Speech perception
Dissertation research: Perceptual properties of coalescence
Under construction!
Phonological knowledge and discriminability
The processing of the incoming speech signal can occur in two distinct general modes: autonomous and interactive. The autonomous mode of processing is characterized by an apparent lack of influence from sources of linguistic knowledge or bias (e.g. the lexicon) in certain perceptual tasks in which an influence is expected. The interactive mode is characterized by the influence of linguistic knowledge (i.e. feedback) in a perceptual task in which an influence is expected. A classic example of the interactive mode are the lexical biases on phoneme identification reported by Ganong (1980). While it is true that the literature is full of studies that report interactive processing (i.e. some linguistic bias) and relatively devoid of studies reporting autonomous processing (i.e. the lack of a bias where one would be expected), it is by no means that we have a good understanding of the factors that determine whether autonomous or interactive processing occurs. In particular, the role of task type and source of linguistic influence is not well understood. The large majority of the studies bearing on the question of processing mode have used recognition tasks (e.g. phoneme identification, word detection, etc.) and have tested for various influences or biases directly from the lexicon (e.g. wordhood, lexical statistics). The studies reported below address some of these missing combinations by (i) examining discrimination tasks in two different formats and using them (ii) to test for the influence or feedback from listeners' phonological knowledge using non-word stimuli, thus factoring out possible lexical influences.
Reference: Key, Michael (2008). Interactive and autonomous modes of speech perception: Phonological discrimination in English and French listeners. Paper to be presented at LabPhon11, Victoria University of Wellington, New Zealand. 30 June 2008.
Abstract (.pdf)