Objective measures of listening effort: effects of background noise and noise reduction

Abstract

Purpose: This work is aimed at addressing a seeming contradiction related to the use of noise-reduction (NR) algorithms in hearing aids. The problem is that although some listeners claim a subjective improvement from NR, it has not been shown to improve speech intelligibility, often even making it worse.

Method: To address this, the hypothesis tested here is that the positive effects of NR might be to reduce cognitive effort directed toward speech reception, making it available for other tasks. Normal-hearing individuals participated in 2 dual-task experiments, in which 1 task was to report sentences or words in noise set to various signal-to-noise ratios. Secondary tasks involved either holding words in short-term memory or responding in a complex visual reaction-time task.

Results: At low values of signal-to-noise ratio, although NR had no positive effect on speech reception thresholds, it led to better performance on the word-memory task and quicker responses in visual reaction times.

Conclusions: Results from both dual tasks support the hypothesis that NR reduces listening effort and frees up cognitive resources for other tasks. Future hearing aid research should incorporate objective measurements of cognitive benefits.