Spatio-temporal analysis of P300 using ICA and SSLOFO

Spatial information of EEG source activity revealed by inverse methods may contribute to an improvement of the BCI systems. This paper proposes an approach that integrates the independent component analysis (ICA) and a newly developed inverse algorithm termed SSLOFO to robustly reconstruct cortical sources of P300. The target independent components are first extracted using a spatio-temporal optimization process and then SSLOFO is employed to localize the sources of the target components. Preliminary studies demonstrate our method is able to localize sources of P300 based on 5-trial-averaged EEG and the results are consistent with the findings of other functional imaging studies such as fMRl. The robustness of our approach is also proved by a study which indicates the P300 sources are stably reconstructed around the left and right TPJ areas