Vocal emotion recognition in children with cochlear implants and with hearing aids

Authors: Laura Rachman1, Leanne Nagels2, Etienne Gaudrain3, Francien Coster1, Bert Maat1, Rolien Free1, Deborah Vickers4, Petra Hendriks1, Marc van der Schroeff5, Jantien Vroegop5, Deniz Başkent1

1University Medical Center Groningen
2University of Groningen
3CNRS, Université Lyon
4University of Cambridge
5University Medical Center Rotterdam

Background: Correctly interpreting the emotions of another person is important in daily communication and this ability contributes to children’s social development. In children with normal hearing (NH), the development of vocal emotion recognition continues throughout childhood. Children with hearing loss may have difficulties perceiving relevant acoustic cues conveying emotions, but combined effects of neuroplasticity, physiological effects from hearing loss, and the compensatory features of hearing aids (HAs) and cochlear implants (CIs) are not fully understood. It is therefore unclear if difficulties in vocal emotion recognition are primarily due to a reduced access to relevant acoustic cues, an acute effect. In addition, the ability to recognize and label emotions may develop differently in HA or CI children, an accumulated effect over a longer period of time that may be less directly related to hearing difficulties.

Methods: We assessed vocal emotion recognition in NH, HA, and CI children (6-18 yr). Participants were presented with semantically meaningless sentences, uttered with a happy, sad, or angry intonation. Recognition accuracy in all three groups was analysed using a quantile regression analysis and generalized additive models to assess nonlinear effects of age.

Results: All three groups showed a significant age effect, with accuracy increasing as a function of age. Individual data indicated a number of HA and CI children showed age-expected accuracy scores. At the group level, accuracy scores of HA and CI children did not differ significantly from each other, but both groups had significantly lower accuracy scores compared to NH children.

Conclusions: Based on these results, introducing a child-appropriate and validated vocal emotion diagnostic test in standard clinical practice could be helpful to identify those HA or CI children with particular difficulties in vocal emotion recognition and provide them with additional and specialized rehabilitation.