Abstract

AbstractMotion-onset visual evoked potentials (mVEPs) are time- and phase-locked brain responses to motion-related stimuli. An mVEP response provides robust features for brain-computer interface (BCI) applications and has the added benefit of being less visually fatiguing than other visual evoked potentials (VEPs). In this study an mVEP BCI that enables control of a visually rich, three-dimensional (3D) car-racing video game is evaluated. A group of 15 teenage schoolchildren (13–16 years old) participated in a single session while they attended a summer school. Participants were asked to control the direction of a car within a realistic racing circuit, where the position of the car was controlled by focusing on one of five motion-related stimuli. Classification accuracy (%) and information transfer rate (ITR) (bits per minute (bpm)) results were encouraging, with participants achieving an average online accuracy of 72% (12 bpm) in the first lap, 67% (10 bpm) in the second lap and 65% (10 bpm) in the third lap (chance accuracy and ITR is 20% and zero bpm). The study shows for the first time the feasibility of using the mVEP paradigm in a commercial-grade car-racing video game. It is also one of the first reports on the performance of a group of teenagers using a BCI.