Discrimination and identification of vowels in hearing-impaired and masked normal-hearing listeners

Abstract

This study examined the effects of mild-to-moderate sloping sensorineural hearing loss on vowel discrimination and identification. Five young, hearing-impaired (YHI) adults listened to vowel stimuli in three conditions: (a) at a conversational level with flat frequency response; and in two gain conditions, (b) at a high-sound level with flat frequency response; and (c) with frequency shaped gain according to the subject’s hearing loss. Listeners discriminated changes in the synthetic vowels /I ε e æ ʌ/ when F1 or F2 varied, and later identified the five vowels in a closed-set task. The results indicated that neither gain condition restored performance to that of young, normal-hearing listeners and that the potential upward spread of masking in condition (b) did not affect performance. Young, normal-hearing (YNH) listeners were age matched to YHI subjects. Hearing thresholds for the YNH listener of each YHI–YNH pair were precisely matched to those of the YHI listener by masking with spectrally shaped noise. YNH listeners with ‘‘simulated-hearing loss’’ were tested on the same discrimination and identification tasks. This design yields performance comparisons between YHI–YNH pairs when conditions of audibility are similar. The results will explain whether differences in performance between YHI–YNH pairs may be due to audibility or pathology. [Supported by NIHDCD–02229.]