Speaker: Meghan Sumner (Stanford)
Date/Time: Thursday, Apr 12, 12:30-1:45p
Location: 32-D461
Phonetic variation in speech is often considered a barrier to understanding spoken words. In this talk, I present data that phonetic variation is necessary and beneficial to understanding accented speech. Within the perceptual learning paradigm, listeners are exposed to p- initial words in English produced by a native speaker of French. Critically, listeners are either trained on these words with an invariant VOT based on the speaker’s mean VOT, or with variant VOTs including native English-like to native French-like examples. While a gross boundary shift is made for participants exposed to the variable VOTs, no such shift is observed for the invariant stimuli. These data run counter to models that predict consistent exposure to an invariant stimulus should ultimately result in perceptual learning, and suggest that when adjusting to accented speech, invariance is an obstacle.