Pupil Response as an Indication of Effortful Listening: The Influence of Sentence Intelligibility

Abstract

The aim of this study was to evaluate the influence of sentence intelligibility on the pupil dilation response during listening. Task-induced pupil-dilation reflects explicit effortful processing load. Therefore, pupillometry can be used to examine the listening effort during speech perception in difficult listening conditions. We expected to find increasing pupil dilation as a function of decreasing speech intelligibility. Thirty-eight young participants (mean age = 23 yrs, SD = 3.2 yrs) with normal hearing were included. They performed three speech reception threshold (SRT) tests in which they listened to sentences in stationary noise. A one-up-one-down, two-up-one-down, or four-up-one-down adaptive procedure was applied, resulting in the correct rehearsal of 50, 71, or 84% of the sentences (SRT(50%), SRT(71%), and SRT(84%), respectively). We examined the peak dilation amplitude, the latency of the peak dilation amplitude, and the mean pupil dilation during the processing of the speech in each of these conditions. The peak dilation amplitude and mean pupil dilation were calculated relative to the baseline pupil diameter during listening to noise alone. For each SRT condition, participants rated the experienced listening effort and estimated their performance level. The signal to noise ratios (SNRs) in the SRT(50%), SRT(71%), and SRT(84%) conditions increased as a function of the speech intelligibility level. The subjective effort ratings decreased, and the estimated performance increased with increasing speech intelligibility level. Repeated measures analyses of variance indicated that peak dilation amplitude and mean pupil dilation were higher in the SRT(50%) condition as compared with the SRT(71%) and SRT(84%) conditions. The peak dilation amplitude, mean pupil dilation, and peak latency increased with decreasing SNR of the speech in noise, but no effect of noise level by itself on the baseline pupil diameter was observed. Irrespective of SNR, the pupil response was higher for incorrectly repeated sentences than for correctly repeated sentences. The analyses also indicated condition-order effects on the peak dilation amplitude and mean pupil dilatation: the pupil response was higher in the first SRT test than in the second and third tests. Within the first and third test, the baseline pupil diameter and the mean pupil dilation decreased as a function of the sentence number within the test. Spearman correlation coefficients showed no relations among the SNRs at the SRTs, subjective ratings, and the pupil response. The peak dilation amplitude, peak latency, and mean pupil dilation systematically increase with decreasing speech intelligibility. These results support that listening effort, as indicated by the pupil response, increases with decreasing speech intelligibility. This study indicates that pupillometry can be used to examine how listeners reach a certain performance level. Application of this technique to study listening effort can yield valuable insight into the processing resources required across listening conditions and into the factors related to interindividual differences in speech perception in noise.