A brain-computer interface speller using peripheral stimulus-based SSVEP and P300

In brain-computer interface (BCI) research, spellers are valuable issues because they can provide communication channel to human. In this paper, we propose a novel hybrid speller that is SSVEP feedback with peripheral-vision stimulus to the conventional P300 paradigm. A canonical correlation analysis (CCA) and a linear discriminant analysis (LDA) classified SSVEP and P300, respectively. Four subjects participated in experiments, in which accuracy was compared with those of other spellers. Proposed approach revealed sufficient P300 and SSVEP potentials without interaction effect and time consuming, and also reduced visual fatigue. The results show that this research suggests a promising approach to make the speller more time-efficient.