Coupling analysis of EEG based on simultaneous screening - Symbolic transfer entropy

Analysis of the coupling of two time series with a large amount of non-related data, non-relevant data will have an adverse effect on the coupling analysis. In order to solve this problem, The synchronous point screening algorithm is proposed to eliminate the non-related information in this paper. To realize the coupling analysis of brain motor area and sensory area during designed grip task at 5kg and 20kg, synchronous point screening algorithm is applied to extract components of in EEG that linearly synchronized with EMG. Then, the symbolic transfer entropy is calculated based on the linearly synchronous components of EEG. Symbolic transfer entropy is capable to revealing directed interactions between signals. Results showed that there's a bidirectional coupling relationship between the motor area and the sensory area of the brain. The EEG signals of the two areas have stronger coupling strength in greater grip, but components of EEG that linearly synchronized with EMG have weaker coupling strength in greater grip. The results reveal that the linear coupling strength of motor area and sensory area in central nervous system is weaker and the nonlinear coupling strength is stronger under the greater grip.