Brain Networks of Noun Generation

 描述（由申请人提供）：语音产生依赖于一个分布式网络，该网络被各种神经系统疾病破坏，包括创伤、中风、神经变性和肿瘤。尽管语言在人类行为中起着核心作用，但我们对语言障碍的理解和适当治疗策略的发展受到缺乏生物实例化模型的阻碍，无法解释最基本的语言过程之一-单词的产生。我们建议使用颅内脑电图（icEEG）记录生成一个基于网络的描述内区域（布罗卡区，颞极）以及区域间的动态，并使用这些数据来严格评估当前的神经语言模型和我们小组提出的模型。我们将使用一种新的直接皮层刺激方法来交叉验证这些发现。我们将收集高时空分辨率的数据集命名的对象，人和地方从一个大的队列的神经外科患者（n=70），并生成一个网络表示的视觉，额叶，岛叶和颞极区域之间的信息流在文字生产。我们假设，文字的产生依赖于本地皮层网络，再加上更分布式的语言网络内的远程交互。将使用多个分析管道（使用幅度包络相关性、相位锁定值、相位-幅度耦合和定向传递函数）来测试该假设。我们将实施计时控制的闭环皮层刺激-定时到局灶性高伽马波段活动的开始-以诱导关键词产生节点的短暂功能障碍，从而能够评估icEEG数据的因果关系。我们将利用我们的专业知识测量大脑中的感应电流传播，通过研究正在进行的网络过程对皮质-皮质诱发电位和高伽马活动的振幅和延迟的影响，进一步评估网络表征。总之，拟议的工作旨在调查的神经动力学命名在多个空间层次（区域和全球）使用各种网络分析方法。拟议的工作将提供独特的信息，文字产生的神经生物学与广泛的影响，正常语言的理解和临床管理的患者与命名障碍。

 DESCRIPTION (provided by applicant): Speech production relies upon a distributed network that is disrupted by a variety of neurological diseases including trauma, stroke, neuro-degeneration and neoplasms. Despite the central role of language in human behavior, our understanding of linguistic disorders and the development of appropriate treatment strategies is impeded by the absence of biologically instantiated models to explain one of the most basic of language processes - the production of words. We propose to use intracranial electroencephalographic (icEEG) recordings to generate a network-based description of the within region (Broca's area, insula and temporal pole) as well as inter-regional dynamics and use this data to critically evaluate current neuro-linguistic models and a model proposed by our group. We will cross-validate these findings using a novel direct cortical stimulation approach. We will collect high spatio-temporal resolution datasets during naming of objects, people and places from a large cohort of neurosurgical patients (n=70) and generate a network representation of information flow between visual, frontal, insular and temporal pole regions during word production. We hypothesize that word production relies on local cortical networks, coupled with long-range interactions within more distributed language network. This hypothesis will be tested with multiple analytic pipelines (using amplitude envelope correlations, phase locking values, phase-amplitude coupling and directed transfer functions). We will implement chronometrically controlled closed loop cortical stimulation - timed to the onset of focal high gamma band activity - to induce transient dysfunction of critical word production nodes enabling assessment of the causality of the icEEG data. We will leverage our expertise with measuring induced current spreads in the brain, by studying the effect of ongoing network processes on the amplitude and latency of cortico-cortical evoked potentials and high gamma activity to further assess network representations. In summary, the proposed work aims to investigate the neuro-dynamics of naming at multiple spatial levels (regional and global) using a variety of network analysis methods. The proposed work will provide unique information on the neurobiology of word production with broad implications for the understanding of normal language and for the clinical management of patients with anomia.