ABSTRACT
Two important developments have provided impetus for research on speech perception. The first, almost 3 decades ago, was the development at Haskins laboratories of rules for synthesizing speech at the phoneme level (Liberman, Ingemann, Lisker, Delattre, & Cooper, 1959). From this came the finding that speech sounds which vary continuously along certain dimensions are perceived categorically by adults. That is, adults’ ability to discriminate between sounds on various continua was found to be limited by their ability to label or identify such sounds as belonging to different categories (Liberman, Harris, Hoffman, & Griffiths, 1957). This was taken as evidence for a special form of processing of speech by adults. In 1971 the second important development occurred when Eimas, Siqueland, Jusczyk, and Vigorito introduced a method to study speech discrimination in infancy. They found that infants as young as 1 month discriminate speech categorically, which strongly suggested that the special manner in which humans process speech is innate. This chapter is concerned with these two claims and their subsequent modification. We also report research that we believe has an important bearing on these claims.
