Blind Source Separation of Event-Related Potentials Using Recurrent Neural Network

Abstract

Event-related potentials (ERPs) are a superposition of electric potential differences generated by neurophysiological activity associated with psychophysiological events. Spatiotemporal dissociation of underlying signal sources can supplement conventional ERP analysis and improve source localization. However, sources separated by independent component analysis (ICA) can be challenging to interpret because of redundant or illusory components and indeterminant polarity and scale. Hence, we have developed a recurrent neural network (RNN) method for blind source separation. The RNN transforms input step pulse signals representing events into corresponding ERP difference waveforms. Source waveforms are obtained from penultimate layer units and scalp maps are obtained from feed-forward output layer weights that project these source waveforms onto EEG electrode amplitudes. An interpretable, sparse source representation is achieved by incorporating L1 regularization of signals obtained from the penultimate layer of the network during training. This RNN method was applied to four ERP difference waveforms (MMN, N170, N400, P3) from the open-access ERP CORE database, and ICA was applied to the same data for comparison. The RNN decomposed real ERPs into eleven spatially and temporally separate sources that were less noisy, tended to be more ERP-specific, and were less similar to each other than ICA-derived sources. The RNN sources also had less ambiguity between source waveform amplitude, scalp potential polarity, and equivalent current dipole orientation than ICA sources. In conclusion, the proposed RNN blind source separation method can be effectively applied to average ERP waves and holds promise for further development as a computational model of event-related neural signals.