A HYBRID VISUAL EVOKED PARADIGM FOR BRAIN COMPUTER INTERFACE BASED ON THE RADIAL MOTION OF ROBOTIC ARMS

Visual evoked stimulus is one of the most important paradigms for construction of brain computer interface (BCI). To improve the diversity of the visual stimulus paradigm, this paper proposes a hybrid visual evoked stimulus paradigm to simultaneously evoke event-related desynchronization (ERD) and steady-state motion visual evoked potential (SSMVEP). The hybrid visual evoked stimulus paradigm is based on the radial motion of robotic arms. The SSMVEP is generated by the periodic motion of the robotic arm modulated by the sinusoidal function, and the ERD is generated by observing the motion of robotic arms and discriminating between left and right. Six subjects were invited to participate in the experiment, and radial motions of robotic arms with six different frequencies were designed. During the experiment, the stimulus paradigms of the arm movement based on body (AMB), arm movement based on itself (AMI), and Newton rings were designed for comparison, and two different algorithms were used to analyze the ERD and SSMVEP for the recorded EEG signals, respectively. Experimental results have shown that the proposed hybrid paradigm will evoke such two phenomena with good recognition accuracy, and the recognition accuracy of the SSMVEP is significantly higher than that of the Newton rings. The results of the analysis for the fatigue and comfort questionnaire after the experiment have shown that the proposed hybrid paradigm has a higher comfort and can cope well with fatigue after a long-term use.