Functional connectivity among spike trains in neural assemblies during rat working memory task
Functional connectivity among spike trains in neural assemblies during rat working memory task
复制标题
大鼠工作记忆任务期间神经组件中尖峰序列之间的功能连接
DOI：
10.1016/j.bbr.2014.08.027
复制
发表时间：
2014-11-01
影响因子：
2.7
通讯作者：
Tian, Xin
中科院分区：
文献类型：
--
作者：
Xie, Jiacun;Bai, Wenwen;Tian, Xin
工作记忆是指大脑系统，它为复杂的认知任务提供临时存储来操纵信息。由于大脑是一个复杂的、动态的、相互交织的连接和相互作用的网络，如何从脑网络功能连接的角度来研究工作记忆的机制？如何在一个低维网络中呈现函数连通性的最典型特征？为了解决这些问题，我们用多电极记录了大鼠在Y-迷宫中执行工作记忆任务时前额叶皮层的锋电位序列。通过最大似然估计（MLE）计算锋电位序列之间的功能连接矩阵。计算平均连接性值C-c，矩阵的平均值，并用于定量描述连接性强度。利用功能连接矩阵构建锋电位网络。计算了网络的信息传递效率E-glob，并将其用于表征网络的特征。为了建立一个低维的锋电位网络，基于稀疏编码的方法选择放电率高于平均放电率的激活神经元。结果表明，在任务执行过程中，网络的连通度C-c和网络传输效率E-glob随时间变化。C-c和E-glob的最大值出现在工作记忆行为参考点之前。与原始网络的结果相比，该特征网络能更好地表征功能连通性。(C)2014爱思唯尔有限公司版权所有。
Working memory refers to a brain system that provides temporary storage to manipulate information for complex cognitive tasks. As the brain is a more complex, dynamic and interwoven network of connections and interactions, the questions raised here: how to investigate the mechanism of working memory from the view of functional connectivity in brain network? How to present most characteristic features of functional connectivity in a low-dimensional network? To address these questions, we recorded the spike trains in prefrontal cortex with multi-electrodes when rats performed a working memory task in Y-maze. The functional connectivity matrix among spike trains was calculated via maximum likelihood estimation (MLE). The average connectivity value C-c, mean of the matrix, was calculated and used to describe connectivity strength quantitatively. The spike network was constructed by the functional connectivity matrix. The information transfer efficiency E-glob was calculated and used to present the features of the network. In order to establish a low-dimensional spike network, the active neurons with higher firing rates than average rate were selected based on sparse coding. The results show that the connectivity C-c and the network transfer efficiency E-glob vaired with time during the task. The maximum values of C-c and E-glob were prior to the working memory behavior reference point. Comparing with the results in the original network, the feature network could present more characteristic features of functional connectivity. (C) 2014 Elsevier B.V. All rights reserved.