Abstract
An unresolved question in speech perception is how speech signals with speaker variation are mapped onto their perceptual representations. In this study, this issue was examined using a written-word/spoken-word matching paradigm, where listeners could adjust phonetic expectations of spoken words carrying lexical tones according to speaker-specific F0 cues contained in a preceding speech context, to analyse the tone of the incoming spoken word. Behavioural results showed that Cantonese listeners perceived spoken words differently, in a way compatible with the adjustment of F0 expectations of lexical tones to accommodate between- and within-speaker variation in F0. Electrophysiologically, effects of F0 expectation adjustment were found in the phonological mapping negativity (PMN) time-window (250–310 ms after spoken word onset). These results suggest that phonetic representations of lexical tones are adjustable in a speaker- and context-specific manner, with the adjustment occurring no later than pre-lexical phonemic processing. These findings are consistent with exemplar theory.
Original language | English |
---|---|
Pages (from-to) | 175-195 |
Number of pages | 21 |
Journal | Language, Cognition and Neuroscience |
Volume | 33 |
Issue number | 2 |
DOIs | |
Publication status | Published - 7 Feb 2018 |
Keywords
- Cantonese
- lexical tones
- phonological mapping negativity
- signal-to-representation mapping
- Speaker variation
ASJC Scopus subject areas
- Language and Linguistics
- Experimental and Cognitive Psychology
- Linguistics and Language
- Cognitive Neuroscience