Perceptual Effects of Noise Reduction With Respect to Personal Preference, Speech Intelligibility, and Listening Effort

Objectives: Most modern hearing aids use noise reduction to increase listening comfort in noisy environments. However, it is unclear whether perceptual effects (e.g., intelligibility, listening effort, and preference) of noise reduction differ among hearing aids and among listeners. The authors compared perceptual scores across different hearing aid noise-reduction systems to determine (1) whether noise-reduction systems differ perceptually and (2) which factors underlie the overall preference of individual listeners. Design: The authors recorded hearing aid noise-reduction outputs and used these signals in a laboratory experiment. The recording method allowed the evaluation of noise reduction in an isolated form, without the dominant effects of hearing aid frequency response and interactions with dynamic-range compression. Ten normal-hearing subjects listened to speech in babble noise processed by noise reduction from four different hearing aids. The subjects performed (1) speech-intelligibility tests, (2) listening-effort ratings, and (3) paired-comparison ratings with respect to noise annoyance, speech naturalness, and overall preference. Results: Noise-reduction systems from different hearing aids differed in the degree to which they influenced the noise annoyance and speech naturalness perceived by the normal-hearing listeners. Small differences in intelligibility and effort scores were found among different noise-reduction systems but not between having noise reduction on and off. Subjects differed in whether their overall preference was more strongly related to noise annoyance or to speech naturalness. Conclusions: The authors conclude that noise annoyance and speech naturalness are determining factors for the overall preference of normal-hearing listeners for a specific noise-reduction condition, and found individual differences in the preferred weighting of these factors even in a homogeneous group of normal-hearing listeners. Subsequent experiments should include hearing-impaired subjects to determine whether these conclusions also hold for a more heterogeneous group of listeners. If these results can be extrapolated to hearing-impaired listeners, the fitting and fine-tuning of noise reduction in hearing aids needs considerable revision.