Effects of Temporal Processing on Speech-in-Noise Perception in Middle-Aged Adults

Abstract

Simple Summary

A significant portion of adults with clinically normal hearing sensitivity have difficulty understanding speech in background noise. Current clinical assessments fail to explain this phenomenon, prompting the exploration of auditory mechanisms beyond those covered by routine clinical testing. One mechanism important for separating sound sources—a key task for understanding speech-in-noise—is temporal processing, or the extraction and organization of acoustic timing characteristics. Here, we investigate the hypothesis that deficits in temporal processing contribute to difficulties in understanding speech-in-noise. We explore this in middle-aged adults—an under-investigated group, despite their high prevalence of speech-in-noise difficulties. In this study, we found that differences in speech-in-noise abilities were associated with deficits in two aspects of temporal processing: the neural encoding of periodic speech features, such as pitch, and perceptual sensitivity to rapid acoustic timing differences between ears. Interestingly, the use of these mechanisms was task-dependent, suggesting various aspects of temporal processing differentially contribute to speech-in-noise perception based on the characteristics of the listening environment. These findings contribute to our overall understanding of which auditory mechanisms play a role in speech-in-noise difficulties in normal hearing listeners, and can inform future clinical practice to serve this population.

Abstract

Auditory temporal processing is a vital component of auditory stream segregation, or the process in which complex sounds are separated and organized into perceptually meaningful objects. Temporal processing can degrade prior to hearing loss, and is suggested to be a contributing factor to difficulties with speech-in-noise perception in normal-hearing listeners. The current study tested this hypothesis in middle-aged adults—an under-investigated cohort, despite being the age group where speech-in-noise difficulties are first reported. In 76 participants, three mechanisms of temporal processing were measured: peripheral auditory nerve function using electrocochleography, subcortical encoding of periodic speech cues (i.e., fundamental frequency; F0) using the frequency following response, and binaural sensitivity to temporal fine structure (TFS) using a dichotic frequency modulation detection task. Two measures of speech-in-noise perception were administered to explore how contributions of temporal processing may be mediated by different sensory demands present in the speech perception task. This study supported the hypothesis that temporal coding deficits contribute to speech-in-noise difficulties in middle-aged listeners. Poorer speech-in-noise perception was associated with weaker subcortical F0 encoding and binaural TFS sensitivity, but in different contexts, highlighting that diverse aspects of temporal processing are differentially utilized based on speech-in-noise task characteristics.