Exploring the Neural Dynamics of Cognition through Human Electrocorticography

为了揭示控制认知的神经机制，我们必须了解不同的大脑区域是如何在几十到几百毫秒的时间尺度上相互作用的。在毫秒范围内的时间分辨率很难通过成像技术如功能磁共振成像来实现。另一方面，在灵长类动物中使用的电生理技术提供了高时间分辨率，但一次只能记录一个或最多几十个神经元。另一种可以克服许多这些问题的记录技术是脑皮质电图（ECoG），在这种技术中，为了在手术前评估大脑而植入的一系列电极被用来记录人体表面的电波动。ECoG允许同时测量来自几个不同大脑区域的电信号，同时提供毫秒级的时间分辨率。因此，它非常适合探索认知的神经动力学。在美国国家科学基金会的支持下，华盛顿大学的Rajesh Rao和Jeff Ojemann将通过三个任务来研究大脑皮层的动态，这三个任务代表了认知中三个不同的抽象层次：运动任务、工作记忆任务和语言任务。它将使用传统的频谱技术以及更复杂的统计技术，如用于源分离的独立分量分析（ICA）和用于信号估计的贝叶斯技术来分析数据。以此为基础，他们将构建神经元网络的生物物理模型，并模拟皮质-皮质和皮质-丘脑反馈回路，以了解ECoG的神经发生。如果成功，这项研究将使人们对大脑功能有一个新的认识，从长远来看，可能会导致运动控制、记忆或语言处理等认知缺陷的治疗。

To unravel the neural mechanisms governing cognition, one must understand how different brain areas interact with one another on time scales that range from tens to several hundreds of milliseconds. Temporal resolution in the range of milliseconds is hard to achieve through imaging techniques such as fMRI. On the other hand, electrophysiological techniques used in primates provide high temporal resolution but only record from a single or at most a few tens of neurons at a time. An alternative recording technique that overcomes many of these problems is electrocorticography (ECoG) where an array of electrodes, implanted for assessment of the brain before surgery, is used to record electrical fluctuations from the surface of the human. ECoG allows electrical signals from several different brain areas to be measured simultaneously while at the same time providing temporal resolution in the millisecond range. It is thus uniquely suited for probing the neural dynamics of cognition. With NSF support, Drs. Rajesh Rao and Jeff Ojemann of the University of Washington will examine cortical dynamics using three tasks that represent three different levels of abstraction in cognition: a motor movement task, a working memory task, and a language task. It will analyze the data using conventional spectral techniques as well as more sophisticated statistical techniques such as Independent Component Analysis (ICA) for source separation and Bayesian techniques for signal estimation. From this, they will construct biophysical models of networks of neurons and simulating cortico-cortical and cortico-thalamic feedback loops to understand the neural genesis of ECoG. If successful, this research will allow a new understanding of brain function, leading in the long term to possible remedies for cognitive deficits involving motor control, memory, or language processing.