Abstract
The events of speech production can be characterized variously by the sounds that are produced (whether assessed subjectively or spectrographically), by the movements and shapes of the vocal and pulmonary tracts, or by the neuromuscular activities that cause these movements and changes of shape. There is a growing interest in experimental studies of the neuromuscular events, owing in part to the fact that they occur early enough in speech production to escape some of the encoding imposed on the message. As a working hypothesis, one might suppose that some limited set of the neural signals to the articulatory muscles would retain a one-to-one correspondence with linguistic units of the message. Such characteristic neural signals, i.e., the motor commands, could then provide a most useful basis for a phonological description of spoken language; further, such motor commands might well play a significant rôle in speech perception. The experimental technique that is most directly applicable is electromyography. Early results from several laboratories seem highly encouraging, though many problems of both method and interpretation remain to be solved.

This publication has 0 references indexed in Scilit: