DEVELOPMENTAL SPEECH PERCEPTION OF THREE ACOUSTIC CUES ASSOCIATED WITH PLACE OF ARTICULATION

JOAN ESTHER SUSSMAN, Purdue University

Abstract

The purpose of this research was to study the relationship between age and perception of acoustic cues in 30 normal-language children between five and ten years of age and in 10 adults. To achieve this purpose, four sets of synthetic speech syllables varying along a seven-step, bilabial-to-alveolar place of articulation continuum were created and arranged into three experimental paradigms: discrimination, labeling and adaptation. The major acoustic variable studied was a change in second and third formant frequency (F2, F3) transition trajectory. The effects of length of the F2, F3 transitions as well as the presence or absence of a white noise burst were also studied. Results showed that children's speech perception abilities are different from adults', even at 10 years of age. The adults were shown to have fine discrimination abilities without the need for extensive training. For the discrimination task, children's sensitivity to fine acoustic cues and their response strategies changed developmentally and were most like that of adults in the 9-10-year-old group. The findings also showed that adaptation is age-dependent as adults were the only group to demonstrate a significant adaptation effect with a very small, nonsignificant, developmental pattern evident for children's responses. Further study with nonspeech stimuli is necessary to clarify whether sensitivity or a phonetic response bias was a more important factor for the children's response patterns in the discrimination and adaptation tasks. The results also indicated that normal labeling performance can occur without very fine discrimination as all subjects performed similarly on the labeling task. This tended to support Stevens' (1972) quantal theory of speech perception. The continuum-type effect of lengthened F2-F3 transitions significantly changed discrimination, labeling and adaptation performance while the burst presence only influenced performance in part of the adaptation task. The signal detection approach and the use of the "change/no change" format in the discrimination task, as well as the use of the adaptation and labeling tasks, were productive in revealing different types of processing used by the different aged subjects.

Degree

Ph.D.

Subject Area

Speech therapy

COinS