Brain Computer Interface based on the Steady-State VEP for Immersive Gaming Control

This paper presents an implementation of an effective EEG-based Brain Computer Interface design as the control mechanism for an immersive 3-D game. The BCI is based on the application of the steady-state visual evoked potential (SSVEP) generated in response to phase-reversing checkerboard patterns. Real-time game control and signal processing is facilitated by a combined graphics/DSP/network communications framework. Three methods of feature extraction are compared in a series of offline classification tests. Despite the relatively uncontrolled environment in which the game was played, the performance of the BCI control was robust and relatively consistent across all subjects with 41 of 48 games successfully completed. Performance while playing the game is found to be an improvement on performance in the offline tests, the cause of which is attributed to the goal-driven nature of the game.