Vowel Phonemes Map Out on an Emotional Valence Continuum

This study expands on findings by Yu, McBeath, & Glenberg (2019) which demonstrated a relationship between the pronunciation of English vowel phonemes and emotional valence due to embodied cognition. That study found that single syllable words containing the phoneme /i:/ (as in “gleam”) were reliably rated as more positive than matched words containing the phoneme /ʌ/ (as in “glum”). The findings are consistent with the idea that the facial musculature when smiling is more conducive to making the /i:/ sound, while that of frowning or grimacing is more conducive to making the /ʌ/ sound. That study only compared the phonemes /i:/ and /ʌ/, which are opposite extremes of phoneme similarity (second formant frequency). The present study expands on this finding by testing the relative emotional valence ratings of matched single-syllable words containing /i:/ vs /ʌ/ plus two intermediate phonemes, /ɪ/ (as in “bit”), and /ɔ/ (as in “bought”). The new findings replicate the Gleam-Glum effect, and provide support for a weak ordering hypothesis for the intermediate phonemes, but not a strong ordering. The weak ordering hypothesis is that single-syllable words containing a middle vowel phoneme that is intermediate to /i:/ and /ʌ/ in musculature and acoustic features are also generally rated as intermediate in emotional valence. The strong ordering hypothesis is that the intermediate phonemes are each differentially rated in emotional valance in precisely the same order as determined acoustically. The pattern of results found is consistent with the Russell Circumplex Model of emotion at a cursory level, but individual emotions do not fully conform to a simple 2-D model that generalizes to similarity judgments of phonemes. Nevertheless, the work supports that facial musculature associated with visually discernible emotions generally relates to a phonetic acoustic continuum.