Impact of SNR, masker type and hearing aid signal processing on cognitive processing effort as indicated by the pupil dilation

The pupil dilation response reflecting cognitive processing effort has been shown to be sensitive to changes in signal-to-noise ratio (SNR) and masker type during sentence recognition in background noise. Our recent research showed that hearing-impaired listeners appear to benefit from a ‘multi-channel enhancement’ (MCE) algorithm in hearing aids, as smaller pupil responses were shown at 50% and 95% correct sentence recognition performance in a 4-talker babble masker for active versus inactive MCE processing.

The objective of this study was to measure the effect of MCE on the pupil dilation response and on speech recognition across a wide range of SNRs for two masker types. For 24 hearing-impaired listeners, the pupil dilation response was recorded during sentence recognition, in a design with 8 SNRs x 2 masker types (4-talker babble vs. stationary noise) x 2 algorithm modes (MCE active vs. inactive).

Based on our recent research we hypothesized an inverse u-shaped function of the pupil response across SNRs with relatively small pupil dilations at very low and very high SNRs. In the mid-range of SNRs, we expected improved sentence recognition and decreased cognitive processing effort for active MCE compared to inactive MCE.

We will present the results of this experiment.