Complex Network Analysis Of Experimental Eeg Signals For Decoding Brain Cognitive State

Depicting the relationship between brain cognitive state and task difficulty level constitutes a challenging problem of significant importance. In order to probe it, we design an experiment to gather EEG data from mental arithmetic task under different difficulty levels. We construct brain complex networks using a complex network method and information entropy theory. We then employ weighted clustering coefficient to characterize the networks generated from different brain cognitive states. The results show that with the increase in task difficulty level, the mean weighted clustering coefficients show a decrease. This is due to the lack of coordination of brain activity and the low efficiency of the network organization caused by the increase in task difficulty. In addition, we calculate the permutation entropy from the signals of each channel EEG signals to support the findings from our network analysis. These findings render our method particularly useful for depicting the relationship between brain cognitive state and difficulty level.