Judgments of self-identified gay and heterosexual male speakers of American English: Which vowels do listeners rely on to form their sexual orientation judgments?

Abstract

Prior research (Tracy et al., 2015) demonstrated that, upon hearing a single phone, listeners differentiated between gay and heterosexual male talkers of American English. For instance, listeners relied on seven vowels (e.g., /æ/, /eɪ/, /ε/, /iː/, /oʊ/, /ɑ/, and /uː/) to form their judgments. From this particular finding, it is unclear whether the results could be replicated and whether these are the only vowels that listeners rely on; it is possible that listeners might rely on additional vowels. To further explore these issues, the present study examined a wider range of vowels (e.g., /i:/, /ɪ/, /eɪ/, /ɛ/, /æ/, /ɑɹ/, /ɑ/, /ɔ/, /oʊ/, /u:/, /ai/, /aʊ/, /ɔɪ/, /ɪr/, / ^/, and /ju/) and, in most instances, presented two different tokens of each vowel to listeners. Some tokens of a vowel (e.g., /i:/, /ɛ/, /æ/, /oʊ/, /u:/, and /ju/) came from a stressed syllable and some tokens came from an unstressed syllable. Only four vowels (e.g., /ɔ/, /aʊ/, /ɔɪ/, and / ^/) were presented once. Across all instances of every vowel, participants were able to significantly discriminate between gay and heterosexual speakers. Thus, listeners use a wide range of vowels to form their sexual orientation judgments of male talkers.