Single Trial Analysis of Event Related Signals

描述(申请人提供)：神经信号非常多变。理解变异性并利用其丰富的内容是当代神经科学的核心。在本申请中，我们建议解决这一广阔领域的一个方面：在逐个试验的基础上估计与事件相关的场电位信号，并应用估计的单个试验与事件相关的参数来解决与神经系统功能相关的问题。有两个具体目标。在目标1中，我们建议进一步开发和彻底验证一种称为差异变量成分分析(DVCA)的单项试验分析方法。在目标2中，我们建议应用该方法来分析来自两个已有数据集的局部场电位：一个来自执行视觉运动模式辨别任务的猕猴，另一个来自执行多模式(视觉和听觉)选择性注意任务的猕猴。第一个数据集是独一无二的，因为它包括从一个半球长期植入的多达16个双极皮质内电极同时记录的局部场电位，因此非常适合于解决与神经激活及其任务相关性的定时和大规模联网有关的问题。第二个数据集的独特之处在于包括从多个触点同时记录的局部场电位，该局部场电位沿着横跨初级视觉皮质中所有皮质层的线性电极。我们的目标是研究注意力是如何调节视觉信息处理的。

DESCRIPTION (provided by applicant): Neural signals are extremely variable. Understanding the variability and harnessing its rich content lie at the very heart of contemporary neuroscience. In this application we propose to tackle 1 aspect of this vast field: estimation of event related field potential signals on a trial-by-trial basis and applying the estimated single trial event related parameters to address questions related to functions of neural systems. There are 2 Specific Aims. In Aim 1 we propose to further develop and thoroughly validate a single-trial analysis methodology termed differentially variable component analysis (dVCA). In Aim 2 we propose to apply the methodology to analyze local field potentials from 2 existing datasets: 1 from macaque monkeys performing a visuomotor pattern discrimination task and the other from macaque monkeys performing an intermodal (visual versus auditory) selective attention task. The first dataset is unique in that it consists of local field potentials simultaneously recorded from up to 16 bipolar intracortical electrodes chronically implanted in one hemisphere, and is therefore ideally suited for addressing issues related to timing and large-scale networking of neural activations and their task relevance. The second dataset is unique in consisting of local field potentials recorded simultaneously from multiple contacts along a linear electrode spanning all the cortical layers in the primary visual cortex. Our goal is to study how visual information processing is modulated by attention.