Phonology impacts segmentation in speech processing
Luca Onnis, Padraic Monaghan, Nick Chater, & Korin Richmond
Abstract
Pena, Bonatti, Nespor, and Mehler (2002) investigated an artificial language where the
structure of words was determined by nonadjacent dependencies between syllables. They
found that segmentation of continuous speech could proceed on the basis of these
dependencies. However, Pena et al.`s artificial language contained a confound in terms of
phonology, in that the dependent syllables began with plosives and the intervening
syllables began with continuants. We consider three hypotheses for the role of phonology
in speech segmentation: (1) participants bring to bear knowledge about the distribution of
sounds in their native language to the artificial language learning task; (2) unvoiced
plosives in speech are preceded by a gap which influences segmentation at that point; and
(3) phonological similarity between dependent syllables contributes to learning the
dependency. In a series of experiments controlling the phonological and statistical
structure of the language, we found that both starting with a plosive and phonological
similarity between first and third segment contributed to segmentation performance.
Learning did not occur when there was no sharing of phonological structure between
dependent syllables. Phonological processing, therefore, provides a fundamental
contribution to higher-level computational tasks.
Please email suggestions/errors to lo35@cornell.edu |