Vision-free brain-computer interface using auditory selective attention: Evaluation of training effect

Abstract

Evoked Potentials (EPs) recorded in electroencephalograms (EEGs) are widely applied in Brain-Computer Interface (BCI). However, many of these interfaces can cause fatigue because of the length of the sessions and the training or eye control required in the case of visual BCIs. The solution to this problem lies in the use of BCIs with auditory evoked potential using auditory steady state response (ASSR)—which can be modulated in a selective attention condition. Consequently, this study analyzed the training effect of a binary auditory BCI using selective attention and stimulation with amplitude-modulated (AM) tones. To assess the performance of this BCI, the hit rate achieved by 20 volunteers was verified over a period of four weeks. The results showed that the hit rate distribution over the four weeks presented similar values. Further, statistical analysis demonstrated that there was no statistical difference between these rates at a 5% significance level. Thus, a volunteer can use the auditory BCI with selective attention in a interval of one week, without any improvement in performance.