Effects of Sensorineural Hearing Loss on Robust Speech Coding

描述(申请人提供)：诊断和治疗听力障碍的一个巨大挑战来自于这样一个事实，即听力损失程度相似的人可能具有不同的语音识别能力。先前的研究已经证实，常见形式的听力损失是由内毛细胞和外毛细胞损伤的混合体引起的。内毛细胞和外毛细胞以根本不同的方式对听力做出贡献的概念框架支持了这样一个普遍假设，即内毛细胞和外毛细胞功能障碍程度的不同导致了患者在言语感知方面的差异。最近的心理物理学研究表明，感音神经性听力损失的听者在言语感知中使用时间精细结构线索的能力降低。这些研究激起了一场关于时间编码在听力正常和受损中的作用的激烈辩论，并可能对提高助听器和人工耳蜗在噪声中恢复言语感知的能力具有重要意义。拟议的神经生理学实验将通过直接量化感觉神经丢失对听神经中时间编码的影响来提供有价值的数据。选择性内毛细胞或外毛细胞损伤的影响将使用耳毒性药物进行研究。噪音引起的听力损失将被用来研究更常见的混合毛细胞损伤情况。将使用组织病理学分析和功能反应测量来确定单个动物的毛细胞损伤的特征。具体目标1是量化选择性毛细胞损伤对纤维内和跨纤维时间编码的影响。创新的分析避免了以前在跨光纤时间编码研究中的实验限制，将被用于量化精细结构和包络编码以及行波延迟。初步数据支持我们的假设，即感觉神经损失对细微结构的跨纤维编码的影响比纤维内编码更大。具体目的2是确定感觉神经丢失是否影响对声码语音中的精细结构和包络线索的神经编码。听力正常和听力受损的听者在理解声码语音的能力上的差异被用来表明在使用TFS线索时存在知觉缺陷。这些感知结果的生理基础很难评估，因为窄带耳蜗滤波限制了在耳蜗输出端分离精细结构和包络的能力。神经互相关系数将直接量化感觉神经丢失对噪声中声码语音的精细结构和包络编码保真度的影响。模型支持精细结构和包络响应均发生显著退化的假设。具体目标3是量化感觉神经丢失对并发复音中基频的时间编码的影响。听力受损的听者表现出利用音调差异来区分两个相互竞争的说话者的能力降低。假设估计两个并发复音的基频的能力降低的主要原因是时间精细结构的损失，而不是未分辨的谐波的包络编码的退化。 公共卫生相关性：拟议工作的长期目标是更好地了解强健言语感知的生理基础，这具有重要的理论和临床意义。在拟议的实验中收集的数据将提供关于内耳损伤对知觉相关声音的神经编码的不同影响的基本知识。这些知识将有助于制定诊断和康复策略，以改善听力损失患者的日常生活。

DESCRIPTION (provided by applicant): A great challenge in diagnosing and treating hearing impairment comes from the fact that people with similar degrees of hearing loss may have different speech recognition abilities. Previous research has established that common forms of hearing loss arise from a mixture of inner- and outer-hair-cell damage. A conceptual framework that inner and outer hair cells contribute to hearing in fundamentally different ways motivates the general hypothesis that differences in the degree of inner- and outer-hair-cell dysfunction contribute to across-patient variability in speech perception. Recent psychophysical studies have suggested that listeners with sensorineural hearing loss have a reduced ability to use temporal fine-structure cues in speech perception. These studies have fueled an active debate about the role of temporal coding in normal and impaired hearing, and may have important implications for improving the ability of hearing aids and cochlear implants to restore speech perception in noise. The proposed neurophysiological experiments will provide valuable data by directly quantifying the effects of sensorineural loss on temporal coding in the auditory nerve. The effects of selective inner- or outer-hair-cell damage will be studied using ototoxic drugs. Noise-induced hearing loss will be used to study the more common case of mixed hair-cell damage. Histopathological analyses and functional response measures will be used to characterize hair-cell lesions in individual animals. Specific Aim 1 is to quantify the effects of selective hair-cell damage on within- and across-fiber temporal coding. Innovative analyses that avoid previous experimental limitations in the study of across-fiber temporal coding will be used to quantify fine-structure and envelope coding, as well as traveling-wave delays. Preliminary data support our hypothesis that sensorineural loss affects across-fiber coding of fine-structure more than within-fiber coding. Specific Aim 2 is to determine whether sensorineural loss affects neural coding of fine-structure and envelope cues in vocoded speech. Differences in the ability to understand vocoded speech between listeners with normal and impaired hearing have been used to suggest a perceptual deficit in the use of TFS cues. The physiological basis for these perceptual results is difficult to evaluate because narrowband cochlear filtering limits the ability to isolate fine-structure and envelope at the output of the cochlea. Neural cross-correlation coefficients will quantify directly the effects of sensorineural loss on the fidelity of fine-structure and envelope coding for vocoded speech in noise. Modeling supports the hypothesis that significant degradations occur in both fine-structure and envelope responses. Specific Aim 3 is to quantify the effects of sensorineural loss on temporal coding of fundamental frequency in concurrent complex tones. Listeners with hearing loss show a reduced ability to make use of voice-pitch differences to segregate two competing talkers. It is hypothesized that the ability to estimate the fundamental frequencies of two concurrent complex tones is degraded primarily due to the loss of temporal fine structure, rather than from degraded envelope coding of unresolved harmonics. PUBLIC HEALTH RELEVANCE: The long-term goal of the proposed work is to obtain a better understanding of the physiological bases for robust speech perception, which has important theoretical and clinical implications. The data collected in the proposed experiments will provide fundamental knowledge about the differential effects of inner ear damage on the neural coding of perceptually relevant sounds. This knowledge will benefit the development of diagnostic and rehabilitative strategies to improve the daily lives of people with hearing loss.