Visual Stimulus Background Effects on SSVEP-Based Brain-Computer Interface

Abstract

The flickering source is an indispensable component in steady-state visual evoked potentials based brain-computer interface, and its background severely influences the potentials evoked by the repetitive stimuli. In this paper, we designed the experiment paradigm under three different backgrounds in the context of the SSVEP controlled small car, including black screen, static scene of the environment, and dynamic scene of the environment. From the spectrogram analysis of the EEG signals at occipital cortex, we found apparent decrease in SSVEP amplitude in dynamic scene condition comparing to the reference condition black screen. And the SSVEP amplitude changes under these three conditions further resulted in identification accuracy decreasing in dynamic scene condition as compared to black screen reference condition, which was evaluated from 10×10 cross validation. Besides, in real-time control of the small car, our results indicated that training in static scene condition exhibited better performance than that in black screen.