| Abstract | [Introduction]
During spontaneous speech, we create utterances in our minds as we produce them out loud. As we formulate the next linguistic chunk to be spoken, we hold it momentarily, if only parts of it, as a perceptual image. We not only know what it is, in general, that we plan to say, but we have a rather abstract idea of how it will sound. Unconsciously, we also have an image of how it will feel in terms of touch, pressure sensations, and movements and positions of speech organs. We know these things because we know our own vocal tract possibilities and our own voicing capabilities. We have heard and felt ourselves talk for years. In almost every speaking situation, we are able to feel and hear ourselves speak. We know so well how we will sound and feel when producing speech that we can continue to produce perfectly intelligible speech in artificial situations in which we are prevented from hearing ourselves, as under auditory masking, or are prevented from feeling surface sensations, as under oral anesthesia. It is likely that, in these instances, we continue to receive information on our performance from our muscles and from feedback mechanisms contained within the central nervous system (CNS).
Less skilled speakers must depend upon auditory and tactile feedback more than speakers who have well-established speech-production systems. Young children developing speech, and speakers of any age attempting to learn a new language, must use all available feedback channels in their efforts to match the sound patterns of the new language with the sensations produced by their own imitations. Children with congenital natural-sounding speech patterns. In contrast, those who have learned to speak before the onset of deafness produce good speech with only slight time (Fry, 1966). The use that one is able to make of feedback from one’s own speech seems to vary with age as well. Children first acquiring language, whether they are learning one or two languages, are particularly adept at matching their own speech to the models provided. They use feedback to emulate the segmental, intonational, and rhythmic characteristics of the languages to which they are exposed. Too often, however, older people learning a second language fail in their efforts to match the segmental and suprasegmental aspects of the language as spoken, even though they may have mastered the grammar and vocabulary. The auditory, tactile, and muscle-moving images that they have stored for their first language with an accent: the sound patterns of the first language persist in the second.
In this chapter, the goal is not to provide a comprehensive review of the literature on motor control, but rather to state a current view of how speech control may operate in skilled and nonskilled speaker-and to include examples from recent research in support of this view. However, before discussing the ways in which feedback may operate during speech acquisition and during established speech, a brief considering of the control mechanisms themselves and the experimental effects of altering the information that they provide is in order. |
|---|
