Speakers can anticipatorily configure their vocal tracts to prepare for an upcoming utterance. The current study investigates this phenomenon by examining variation in anticipatory posturing. Realtime MRI images were acquired from 13 native speakers of English performing a consonant-vowel response task. The task was embedded in a 2 x 2 design which manipulated preparation (whether speakers had foreknowledge of the target response prior to a go-signal) and pre-response postural constraint (whether the response was preceded by a target articulatory posture). Analyses of pre-response articulatory postures show that only some speakers exhibited response-specific anticipatory posturing; furthermore, speakers differed with respect to the articulators used for anticipatory posturing, and in some cases the direction of effects. The results indicate that anticipatory posturing is unlikely to arise from a unitary optimization mechanism, and may instead result from joint optimizations or other aspects of the speech motor control system.

This content is only available via PDF.