Using Error Potentials to Improve Auditory BCI

Abstract

Ever since the start of the development of brain-computer interfaces (BCIs), they have been envisioned as a tool that could help impaired people in their interaction with the world. BCIs translate brain signals into computer commands. Auditory BCIs, which only require an intact hearing, could be used as a communication tool for patients with locked-in syndrome who have very little control over their muscles or none at all, making it hard to control BCIs based on (eye) movements. In this paper we execute a pilot study in which we propose a way to improve the accuracy of an auditory BCI. In order to answer a binary question, subjects control the BCI by changing their state of mind as a result of a shift of attention between auditory streams. A certain component of the measured brain signal, an error potential, could be detected during the feedback phase of the BCI in case of incorrect provided feedback. If an error potential could be detected, the initial feedback could be corrected, increasing the overall accuracy of the BCI. The results from this experiment show that there was no noticeable improvement in the performance of the auditory BCI after addition of a second brain signal (error potentials); the overall accuracy even seemed to diminish.