Flexible representation of speech in human auditory cortex

人类听觉皮层中语音的灵活表示

基本信息

批准号：
10605606
负责人：
Vibha Viswanathan
金额：
$ 3.35万
依托单位：
UNIVERSITY OF PITTSBURGH AT PITTSBURGH
依托单位国家：
美国
项目类别：
财政年份：
2022
资助国家：
美国
起止时间：
2022-12-01 至 2023-04-30
项目状态：
已结题

来源：
https://reporter.nih.gov/project-details/10605606
关键词：
Accent Acoustics Address Area Auditory Auditory area Behavior Control Behavioral Behavioral Research Brain Categories Clinical Code Cognitive Communication impairment Complement Cues Data Dimensions Doctor of Philosophy Effectiveness Electrocorticogram Electroencephalography Electrophysiology (science)Engineering Environment Frequencies General Population Goals Hearing Human Intervention Laboratories Language Linguistics Link Maps Measurable Measures Modeling Neurobiology Neurologic Neurosciences Noise Pear Population Psychophysics Research Research Training Scalp structure Schizophrenia Sensory Signal Transduction Source Speech Speech Acoustics Speech Perception Stimulus Structure of superior temporal sulcus Superior temporal gyrus Surface System Temporal Lobe Testing Theoretical model Time Training Training Support Translating Ursidae Family Variant Weight Work autism spectrum disorder career clear speech clinical application cognitive neuroscience cohort contextual factors flexibility gray matter hearing impairment improved innovation insight nervous system disorder neural information processing neuromechanism novel rehabilitation strategy relating to nervous system response speech in noise speech processing speech recognition translational health science

项目摘要

PROJECT SUMMARY/ABSTRACT The mapping from acoustics to linguistically relevant speech categories is not fixed; rather, speech recognition adapts to listening context. For example, the utterances bear and pear differ by as many as 16 spectrotemporal acoustic dimensions. The relative importance (i.e., the perceptual weight) of each of these acoustic dimensions in signaling bear vs. pear changes with contextual factors like whether the listening environment is quiet or noisy and whether the talker speaks in a native or unfamiliar accent. This flexibility or adaptive plasticity is an important component of how the brain maintains robust speech perception despite considerable variation in speech acoustics. Indeed, inflexible sensory processing is implicated as a source of perceptual deficits in both hearing impairment and neurological disorders. Yet, the neural mechanisms underlying adaptive plasticity in speech perception are poorly understood. To address this gap, the proposed research will leverage access deep within human auditory cortex obtained through intracerebral stereotactic electroencephalography (sEEG), by employing sEEG simultaneously with scalp electroencephalography (EEG) and well-established behavioral tasks for measuring adaptive plasticity. Aim 1 will invoke adaptive plasticity in speech perception with acoustic noise or a change in short-term input distributions mimicking an ‘accent’ and will relate behavioral changes in the perceptual weights of acoustic dimensions relative to baseline (i.e., clear speech) to changes in the neural response profile across the auditory cortical hierarchy. Aim 2 will quantify the relationships between sEEG and EEG to establish the extent to which intracerebral signatures of adaptive plasticity relate to scalp EEG signatures measurable in the general population; this will also facilitate region-specific interpretation of EEG. The work will provide novel insight into the cortical mechanisms of adaptive plasticity in speech perception, with implications for neurobiological models and clinical applications. Our innovative combination of simultaneous scalp EEG with spatially specific, high signal-to-noise ratio sEEG will create a highly informative link across noninvasive and intracranial electrophysiology. Project completion also will provide the applicant with training in cognitive neuroscience of speech perception, intracerebral electrophysiology, and approaches to effectively integrate intracerebral and scalp EEG. This training will complement her strong engineering/quantitative background and Ph.D. training in auditory temporal coding, scene analysis, and EEG. This will advance her goal of undertaking a successful independent academic research career in the neuroscience of human audition and speech perception.

项目概要/摘要从声学到语言相关语音类别的映射不是固定的；而是语音识别。适应听力环境，例如，“bear”和“pear”的话语相差多达 16 个时域。每个声学维度的相对重要性（即感知权重）在发出熊与梨的信号时，会根据上下文因素（例如聆听环境是否安静或安静）而变化嘈杂的声音以及说话者是否用本地口音或不熟悉的口音说话，这种灵活性或适应性可塑性是一种。尽管语音感知存在很大差异，但大脑如何保持强大的语音感知的重要组成部分事实上，不灵活的感觉处理是两者感知缺陷的根源。然而，听力障碍和神经系统疾病的适应性可塑性背后的神经机制。为了解决这一差距，拟议的研究将利用访问权限。通过脑内立体定向脑电图（sEEG）获得人类听觉皮层深处，通过同时使用 sEEG 和头皮脑电图 (EEG) 以及成熟的行为学测量自适应可塑性的任务将调用声学语音感知中的自适应可塑性。噪声或短期输入分布的变化模仿“口音”，并将与行为变化相关声学相对维度相对于基线（即清晰语音）的感知权重对神经网络变化的影响目标 2 将量化 sEEG 和听觉皮层层次之间的关系。脑电图确定适应性可塑性的脑内特征与头皮脑电图的相关程度在一般人群中可测量的特征；这也将有助于脑电图的特定区域解释。这项工作将为言语感知中适应性可塑性的皮质机制提供新的见解，我们的创新组合对神经生物学模型和临床应用具有影响。具有空间特异性、高信噪比 sEEG 的同步头皮脑电图将创建信息丰富的数据项目完成后还将为申请人提供跨非侵入性和颅内电生理学的联系。接受过言语感知、脑内电生理学和方法的认知神经科学培训有效整合脑内和头皮脑电图这种训练将补充她的强大能力。听觉时间编码、场景分析和脑电图方面的工程/定量背景和博士学位培训。这将推进她在该领域开展成功的独立学术研究生涯的目标人类听觉和言语感知的神经科学。