Corticocortical interactions in visual working memory

描述（由申请人提供）：拟议研究的长期目标是检验以下假设：工作记忆中有关对象身份和/或位置的信息的维护涉及位于分布式网络中的神经元群体的合作相互作用，该网络包括背外侧前额叶、后顶叶和颞下皮质区域。为了实现这一目标，我们计划开发和实施三个基本的方法，然后测试它们结合在一个单一的实验范式的可行性。首先，我们将训练猕猴执行眼动Go/No-Go延迟匹配样本任务。该任务要求动物灵活地整合和保留有关复杂视觉对象的身份和/或空间位置的信息，并使用这些信息来形成基于规则的行为决策。其次，我们设计了一种新型的显微操作器，使半慢性记录神经元活动，从36个独立可移动的微电极容纳在一个单一的记录室。我们计划构建，测试和优化这些设备，然后使用它们来记录神经元活动的时空模式，在背外侧前额叶，后顶叶，颞下皮层区域，而动物执行的行为任务。第三，我们已经开发了一个全面的方法，在快速的时间尺度上的神经元相关性的统计量化。我们计划收集的实验数据的初始机构，并使用这些分析方法来调查统计关系的活动在这三个领域的场电位和单个单元水平。这项研究计划构成了一个长期项目的初始阶段，以研究视觉认知的皮层机制。出于几个原因，它特别适合R21计划的目标。将开发各种新技术来进行实验，包括手术，机械，电生理和行为方法的创新。将采用新的分析方法来评价和解释所得数据。完成这一研究计划的初步阶段将为今后的应用奠定基础。

DESCRIPTION (provided by applicant): The long-term goal of the proposed research is to test the hypothesis that the maintenance of information about object identity and/or location in working memory involves the cooperative interaction of populations of neurons located in a distributed network that includes dorsolateral prefrontal, posterior parietal and inferior temporal cortical regions. To accomplish this goal, we plan to develop and implement three essential methodologies and then test the feasibility of combining them in a single experimental paradigm. First, we will train macaque monkeys to perform an oculomotor Go/No-Go delayed match-to-sample task. The task requires the animals to flexibly integrate and retain information about the identity and/or spatial location of complex visual objects, and use this information to form a rule-based behavioral decision. Second, we have designed a new type of micromanipulator that enables the semi-chronic recording of neuronal activity from up to 36 independently moveable microelectrodes housed within a single recording chamber. We plan to construct, test and optimize these devices, and then use them to record spatio-temporal patterns of neuronal activity in the dorsolateral prefrontal, posterior parietal, and inferior temporal cortical areas while the animals perform the behavioral task. Third, we have developed a comprehensive approach to the statistical quantification of neuronal correlations on a fast time scale. We plan to collect an initial body of experimental data and use these analytic methods to investigate the statistical relations of activity in these three areas at both the field potential and single unit levels. This research plan constitutes the initial phase of a long-term project to study the cortical mechanisms underlying visual cognition. For several reasons, it is particularly well suited to the goals of the R21 program. A variety of new techniques will be developed to conduct the experiments, including innovations in surgical, mechanical, electrophysiological and behavioral methods. New analytical methods will be incorporated to evaluate and interpret the resulting data. And, completion of this initial phase of the research plan will form the basis for future applications.