Audibility as a Predictor of Speech Recognition and Listening Effort

Authors

Ryan W. McCreery, University of Nebraska-LincolnFollow

Date of this Version

2011

Document Type

Article

Comments

A DISSERTATION Presented to the Faculty of The Graduate College at the University of Nebraska In Partial Fulfillment of Requirements For the Degree of Doctor of Philosophy, Major: Human Sciences, Under the Supervision of Professors Stephen J. Boney and Patricia G. Stelmachowicz. Lincoln, Nebraska: May, 2011
Copyright 2011 Ryan William McCreery

Abstract

Two studies were conducted to evaluate how audibility influences speech recognition and measures of working memory in children with normal hearing. Specifically, audibility limitations related to background noise and limited bandwidth were analyzed, as these factors are characteristic of the listening conditions encountered by children with hearing loss who wear hearing aids.

In the first study, speech recognition was measured for 117 children and 18 adults with normal hearing. Stimulus bandwidth and the level of background noise were varied systematically in order to evaluate predictions of audibility based on the Speech Intelligibility Index. Results suggested that children with normal hearing required greater audibility to reach the same level of speech understanding as normal-hearing adults. However, differences in performance between adults and children did not vary across frequency bands as anticipated.

In the second study, 18 children with normal hearing completed two tasks of working memory to examine how background noise and limited bandwidth might limit memory processes in children. In a non-word repetition task, significant reductions in speech recognition and increases in response time were observed for both the noise and limited bandwidth conditions. These results suggest that listening effort increased and phoneme recall decreased when the speech signal was degraded. For recall of real words, no differences in recognition were observed for two conditions with the same signal-to-noise ratio but differing bandwidths. However, recall was significantly inhibited in the limited bandwidth condition, supporting the hypothesis that a limited bandwidth may negatively impact working memory performance in children, even when recognition is preserved.

Collectively, these studies suggest that methods of calculating audibility based on adults are likely to be inadequate for predicting speech recognition and listening effort for children. Models of audibility that incorporate the linguistic and cognitive dynamics of children are necessary to maximize communication outcomes for children with hearing loss.