Electrophysiological Approaches to Understanding Functional Organization of Speech in the Brain

摘要 为了理解语音，人脑必须将嘈杂的声音信号解析并转换为有意义的 语言内容，包括音素、音节、单词和句子。这涉及到确定 重要的声学事件，如句子或短语的开始。在检测到这些起点之后， 必须确定句子的内容。后颞上回(PSTG)--包括 经典的“韦尼克氏区”--对这一过程至关重要，但直到最近，人们对其功能知之甚少。 组织，特别是这个职能组织在整个发展过程中如何变化。我们最近 研究表明，pSTG的一个空间离散区域对于指示句子或短语何时出现至关重要 开始。该区域在空间和功能上与编码的更前面的“持续”区域截然不同 贯穿整个句子的语音特征信息。在功能上，后发性和前部性均持续 地区分别表现出较短的和较长的时间整合时间，这表明在 自然语音处理。在这里，我们提出了一种创新的方法，使用罕见的数据集，其中神经活动 直接从儿科患者的人类听觉皮质和语言相关区域记录下来 正在接受癫痫手术的临床评估。该方法克服了空间分辨率和时间分辨率的不足。 其他非侵入性手术的局限性，为了解人类听觉功能提供了一个难得的窗口 大脑皮层。这项拟议的研究将使用高分辨率的颅内记录来研究大脑是如何检测到 自然语音混合中的声学起始点，以及神经生理学对这些声音的反应 从儿童早期到青春期的变化。此外，我们将调查这些对onset的反应 受到环境的调节，包括在注意期间和自我产生的声音。除了提供 洞察人类听觉皮质的基本功能组织和听觉的皮质机制 场景分析，本研究对语音脑计算机的开发具有重要意义 界面。我们的结果也可以告诉我们，语音和语言是如何在自然环境中被处理的，这有 对发育性语言障碍、听觉处理障碍、阅读障碍、 自闭症和失语症。

ABSTRACT To understand speech, the human brain must parse and transform a noisy acoustic signal into meaningful linguistic content, including phonemes, syllables, words, and sentences. This involves determining the timing of important acoustic events, such as the onset of a sentence or a phrase. Following detection of these onsets, the content of the sentence must be determined. The posterior superior temporal gyrus (pSTG)—including the classic “Wernicke’s area”—is critical to this process, but until recently, little was known about its functional organization, and in particular how this functional organization changes throughout development. Our recent work showed that a spatially discrete region of the pSTG is critical for indicating when a sentence or phrase begins. This region is distinct spatially and functionally from more anterior “sustained” areas that encode phonetic feature information throughout a sentence. Functionally, both posterior onset and anterior sustained regions show short and long temporal integration times, respectively, suggesting complimentary roles in natural speech processing. Here, we propose an innovative approach using rare datasets where neural activity is recorded directly from the human auditory cortex and speech-related areas in pediatric patient participants undergoing clinical evaluation for epilepsy surgery. This method overcomes the spatial and temporal resolution limitations of other noninvasive procedures, and provides a rare window into the function of the human auditory cortex. The proposed study will use high resolution intracranial recordings to investigate how the brain detects acoustic onsets in natural speech sound mixtures, and how neurophysiological responses to these sounds change from early childhood to adolescence. Furthermore, we will investigate how these responses to onsets are modulated by context, including during attention and for self-generated sounds. In addition to providing insight into the basic functional organization of the human auditory cortex and cortical mechanisms for auditory scene analysis, this research has important implications for the development of a speech brain computer interface. Our results could also inform how speech and language are processed in natural contexts, which has implications for the treatment of developmental language disorders, auditory processing disorder, dyslexia, autism, and aphasia.