Analytic and divided listening in normal-hearing and hearing-impaired listeners measured in a nonspeech pattern identification task

Abstract

In multisource listening environments, it is important to be able to attend to a single sound source (analytic listening) while concurrently monitoring unattended sources for potentially useful information (divided listening). Previous studies have indicated that hearing-impaired (HI) listeners have more difficulty with analytic listening than do normal-hearing (NH) listeners. Although central factors (e.g., selective attention and memory) clearly play a role, the extent to which differences in peripheral factors (e.g., auditory filter characteristics) contribute to this effect is not clear. In this study, performance in a closed-set nonspeech tonal pattern identification task was measured in NH and HI listeners for patterns centered at 850 and 3500 Hz. The frequency spacing among the tones forming the patterns was adjusted to equate performance across listeners at each center frequency separately to control for peripheral frequency resolution. Patterns were then played at both frequencies concurrently. Listeners were instructed to attend to either the low or high frequency and identify the pattern. In a second condition, patterns were randomly presented at one frequency with a foil at the other frequency, requiring the listener to monitor both frequencies to identify the pattern. Preliminary findings suggest that peripheral and central factors contributed to performance. [Support: NIH-NIDCD.]