Predicting Speech Recognition in Adults Receiving Cochlear Implants

预测接受人工耳蜗植入的成年人的语音识别能力

• 批准号: 10755567
• 负责人: Aaron C Moberly
• 金额: $ 38.63万
• 依托单位: VANDERBILT UNIVERSITY MEDICAL CENTER
• 依托单位国家: 美国
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-08-01 至 2026-05-31
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10755567
• 关键词: Predicting; Speech; Recognition; Adults; Receiving

Project Summary/Abstract: Acquired hearing loss is becoming increasingly common, especially with the growth of the aging population. For many individuals, hearing loss is severe enough to warrant cochlear implantation. While many adults with cochlear implants (CIs) understand speech well through their devices, enormous unexplained variability exists in speech recognition outcomes, as well as in the trajectory of speech recognition improvement (i.e., perceptual learning) after implantation. Unfortunately, because of this unexplained variability, clinicians are unable to predict how an individual will perform with a CI, to explain why someone has poor speech recognition with a CI, or to design appropriate rehabilitation strategies to help an individual with poor performance. Currently, only half of this outcome variability can be explained by traditional clinical measures. Recent studies have suggested that individual differences in speech recognition among CI users reflect variability in more basic sensory (“bottom-up”) and cognitive-linguistic (“top-down”) factors. These factors should explain additional variability in CI outcomes, and a comprehensive integrative model incorporating traditional clinical measures along with bottom-up and top-down factors is needed. The overall objective of the proposed project is to fill the gap in knowledge regarding the sources of unexplained variability in CI speech recognition outcomes and perceptual learning. Aim 1 will determine the degree to which pre- operative measures of bottom-up sensory functions and top-down cognitive-linguistic processes predict CI speech recognition outcomes. Aim 2 will investigate the degree to which additional objective bottom-up sensory processes, and especially their interactions with top-down factors, explain sentence recognition. Aim 3 will investigate the trajectories of perceptual learning in speech recognition that occur during the first two years of CI use. The findings from this research project will have important theoretical and clinical implications by comprehensively delineating the contributions and interactions of bottom-up and top-down factors as they relate to variability in recognition of speech, as well as helping clinicians to better predict pre-operatively and understand post-operatively the outcomes and time course of perceptual learning that occurs after implantation.

项目概要/摘要： 后天性听力损失越来越普遍，特别是随着人口老龄化的增长。 对于许多人来说，听力损失严重到足以保证人工耳蜗植入。虽然许多成年人 人工耳蜗（CI）通过他们的设备很好地理解语言， 存在于语音识别结果中，以及语音识别改进的轨迹中（即， 感知学习）植入后。不幸的是，由于这种无法解释的变异性，临床医生 无法预测一个人将如何与CI合作，以解释为什么有人说话不好 或设计适当的康复策略，以帮助个人与贫困 性能目前，只有一半的结果变异性可以通过传统的临床措施来解释。 最近的研究表明，CI用户之间语音识别的个体差异反映了 更基本的感觉（“自下而上”）和认知语言（“自上而下”）因素的变化。这些因素 应该解释CI结果的额外可变性，以及一个全面的综合模型， 需要传统的临床测量沿着自下而上和自上而下的因素。的总体目标 拟议的项目是填补关于CI中无法解释的变异性来源的知识空白 语音识别结果和感知学习。目标1将决定预... 自下而上的感觉功能和自上而下的认知语言过程的操作措施可预测CI 语音识别结果。目标2将调查额外目标自下而上 感觉过程，特别是它们与自上而下因素的相互作用，解释了句子识别。目的 3将研究语音识别中感知学习的轨迹， 多年的使用。该研究项目的结果将具有重要的理论和临床意义。 通过全面描述自下而上和自上而下的贡献和相互作用， 这些因素与语音识别的可变性有关，并帮助临床医生更好地预测 术前和术后了解知觉学习的结果和时间过程， 发生在植入后。