Influence of Difference of Spatial Information Obtained from a Moving Virtual Sound Presentation on Auditory BCI

Abstract

Brain-computer interface (BCI) technology can control external devices by using human brain activity. In a previous study, a compact BCI system for the estimation of the intended direction was realized by using virtual sound. Some data were averaged to improve the identification rate. Herein, we expected to obtain a higher identification rate with a small amount of averaged data by improving the localization accuracy of users. In this study, to investigate the effect of the difference of auditory stimulation methods (static sound and moving sound) on brain activity, we performed an experiment with six directions of auditory stimulation using each method and measured the brain activity of subjects. We used a variant of regularized Fisher’s discriminant analysis to classify brain waves. As a result of comparing the identification rates obtained through each method, individual differences were observed in the effect on the identification rate, and an improvement was observed for moving sound as compared with static sound.