Correlation Studies of P300 and EEG Rhythms Using dVCA

Analysis of single-trial mental EEG may potentially reveal more information about for brain dynamics investigating than simple averaging methods. However, Traditional EEG analysis relies on the averaging methods to increase event related potentials and basic rhythms extraction from EEG, which face many signal processing challenges, such as trial-to-trial variability in latencies, amplitudes of event-related response(ERP), and separation of evoked EEG and spontaneous EEG which is a serious problem for EEG interpretation and analysis. Here, we introduce differentially variable component analysis (dVCA) method to estimate the P300 amplitude in each trial, and extract theta rhythm from spontaneous EEG free of ERPs, which allows the interaction between ERPs and the ongoing EEG to be investigated directly. We test our methods on dataset from Eriksen flanker experiments. Our result shows that the method using dVCA can perform well to extract ERP variables and remove ERP from spontaneous EEG. In addition, the results also show that the approach can identify the correlation between theta rhythm energy and amplitudes in single-trial EEG signals, and proving the approach is an effective and efficient method to study the brain dynamics.