cycle phases vary between studies, and most researchers relied on self-report methods to identify the fertile cycle phase. In addition, in most of the studies, women's voice was recorded during read speech or phonation of vowel, numbers, and sentences with no context. Those voice samples cannot represent a natural variation of vocal characteristics that could be present in everyday communication. Considering this, vocal changes across the cycle should be investigated during real or simulated communication. Assuming that vocal changes across the ovulatory cycle represent an adaptation to attract potential mates, voice changes should be explored in a social context in which they actually might occur – mating context. Therefore, the purpose of this study was to investigate sexually dimorphic vocal features and quality of voice during simulated interaction of women and other men and women. Interaction was simulated in different phases of the menstrual cycle among women with natural cycle and users of hormonal contraceptives. The voices of 48 women with natural cycle and 20 users of hormonal contraception were obtained in menstrual, late-follicular and luteal phase. Cycle phases were identified by counting method and luteinizing hormone surge confirmation in the mid-cycle. In each cycle phase, voice was recorded in a neutral situation of vowel production. In order to examine vocal changes in a simulated social interaction, women had a task to introduce themselves and leave a voice message for masculine and feminine versions of prototypical male and female faces, which were presented on a screen in a random order. After they recorded four messages, they rated the attractiveness of prototype faces. Voice samples were analyzed using Praat, in which sexually dimorphic vocal characteristics, as well as vocal quality was measured. With regard to the control situation and vowel recordings, the results showed greater vocal stability (lower fundamental frequency variability) among women with natural cycle in cycle phases with higher sex hormone levels (late-follicular and luteal phase), compared to the menstrual phase. Based on the changes in minimum fundamental frequency values, women had more feminine voice in the late-follicular phase compared to the rest of the cycle. In addition, voice intensity was the lowest in the luteal phase, probably because of the lower activation levels in a cycle phase with high levels of progesterone. Among users of hormonal contraception, there were no voice changes across the cycle. Comparison of the two groups of women revealed that the natural cycle group had higher formant dispersion. Naturally cycling women also had higher values of harmonics to noise ratio, but only in menstrual and late-follicular phase, which implies greater vocal quality. Absence of differences between the groups in luteal phase might be due to the similarity of their hormonal profiles in this phase. Analysis of voice recordings obtained during a message recording to feminine and masculine man and woman showed that women with natural cycle had a tendency to speak with more feminine voice (higher fundamental frequency) and a greater voice intensity when leaving a message for feminine, compared to masculine man. This result was found regardless of the cycle phase. Moreover, women had greater voice intensity in fertile period when leaving a message for a feminine man, compared to feminine women. In the same phase, women who were in a relationship had a bit higher fundamental frequency in the same situation. Greater vocal femininity and intensity directed to a feminine man is in line with the attractiveness ratings obtained in this study, which showed that feminine man was more attractive than masculine. These results imply vocal modulation with the purpose of attracting a potential mate, which was more pronounced in a fertile period among naturally cycling women who were in a relationship. Furthermore, changes in the minimum fundamental frequency showed that women in a fertile period produced less feminine voice directed to feminine woman, compared to the rest of the menstrual cycle. This might imply dominance signaling to an attractive rival when the probability of conception is high. Moreover, small changes in fundamental frequency variability showed that women increased attractiveness of their own voice when leaving a message for an attractive feminine woman in fertile period. These results could also be explained in the context of intrasexual competition. Among users of hormonal contraception, no vocal changes in different cycle phases were found. Results obtained in this study imply that voice changes among women with natural cycle are regulated by sex hormone levels. To sum up, hormonal regulation of vocal changes in a control situation might be related to the direct effects of hormones on the vocal cords, or indirect effects of hormones on activation and arousal. These results are in line with the proposed proximal causes of vocal changes across the menstrual cycle. Moreover, vocal changes during message recordings for feminine and masculine man and women in various cycle phases imply a regulating role of the sex hormones in activating behaviors directed towards the potential mate and the potential rival. Those behaviors include voice modulation in a way to appear more attractive or dominant. In addition, these behaviors appear to be more pronounced among women who were in a relationship. The results of this study also point out the importance of investigating vocal changes across the cycle in a defined social context.

Actions (login required)

View Item