Applying Fitts' Law to a Brain-Computer Interface Controlling a 2D Pointing Device

Abstract

We have developed an EEG based non-invasive, consumer-convenient dry electrode brain-computer interface (BCI) for two-dimensional cursor control. This BCI drives a robot operation system (ROS) implemented pointer on a computer screen in real-time. Within the chosen paradigm, test subjects achieved an average hit rate of 56.8% for a point-and-click task after a brief training phase. The measured average BCI throughput is 0.18 bit/s. The sensorimotor rhythm-based BCI's performance is then compared to that of a computer mouse and a joystick by evaluating parameters according to Fitts' law.