Adaptation to sentences and melodies when making judgments along a voice/non-voice continuum

Abstract

Perceptual contrastive adaptation effects can be used to identify important perceptual features or categories. Our previous work revealed contrastive adaptation effects between voice and non-voice categories when adapting with repeated single vowels or musical tones. The current study investigated whether the effect generalizes to adaptors with higher ecological validity—sentences and melodies. Ten-step continua between a “voice” (female /a/, /o/, or /u/ vowels) and an “instrument” (bassoon, horn, or viola) were generated for each possible pair. A sentence spoken by a female voice, or a melody played on bassoon or horn, was presented, followed by a test stimulus from along the continuum. Contrastive adaptation effects were observed, with the test stimulus more likely to be identified as a voice following a musical melody and vice versa. Pilot data showed similar trends when the female voice adaptors were replaced with their male counterparts, suggesting that the effects may generalize across speaker gender and fundamental frequency. The results show that contrastive adaptation to voice and non-voice stimuli is a robust effect that does not rely on the repetition of simple adaptors or on a shared frequency range between the adaptors and test stimuli. [Work supported by NIH grant R01DC005216.]