Blackwell Reference Online

Extract

The study of speech perception investigates how we are able to identify in the human voice the meaningful patterns that define spoken language. Research in this area has tended to focus on how humans perceive minimal linguistic contrasts known as “phonemes” – how we distinguish “pat” from “bat” or “bit” from “bet” for example. Although speech perception has traditionally been the study of phoneme perception, an account is also needed for how we identify spoken words and comprehend sentences in connected fluent speech. Defining “speech perception” very narrowly in terms of phoneme perception or nonsense syllable identification was a useful and reasonable simplification in the early days of the field, but the drawbacks of conceptualizing the problem purely in these terms have become increasingly apparent in recent years. Speech perception stands to audition much as face perception stands to vision, in that it is a differentially developed response to a particular type of physical signal within a more general sensory modality. Because speech perception involves how we categorize vocalizations characteristic of our own species, the key issues cannot be addressed through the use of animal models; the relevant data come, almost exclusively, from external observations of human behavior. The relevance of how other species perceive human speech signals is doubtful (see Trout, 1998 ), ... log in or subscribe to read full text