Pupillary correlates of auditory emotion recognition in older hearing-aid users

Hearing-impaired (HI) individuals are known to have difficulties in auditory emotion recognition tasks compared to normal hearing individuals. It is still unclear if this is due to difficulties in the lower levels of auditory processing or to the categorisation of emotions that is involved in the experimental task (Picou et al., 2018, Tr. Hear. 22:4). An objective index of emotion recognition can be observed in pupil dilations, which have recently been shown to dilate more for emotionally meaningful speech in comparison to emotionally neutral speech (Jürgens, Fischer and Schacht, 2018, Front. Psychol 9:13). In this study, we investigated pupil dilations as a measure of emotion recognition in an older HI population, all users of hearing aids, and by correlating this measure with behavioural responses we aimed to gain insight into the potential difficulties that this population faces with an emotion recognition task. These difficulties may arise from this population being unable to hear the signal clearly, or due to the effect of cognitive involvement in the experimental task. We further hypothesised that high-frequency information is important for emotion recognition, so we tested our participants both with and without a frequency lowering enabled hearing aid.

We fitted 8 older HI participants, who had moderate to severe sloping high-frequency hearing loss, with frequency lowering enabled hearing aids for an acclimatisation period of 3-6 weeks. We recorded their pupil dilations in response to emotional speech with and without frequency lowering, during both a passive and active-listening condition. The active condition included a behavioural emotion identification task, where participants were given a forced-choice task after each stimulus was presented.

Preliminary results indicate that the pupillary response of the passive-listening condition reveals that emotional arousal can be elicited in an older HI population, and even when only passively listening to emotional speech. Functional differences are seen between the passive and active listening conditions, indicating that the cognitive involvement elicited by the emotion categorisation task is reflected by the pupillary responses. Frequency lowering does not provide benefits on emotion categorisation at the group level, but may have benefits for individual participants.