Distorted tonotopy severely degrades neural representations of natural speech in noise following acoustic trauma

Abstract

Listeners with sensorineural hearing loss (SNHL) struggle to understand speech, especially in noise, despite audibility compensation. These real-world suprathreshold deficits are hypothesized to arise from degraded frequency tuning and reduced temporal-coding precision; however, peripheral neurophysiological studies testing these hypotheses have been largely limited to in-quiet artificial vowels. Here, we measured single auditory-nerve-fiber responses to a natural speech sentence in noise from anesthetized chinchillas with normal hearing (NH) or noise-induced hearing loss (NIHL). Our results demonstrate that temporal precision was not degraded, and broader tuning was not the major factor affecting peripheral coding of natural speech in noise. Rather, the loss of cochlear tonotopy, a hallmark of normal hearing, had the most significant effects (both on vowels and consonants). Because distorted tonotopy varies in degree across etiologies (e.g., noise exposure, age), these results have important implications for understanding and treating individual differences in speech perception for people suffering from SNHL.