ABSTRACT
Since the pioneering study of infant speech perception by Eimas, Siqueland, Jusczyk, and Vigorito (1971), it has become widely acknowledged that infants have sophisticated perceptual abilities that enable them to discriminate subtle phonetic distinctions (Aslin, Pisoni, & Jusczyk, 1983), some of which are not easily discriminated by adults (Werker & Tees, 1984; Werker & Polka, 1993). Unfortunately, these sophisticated perceptual abilities have been demonstrated under listening conditions that are not typical of the infant’s natural environment. For example, high-amplitude sucking, conditioned headturning, and habituation-of-looking techniques present the infant with brief speech tokens, separated by pauses, and repeated many times in a low-noise environment. These are not the conditions typical of natural language processing where words are generally presented once, in rapid succession, and often in noisy environments. What remains unclear, therefore, is the extent to which infants deploy their sophisticated perceptual abilities when confronted with fluent speech.
