Association between neural processing and speech understanding in adverse acoustics in children

It is well-known that hearing loss exacerbates the difficulty children experience understanding speech in noise. While factors such as vocabulary and working memory have been recognized as contributors, the role of neural mechanisms in speech-in-noise difficulties remains less understood. This talk will present findings from three recent studies examining: (1) differences between adults and children in encoding the fundamental frequency of voice (f0) in the presence of noise and/or reverberation, (2) the association between f0 encoding and speech understanding difficulties in children, and (3) association between sensitivity to spectral modulations and speech understanding difficulties in aided children with hearing loss. Our results show that, in the presence of reverberation, children exhibit immature f0 encoding, characterized by enhancements rather than the expected attenuation at low frequencies observed in adults. Among children with hearing loss using hearing aids, more robust low-frequency f0 encoding and stronger cortical representation of spectral modulations in the temporal cortex were linked to better speech discrimination in adverse acoustic conditions. Together, these findings highlight candidate neural metrics that may help explain the variability in speech-in-noise performance among children with hearing loss.