NEURAL SYSTEMS SUPPORTING SPEECH PROCESSING IN LISTENERS WITH COCHLEAR IMPLANTS

摘要 大约有3600万美国人报告有一定程度的听力障碍。听力障碍是 与社交孤立、抑郁和认知能力下降有关。听力障碍的代价不仅是 个人，但经济：对于有听力障碍的美国人来说，一生的成本估计为4.6美元 10亿美元，主要是由于工作生产率下降。听力障碍的听众通常可以理解 口语，但随着努力的增加，从其他过程中带走认知资源， 注意力和记忆力。因此，一个重要的挑战是了解大脑如何应对退化的 语音信号和认知过程是成功理解的关键。成人听众 人工耳蜗植入者是研究努力倾听的独特群体：他们通常已经适应了 听觉剥夺多年的深度听力损失，然后是一定程度的听力 植入后修复。随着人工耳蜗植入引起的听觉输入增加， 个人听众能够成功地识别语音，特别是在存在 背景噪音是非常多变的。以前试图解释这种变化的背景下， 大脑活动的潜在模式一直不成功，在很大程度上是因为技术挑战， 在植入医疗器械的情况下与神经成像相关的疾病已经阻止了充分的 对语言的神经反应的定位。我们研究的目标是了解认知系统 支持耳蜗植入者的语音识别，并使用有关这些系统的知识， 来改善行为结果。我们这样做是使用来自行为测量的汇聚证据， 脑功能成像我们利用高密度扩散光学断层扫描（HD-DOT），一种光学成像的形式。 脑成像，其产生局部皮层活动的解剖学定位指数。我们将绘制出 支持耳蜗植入者语音理解的系统，我们希望与 有良好听力的听众参与的活动。然后，我们将评估神经标志物的程度， 努力倾听可以预测在存在语音识别成功的个体差异， 背景噪音这些发现将有助于我们理解人工耳蜗植入辅助语言 在神经解剖学约束的框架中识别，并开发更准确的结果测量。

Abstract Approximately 36 million Americans report having some degree of hearing impairment. Hearing impairment is associated with social isolation, depression, and cognitive decline. The toll of hearing impairment is not only personal, but economic: For Americans who have hearing impairment, the lifetime cost is estimated to be $4.6 billion, mostly due to reduced work productivity. Listeners with hearing impairment can often understand spoken language, but with increased effort, taking cognitive resources away from other processes such as attention and memory. An important challenge is therefore to understand how the brain copes with a degraded speech signal and the cognitive processes that are most critical to successful comprehension. Adult listeners with cochlear implants are a unique group in which to investigate effortful listening: They have typically adapted to auditory deprivation for a period of years of profound hearing loss, followed by some degree of hearing restoration following implantation. Following increased auditory input due to cochlear implantation, the degree to which individual listeners are able to successfully recognize speech, especially in the presence of background noise, is extremely variable. Previous attempts to explain this variability in the context of underlying patterns of brain activity have been unsuccessful, in large part because the technical challenges associated with neuroimaging in the presence of an implanted medical device have prevented adequate localization of neural responses to speech. The goal of our research is to understand the cognitive systems that support speech recognition in listeners with cochlear implants and to use knowledge about these systems to improve behavioral outcomes. We do so using converging evidence from behavioral measures and functional brain imaging. We make use of high-density diffuse optical tomography (HD-DOT), a form of optical brain imaging that produces anatomically-localized indices of regional cortical activity. We will map the neural systems supporting speech comprehension in listeners with cochlear implants, which we expect to differ from those engaged by listeners with good hearing. We will then evaluate the degree to which neural markers of effortful listening can predict individual differences in speech recognition success in the presence of background noise. Together the findings will help ground our understanding of cochlear implant-aided speech recognition in a neuroanatomically-constrained framework and develop more accurate outcome measures.