基于互信息的脑网络及测谎研究

互信息分析方法是基于信息论提出的一种描述两信号间信息交互情况的算法，其在脑电信号领域的有效性已得到了充分证实.针对当前测谎方法中脑电信号特征提取困难以及大脑整体认知功能分析在脑认知科学研究中越来越被重视的情况，本文首次将互信息分析方法应用到脑电测谎领域中，使用互信息量化大脑各节点之间的相关性，对计算结果进行统计分析，选取出在两类人群中具有显著性差异的电极对的互信息作为分类特征，进行模式识别，得到了99.67%的准确率.这一结果表明，互信息分析方法是一种有效的脑功能连接分析方法，为基于脑电信号连接分析的测谎研究提供了一种新的途径.另外，对说谎与诚实两类受试者的大脑功能网络的分析结果表明：处于说谎状态时，大脑的额叶、顶叶、颞叶及枕叶之间协同实现谎言功能，并在躯体行为所对应的脑区与其他脑区的连接上也表现出相对诚实组的显著性差异，以上结果均有助于进一步揭示谎言的神经活动机制.

Research on Mutual Information-Based Brain Network and Lie Detection

The mutual information analysis is a method based on information theory to describe the information interaction between two signals.In view of the difficulty in extracting features of EEG signals in the current lie detection method and the circumstance that the analysis of the overall cognitive function of the brain were increasingly important in brain cognitive science research,this paper applied the mutual information analysis method to the field of EEG lie detection for the first time and quantified the correlation between the brain nodes and perform statistical analysis on the calculation results.The mutual information of the electrode pairs with significant differences in the two groups were selected as the classification features,on which the pattern recognition was performed,resulting in the accuracy rate of 99.67%.This result proves that the mutual information analysis is an effective brain functional connection analysis method,which provides a new way for lie detection research based on EEG signal connection analysis.In addition,the brain function network of both lying and honest subjects was also analyzed.The results show that when lying,the frontal,parietal,temporal,and occipital regions of the brain cooperate to achieve the lie function,and in the connection between the brain regions corresponding to the physical behavior and other brain regions,significant differences between the two groups was also shown.These above results will help us further reveal the neural activity mechanism of the lie.