Differences between children and adults in the neural encoding of voice fundamental frequency in the presence of noise and reverberation.

Abstract

Environmental noise and reverberation challenge speech understanding more significantly in children than in adults. However, the neural/sensory basis for the difference is poorly understood. We evaluated the impact of noise and reverberation on the neural processing of the fundamental frequency of voice (f0 )-an important cue to tag or recognize a speaker. In a group of 39 6-15-year-old children and 26 adults with normal hearing, envelope following responses (EFRs) were elicited by a male-spoken/i/in quiet, noise, reverberation, and both noise and reverberation. Due to increased resolvability of harmonics at lower than higher vowel formants that may affect susceptibility to noise and/or reverberation, the/i/was modified to elicit two EFRs: one initiated by the low frequency first formant (F1) and the other initiated by mid to high frequency second and higher formants (F2+) with predominantly resolved and unresolved harmonics, respectively. F1 EFRs were more susceptible to noise whereas F2+ EFRs were more susceptible to reverberation. Reverberation resulted in greater attenuation of F1 EFRs in adults than children, and greater attenuation of F2+ EFRs in older than younger children. Reduced modulation depth caused by reverberation and noise explained changes in F2+ EFRs but was not the primary determinant for F1 EFRs. Experimental data paralleled modelled EFRs, especially for F1. Together, data suggest that noise or reverberation influences the robustness of f0 encoding depending on the resolvability of vowel harmonics, and that maturation of processing temporal/envelope information of voice is delayed in reverberation, particularly for low frequency stimuli.