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

Abstract

Prior research demonstrated that, upon hearing a single phone, listeners differentiated between gay and heterosexual male talkers of American English. For instance, researchers found that listeners relied on three consonants (e.g., /l/, /n/, and /s/) to form their judgments. It is unclear whether these findings could be replicated and whether there are additional consonants that listeners used. To further explore this, 23 consonants were examined (e.g., /b/, /v/, /s/, /n/, /t/, /f/, /ɛr/, /dʒ/, /θ/, /k/, /ɹ/, /m/, /l/, /g/, /d/, /ʒ/, /w/, /p/, /ɛl/, /ʃ/, /z/, /j/, and /ð/). For 21 of these consonants, listeners heard two tokens each, and heard one token of /θ/ and /ð/. Results were mixed. Listeners relied on both instances of /s/, /w/, /ɛl/, and /z/ to form their judgments, while they relied on only one instance of /b/, /n/, /ɛr/, /dʒ/, /k/, /ɹ/, /l/, /d/, /ʒ/, /p/, and /j/. Listeners didn’t rely on any instances of /v/, /t/, /f/, /g/, /m/, and /ʃ/; they relied on /ð/ and did not rely on /θ/. The only consistent finding, across multiple experiments, is that participants used /s/. Under certain circumstances, they may use additional consonants, but it doesn’t appear as if they use them consistently.