The Role of LFP Spatial Structure in the Hippocampus

描述（由申请人提供）：这项工作的长期目标是了解海马体内神经元的活动是如何协调的，海马体是一个参与记忆存储的大脑区域。这种协调的一种表现是局部场电位（LFP），即少量神经组织的电活动总和。最近，多电极阵列（MEA）已经使得能够在动物执行任务时同时测量来自动物脑内的多个部位的LFP和单神经元活动。这一进步推动了新的计算方法的发展，这些方法可以识别这些丰富的大型数据集中存在的关系。海马LFP的时间结构与单神经元活动以及行为状态相关。相比之下，这些LFP的空间结构仍然相对未被探索。这种空间结构的一个有趣的例子是通过海马体传播的LFP波。本研究将探讨海马LFP的空间动力学是否影响神经元活动及其与行为的关系。其目的是：1）确定一个简约的模型来解释观察到的时空结构的LFP; 2）调查LFP结构的关系，以单神经元活动;和3）了解LFP结构如何控制回路水平的海马神经元处理。为了实现这些目标，本项目将采用统计学习技术从LFP中提取时空信息，并使用预测模型来研究不同LFP结构对海马神经元群体活动的影响。这项工作将确定简洁的指标，捕捉高维海马LFP测量的丰富性。此外，它将研究将复杂的动力学特征纳入目前的海马处理模型的效用。这项工作可能有助于理解海马体内电活动的扭曲如何导致癫痫和健忘症等疾病。它将评估LFP作为神经假体干预的直接目标的适用性。最后，它可能会导致筛查程序，以快速发现患者大脑活动的潜在异常，并有助于了解此类异常对患者生活方式的影响。

DESCRIPTION (provided by applicant): The long-term objective of this work is to understand how the activity of neurons is coordinated within the hippocampus, a brain region involved in the storage of memories. One manifestation of this coordination is the local field potential (LFP), the summed electrical activity of a small volume of neural tissue. Recently, multi- electrode arrays (MEAs) have enabled the simultaneous measurement of LFPs and single-neuron activity from multiple sites within the brain of an animal as it performs a task. This advance motivates the development of new computational methods that can identify the relationships that exist within these rich, large data sets. The temporal structure of hippocampal LFPs is known to correlate with single-neuron activity, as well as with behavioral state. In contrast, the spatial structure of these LFPs remains relatively unexplored. An interesting example of this spatial structure are the traveling LFP waves that propagate through the hippocampus. This project will examine whether the spatial dynamics of the hippocampal LFP influence neuronal activity and its relationship to behavior. Its aims are 1) to identify a parsimonious model for explaining the observed spatio- temporal structure in the LFP; 2) to investigate the relationship of LFP structure to single-neuron activity; and 3) to understand how LFP structure governs circuit-level neuronal processing within the hippocampus. To achieve these goals, this project will employ statistical learning techniques to extract spatio-temporal regularities from within the LFP, and use predictive models to examine the influence of different LFP structures on the population activity of hippocampal neurons. This work will identify concise metrics that capture the richness of high-dimensional hippocampal LFP measurements. Furthermore, it will examine the utility of incorporating complex dynamical features into current models of hippocampal processing. This work may help understand how distortions in electrical activity within the hippocampus lead to conditions such as epilepsy and amnesia. It will evaluate the suitability of the LFP as a direct target for neuroprosthetic interventions. Finally, it may lead to screening procedures for rapidly finding potential abnormalities in patients' brain activity, and help understand the implications of such abnormalities for patients' lifestyles.