Do voices carry valid information about a speaker's personality?

Abstract

Research on links between peoples’ personality traits and their voices has primarily focused on other peoples’ personality judgments about a target person based on a target person’s vocal characteristics, particularly voice pitch. However, it remains unclear whether individual differences in voices are linked to actual individual differences in personality traits, and thus whether vocal characteristics are indeed valid cues to personality. Here, we investigate how the personality traits of the Five Factor Model of Personality, sociosexuality, and dominance are related to measured fundamental frequency (voice pitch) and formant frequencies (formant position). For this purpose, we conducted a secondary data analysis of a large sample (2,217 participants) from eleven different, independent datasets with a Bayesian approach. Results suggest substantial negative relationships between voice pitch and self-reported sociosexuality, dominance and extraversion in men and women. Thus, personality might at least partly be expressed in people’s voice pitch. Evidence for an association between formant frequencies and self-reported personality traits is not compelling but remains uncertain. We discuss potential underlying biological mechanisms of our effects and suggest a number of implications for future research.

Voice pitch is negatively associated with sociosexual behavior in males but not in females

“…Our results show a negative correlation between mean fundamental frequency and sociosexual behavior -as measured by the behavior facet of the Revised Sociosexual Orientation Inventory (Penke and Asendorpf, 2008) -in males (r 100 = −0.27, p = 0.023; 95%CI = [−0.45, −0.07]) but not in females (r 147 = −0.17 , p = 0.103, 95%CI = [−0.32, −0.00]); in other words, our findings show a trend where men with lower-pitched voices exhibit a more unrestricted sexual behavior. This association is in line with previous studies which have shown that male-speech lower F0 is related to higher scores on the SOI-R scale (Valentova et al, 2019;Stern et al, 2021) and that males with lower-pitched voices have higher reproductive success and more children born to them (Apicella et al, 2007). While Stern et al (2021) did find an association between voice pitch and sociosexuality in women, neither the studies by Apicella et al (2007) and Valentova et al (2019) nor the present study found an association between female F0 and sociosexuality.…”

No evidence for an association between facial fluctuating asymmetry and vocal attractiveness in men and women

“…3D facial FA measurement. Participants' faces were 3D-scanned twice for men and overall twelve times for women (thrice for each of 2 sessions during the luteal phase and sessions during the fertile phase of their ovulatory cycles, of which the scans from the first luteal sessions were used, see Jünger et al, 2018 andStern et al, 2019 for details) using a 3dMD face scanner. For each participant one scan was chosen with the most neutral facial expression and standardized head position, while some participants (males: n = 28, females: n = 4) had to be excluded due to poor scan quality or interfering facial hair for the males.…”

“…The final sentence was used for voice ratings (mean length of the voice stimuli = 7.36 secs, SD = 0.91 secs). Since women's voice were recorded four times (twice per fertility status, see Jünger et al, 2018 andStern et al, 2019), the recordings concurrent with the face scans from the first infertile session were used. After excluding five recordings for men due to technical issues and participants not wanting their recordings to be used further, the final stimulus sets comprised 160 recordings for men and 157 for women.…”

No evidence for an association between facial fluctuating asymmetry and vocal attractiveness in men or women

“…Participants' faces were 3D-scanned twice for men and overall twelve times for women (three times for each of two sessions during the luteal phase and two sessions during the fertile phase of their ovulatory cycles, of which the scans from the first luteal sessions were used, see Jünger et al, 2018 andStern et al, 2019 for details) using a 3dMD face scanner. For each participant one scan was chosen with the most neutral facial expression and standardized head position, while some participants (males: n = 28, females: n = 4) had to be excluded owing to poor scan quality or interfering facial hair for the males.…”

“…The final sentence was used for voice ratings (mean length of the voice stimuli = 7.36 s, SD = 0.91 s). Since women's voices were recorded four times (twice per fertility status, see Jünger et al, 2018 andStern et al, 2019), the recordings concurrent with the face scans from the first infertile session were used. After excluding five recordings for men owing to technical issues and participants not wanting their recordings to be used further, the final stimulus sets comprised 160 recordings for men and 157 for women.…”