This chapter focuses on one of the first steps in comprehending spoken language: How do listeners extract the most fundamental linguistic elements-consonants and vowels, or the distinctive features which compose them-from the acoustic signal? We begin by describing three major theoretical perspectives on the perception of speech. Then we review several lines of research that are relevant to distinguishing these perspectives. The research topics surveyed include categorical perception, phonetic context effects, learning of speech and related nonspeech categories, and the relation between speech perception and production. Finally, we describe challenges facing each of the major theoretical perspectives on speech perception.

No abstract

This study was designed to test the iambic/trochaic law, which claims that elements contrasting in duration naturally form rhythmic groupings with final prominence, whereas elements contrasting in intensity form groupings with initial prominence. It was also designed to evaluate whether the iambic/trochaic law describes general auditory biases, or whether rhythmic grouping is speech or language specific. In two experiments, listeners were presented with sequences of alternating / / syllables or square wave segments that varied in either duration or intensity and were asked to indicate whether they heard a trochaic (i.e., strong-weak) or an iambic (i.e., weak-strong) rhythmic pattern. Experiment 1 provided a validation of the iambic/trochaic law in Englishspeaking listeners; for both speech and nonspeech stimuli, variations in duration resulted in iambic grouping, whereas variations in intensity resulted in trochaic grouping. In Experiment 2, no significant differences were found between the rhythmic-grouping performances of English-and French-speaking listeners. The speech/ nonspeech and cross-language parallels suggest that the perception of linguistic rhythm relies largely on general auditory mechanisms. The applicability of the iambic/trochaic law to speech segmentation is discussed.

Inter- and intratalker variation in the production of lexical tones may contribute to acoustic overlap among tone categories. The present study investigated whether such category overlap gives rise to perceptual ambiguity and, if so, whether listeners are able to reduce this ambiguity using contextual information. In the first experiment, native Cantonese-speaking listeners were asked to identify isolated Cantonese level tones produced by 7 talkers. Identification accuracy was significantly higher when the presentation of items was blocked by talker rather mixed across talkers. In the second experiment, listeners identified the final (target) tone of 6-syllable semantically neutral sentences with f0 patterns of the context (i.e., the first 5 syllables) altered. The same target tone was identified differently depending on the context. In the third experiment, the context portions of stimulus sentences from the second experiment were divided into 2 halves, and their f0 patterns were altered independently. In identifying the target tone, listeners relied more heavily on the f0 pattern of the second (last) half of the context. These results are discussed in relation to characteristic inter- and intratalker variations of lexical tones.

Japanese quail (Coturnix coturnix) learned a category for syllable-initial [d] followed by a dozen different vowels. After learning to categorize syllables consisting of [d], [b], or [g] followed by four different vowels, quail correctly categorized syllables in which the same consonants preceded eight novel vowels. Acoustic analysis of the categorized syllables revealed no single feature or pattern of features that could support generalization, suggesting that the quail adopted a more complex mapping of stimuli into categories. These results challenge theories of speech sound classification that posit uniquely human capacities.