Perception of speech in context by listeners with healthy and impaired hearing

点击翻译按钮获取中文摘要

Project Summary / Abstract Perceptual systems operate in context: perception of a given object/event is influenced by the objects/events that surround it. Speech perception operates in the same way, where perception of speech sounds is heavily influenced by the acoustic properties of contextual sounds that surround them. These acoustic context effects are critically important to perception because they help maintain perceptual constancy in the face of considerable variability in speech acoustics and listening environments. Despite this importance, there is a poor understanding of how acoustic context shapes speech perception. The well-established ways of measuring speech perception routinely present stimuli without any context, or when they do, it is to study a single context effect in quiet for listeners with normal hearing (NH). These approaches severely impede understanding of speech perception in everyday listening conditions (in context, amidst background noise) and for listeners with hearing difficulties. The long-term goal of this project is to improve our understanding of the mechanisms underlying these acoustic context effects in speech perception in normal and impaired auditory systems. This goal is addressed through three specific aims. The first aim is to define the unifying relationships between different context effects in speech perception by NH and hearing-impaired (HI) populations. Experiments in the first aim utilize an individual differences approach to explore auditory context effects as a unified system rather than individual isolated effects. The second aim is to study the impact of background noise on these context effects for NH and HI populations. Experiments in the second aim will test competing predictions regarding whether noise diminishes the influence of context via masking or enhances the influence of context via streaming. The third aim is to investigate context effects in speech perception by cochlear implant (CI) users. Experiments in the third aim with real- and simulated-CI listening will elucidate mechanisms underlying context effects and the impacts of different aspects of CI processing. These aims are supplemented by computational modeling of neural responses to the speech stimuli in order to clarify peripheral versus central mechanisms underlying these context effects. Overall, the results will provide new insights into how NH, HI, and CI populations perceive sounds in the context of other sounds and in background noise. These efforts will produce better understanding of auditory perception with the realistic and prominent constraint of contextual influence, and will help to provide a neural and behavioral framework for translational applications.