13th Speech in Noise Workshop, 20-21 January 2022, Virtual Conference 13th Speech in Noise Workshop, 20-21 January 2022, Virtual Conference

P26 Processing fast speech: Investigating challenges for children with hearing loss

Rosanne Abrahamse, Titia Benders, Katherine Demuth, Nan Xu Rattanasone
Department of Linguistics, Macquarie University, Sydney, Australia

(a) Presenting

Children with hearing loss (HL), wearing hearing devices, encounter spoken language at varying speaking rates in everyday life. Faster rates may contain altered or reduced acoustic cues, leaving them with less time to process language. Degraded hearing may also increase this challenge. While results of children with normal-hearing (NH) show that they can process fast and slow speaking rates with similar speed, it is unclear how children with HL might perform. The current study examines effects of speaking rate on sentence processing in children with HL vs. NH. We predicted that children with HL would process sentences more slowly than children with NH, especially at faster speaking rates. We also explored the effect of hearing characteristics (unilateral HL, bilateral cochlear implants (CIs), bilateral hearing aids (HAs) or bimodal fitting) on performance. Thirty-one children with NH (mean age = 10.0, s.d. 1.65) and 36 children with HL (mean age = 10.4, s.d. 1.6) participated in an Auditory Word Detection Task. They were instructed to press the spacebar when they heard a target word in sentences presented at naturally produced Normal (4.4 syllables/second) vs. Fast speaking rates (6.1 syll./s). Response time (RT) was taken as a measure of processing speed. Data were analysed using mixed-effects models. Model 1 included Age, Group and Rate as fixed factors. In Model 2, participants were divided according to hearing characteristics, comparing each sub-group against the NH group. Results of the first model showed that children with HL were slower to detect words than their NH peers. There was no effect of Rate, nor an interaction between Rate and Group. In the second model, although children with unilateral HL did not differ from NH children, children with bilateral CIs, bilateral HAs and bimodal fitting were significantly slower. No significant interactions with Rate were found. These results suggest that children with HL process sentences more slowly than their NH peers, an effect driven primarily by children with bilateral HL. However, contrary to our predictions, children with HL were not disproportionately affected by fast speech. This contrasts with reported delays in fast speech processing for elderly listeners with HL. Greater availability of cognitive resources might play a role in explaining these differences. Thus, while language processing ability of children with bilateral HL is not on par with NH peers, our results provide a positive outlook on their ability to deal with faster speaking rates in communication.

Last modified 2022-01-24 16:11:02