11th Speech in Noise Workshop, 10-11 January 2019, Ghent, BE

Neural processing of speech in children with sensorineural hearing loss

Axelle Calcus(a), Stuart Rosen, Lorna Halliday
University College London

(a) Presenting

Previous research has shown that even a mild (21-40 dB HL) or moderate (41-70 dB HL) sensorineural hearing loss (MMHL) can impair cortical processing of speech sounds, as evidenced by differences in event-related potential responses (P1-N1-P2-N2 and MMN) between children with MMHL and chronological age-matched normally hearing (NH) controls (Koravand et al., 2013). However, to date no studies have examined speech processing at the subcortical level in children with MMHL. Moreover, the effects of amplification on the neural encoding of speech are still poorly understood, with previous data suggesting a significant benefit at the subcortical (Anderson et al., 2013) but not the cortical level (Billings et al., 2007).

The aims of this project were to (i) investigate the cortical and subcortical processing of speech sounds in children with MMHL and (ii) evaluate the effects of amplification on the neural processing of speech in this group. To do so, cortical and subcortical EEG activity evoked by speech stimuli (/ba/-/da/) were simultaneously recorded in 18, 8- to 16-year-old children with MMHL and 16 age-matched NH controls. Subcortical processing was assessed using the frequency following response (FFR), an EEG component evoked at the subcortical level which reflects the encoding of the fundamental frequency (F0) and first few harmonics of complex auditory signals such as speech. For the MMHL group, stimuli were presented both unamplified (70 dB SPL), and with a frequency specific gain (without compression) based on their individual audiograms.

Results revealed that children with MMHL had smaller cortical responses than NH controls in both unamplified and amplified conditions, and did not show an MMN. In contrast, at the subcortical level, they showed a smaller FFR than NH controls in the unamplified condition only. With simulated amplification, children with MMHL demonstrated an FFR that was comparable to that observed in NH controls. Our findings suggest that the neural processing of unamplified speech may be impaired at both subcortical and cortical levels in children with MMHL. However, consistent with previous studies in adults, amplification appears to benefit auditory processing at subcortical but not cortical levels in children with MMHL. This might be explained by increasing multi-sensory integration at successive levels of the auditory system: whereas the inferior colliculus processes unimodal information, the auditory cortex processes multimodal information. Alternatively, this could reflect the later maturation of the auditory cortex compared to the inferior colliculus. MMHL may have a bigger impact upon cortical than subcortical processing.

Last modified 2019-01-08 16:51:41