William J. Idsardi
Department of Linguistics and Program in Neuroscience and Cognitive Science, University of Maryland

ID 1745
[full paper]

This presentation reviews the use of distinctive features for the mental representation of speech sounds, briefly considering three bases for feature definition: articulatory, auditory and translational. We then review several recent neuroimaging studies examining distinctive features using magnetoencephalography (MEG). Although this area of research is still relatively new, we already have interesting findings regarding vowel place, nasality and consonant voicing. Although this research is not yet definitive, some refinements of these experiments can be expected to yield important results for feature theory, and more generally for our understanding of the neural computations that underlie the transformations between articulatory and auditory space necessary to produce and perceive speech.