Causality analysis of multichannel alpha activities using multidimensional directed information

Analysis of information flow pattern among different portions on the scalp is important for investigating the relation between EEG and mental activity. Therefore, for the causality analysis, that is, the detection of the propagation path and direction of each component of EEG among different portions on the scalp, the authors have proposed a method based on the multidimensional directed information. The performance of the method was evaluated by simulation using signal generation models. Moreover, the method was applied to the analysis of alpha activities recorded for normal subjects and a patient whose brain has an organic disorder. The result showed that the detected pattern of information flow was well correlated with the disordered portion, consequently the method was a promising way for the estimation of mental activities