When presented with a sound that is ambiguous between two phonemes, listeners automatically adjust their phoneme boundary. For instance, upon hearing a series of /s/-final words produced with a sound that is ambiguous between /s/ and /f/, listeners will shift their perceptual boundary between these two sounds towards /f/, such that tokens in the middle of a continuum between /s/ and /f/ will be more often categorized as /s/. Here, we will examine whether such perceptual learning is constrained by the phonetic context. In particular, we will test the hypothesis that when the ambiguity can be attributed to coarticulation with a following sound, perceptual learning is blocked.
Student profile: Preferably some knowledge of phonetics and some experience with Praat, Audacity, and/or Matlab.