Evaluation of speech perceptual development

UCL Phonetics and Linguistics

Evaluation of speech perceptual development using speech pattern audiometry tests

A research project funded under the MRC small-grants scheme

Administrative Details

Grant Period:	August 1997-July 1998
Grant Award:	£37,842
Investigators: Research Fellow	Valerie Hazan, UCL Frederika Holmes, UEA Sarah Barrett, UCL

Overview

This study, although brief, is designed to address a major issue. Experimental evidence is beginning of confirm that the acquisition of speech skills by the normally-developing child is a structured process. There is, however, very little normative material which can be used to gain a well-founded practical understanding of the underlying phonological and perceptual processing mechanisms. The normative reference needs of workers in quite diverse areas such as speech and language therapy, audiology and English as a second language (ESL) are all currently unmet.

Models of speech perception development have suggested: (1) that development in the ability to categorise phonemic contrasts proceeds from acoustically-simple phonemic contrasts which have most of their acoustic patterns in the low-frequency range, to the more acoustically complex, high-frequency based contrasts and (2) that the perceptual importance given to specific acoustic cues differs in children and adults, with children more sensitive to dynamic cues (e.g. formant transitions).

Existing developmental studies on categorical labelling in children do not provide information on the relative difficulty of different contrasts as they have typically assessed the development of a single phonemic contrast (e.g., voicing in initial plosives) in different age groups. Studies of cue-weighting in children have also usually been limited to a single phonemic contrast. The model of order of acquisition described above has therefore not been tested experimentally in a systematic fashion. Further insight into the development of children’s ability to use acoustic cues in their perception of phonemic contrasts is to be obtained using a technique known as ‘speech pattern audiometry’. This study will be the first to evaluate development in children aged 5 to 12 across a range of phonemic contrasts of differing acoustic complexity.

The proposed work is timely because: (1) speech pattern audiometry is beginning to be used clinically with children with hearing impairment and language delay, and a better mapping of the development in the ability to use acoustic cues is important for normative purposes and (2) speech pattern processing is used both in the design of speech pattern hearing aids and visual displays, and in the development of techniques designed to enhance speech in order to make it more discriminable. A better understanding of which speech pattern cues are salient for children is important for the future development of computer-based training using enhanced speech.

Objectives

Our main research questions are:

2. Does experimental data with a wide range of phonemic contrasts support the view that children give greater perceptual weighting to dynamic than to static acoustic cues?

Progress

The project started in August 1997. Testing is currently under way in a number of schools around London.

Related Issues

If you would like to know more about the project, please contact Valerie Hazan. .

^{Author: Valerie Hazan . Last Changed: 30 October 1997}