Use of a psychological learning technique to improve brain-computer interface (BCI) control

Abstract

A Brain-Computer Interface (BCI) is a system that connects the human brain to a computer, allowing their users to communicate or control external devices using brain signals. Performance of BCI depends on the ability of subjects to control their own electroencephalographic (EEG) patterns, being necessary to provide suitable neurofeedback training techniques. The main objective of this work is to use a psychological learning technique in order to help users to acquire specific skills on the control of a BCI. The feedback provided to the subject was based on virtual reality techniques. The well-known Skinner´s shaping method (based on successive approximations) was used to reinforce two mental tasks: imagination of right hand movements and an idle state, such as mental relaxation. Two groups (5 and 4 subjects each) participated in the same number of training sessions; a control group used a standard procedure and the other group tested the shaping proposal. Both mental tasks were considered separately, so each one received different shaping levels depending on the initial control and the evolution through the training sessions. The shaping effect was implemented by modifying the feedback that users received. The study proved the effectiveness of the proposed method by enhancing the difference between the EEG responses corresponding to these two mental tasks. The use of a specific psychological learning technique in order to shape mental activity improved the training process over the control of a BCI, by adapting to the real needs of each user.