Using spectral resolution to explain performance variability with severe loss

描述（由申请人提供）：听力损失影响着美国5500万人。患有严重听力损失的听众经历独特的语音感知问题，其与差的频谱分辨率、沿着降低的可听度和降低的动态范围相关联。对于重度听力损失患者的相对贡献知之甚少。几项研究报告了严重听力损失听众的康复成功率。以前的工作已经确定了一个子集的严重损失的听力改善并没有提高辅音识别分数的听众，在相同的方式与他们的同龄人具有相同的阈值。量化频谱分辨率和语音识别之间的关系可以帮助1）指导临床决策，为其听力仪器设置压缩参数。2）。它可以解释经常报告的差异，分数的非线性放大之间的听众显然是平等的严重阈值。拟议项目的具体目标是：1）评估评估光谱分辨率的不同方法。2)探讨频谱分辨率与辅音识别的关系。3)研究幅度压缩语音的频谱分辨率与性能的关系。患有严重感音神经性听力损失的成年听众将参与这项研究。每例受试者将接受标准听力学和认知测试，以确定研究的候选人资格。光谱分辨率和辅音识别将使用两种不同的方法和测试刺激来测量。将评估频谱分辨率的诊断价值，用于1）解释具有相似阈值的收听者之间的性能差异。2)确定个体收听者对由振幅压缩导致的失真的敏感性。从这项研究中获得的信息可以用于未来开发助听器放大和信号处理策略，为严重听力损失的听众。无法检测频谱线索的听众可能会受益于更复杂的信号处理，如频谱增强。这些信息也将有助于为严重听力损失的听众定制康复计划。

DESCRIPTION (provided by applicant): Hearing loss affects 55 million individuals in the United States. Listeners with severe hearing loss experience unique speech perception problems that are associated with poor spectral resolution, along with reduced audibility, and reduced dynamic range. Little is known about the relative contribution of each for people with severe hearing loss. Several studies have reported varied rehabilitation success rate among listeners with severe hearing loss. Previous work has identified a subset of listeners with severe loss for whom improving audibility did not improve consonant recognition scores, in the same manner as their cohort with equivalent thresholds. Quantifying the relationship between spectral resolution and speech recognition could help 1) guide clinical decisions on setting compression parameters for their hearing instruments. 2). It could explain the often reported variance in scores for nonlinear amplification among listeners with apparently equal severe thresholds. The specific aims of the proposed project are to 1) to evaluate different methods of assessing spectral resolution. 2) To evaluate the relationship between spectral resolution and consonant recognition. 3) To evaluate the relationship between spectral resolution and performance for amplitude compressed speech. Adult listeners with severe sensorineural hearing loss will participate in this study. Each subject will undergo standard audiologic and cognitive testing to determine candidacy for the study. Spectral resolution and consonant recognition will be measured using two different methods and test stimuli. The diagnostic value of spectral resolution will be evaluated for 1) explaining performance variance across listeners with similar thresholds. 2) determining individual listener's susceptibility to distortion resulting from amplitude compression. The information gained from this study can be used in future development of hearing aid amplification and signal processing strategies for listeners with severe hearing loss. Listeners who are unable to detect spectral cues might benefit from more sophisticated signal processing, such as spectral enhancement. This information will also be useful for customizing rehabilitative programs for listeners with severe hearing loss.