S-Space College of Medicine/School of Medicine (의과대학/대학원) Dept. of Neurology (신경과학교실) Journal Papers (저널논문_신경과학교실)
Classification of selective attention to auditory stimuli: Toward vision-free brain-computer interfacing
- Kim, Do-Won; Hwang, Han-Jeong; Lim, Jeong-Hwan; Lee, Yong-Ho; Jung, Ki-Young; Im, Chang-Hwan
- Issue Date
- Journal of Neuroscience Methods, Vol.197 No.1, pp. 180-185
- 공학; Brain–computer interface (BCI); Auditory steady-state response (ASSR); Selective attention; Amyotrophic lateral sclerosis (ALS); Completely locked-in syndrome
- Brain-computer interface (BCI) is a developing, novel mode of communication for individuals with severe motor impairments or those who have no other options for communication aside from their brain signals. However, the majority of current BCI systems are based on visual stimuli or visual feedback, which may not be applicable for severe locked-in patients that have lost their eyesight or the ability to control their eye movements. In the present study, we investigated the feasibility of using auditory steady-state responses (ASSRs), elicited by selective attention to a specific sound source, as an electroencephalography (EEG)-based BCI paradigm. In our experiment, two pure tone burst trains with different beat frequencies (37 and 43 Hz) were generated simultaneously from two speakers located at different positions (left and right). Six participants were instructed to close their eyes and concentrate their attention on either auditory stimulus according to the instructions provided randomly through the speakers during the inter-stimulus interval. EEG signals were recorded at multiple electrodes mounted over the temporal, occipital, and parietal cortices. We then extracted feature vectors by combining spectral power densities evaluated at the two beat frequencies. Our experimental results showed high classification accuracies (64.67%, 30 commands/min, information transfer rate (ITR) = 1.89 bits/min; 74.00%, 12 commands/min, ITR = 2.08 bits/min; 82.00%, 6 commands/min, ITR = 1.92 bits/min; 84.33%, 3 commands/min, ITR = 1.12 bits/min; without any artifact rejection, inter-trial interval = 6 s), enough to be used for a binary decision. Based on the suggested paradigm, we implemented a first online ASSR-based BCI system that demonstrated the possibility of materializing a totally vision-free BCI system.
- Files in This Item: There are no files associated with this item.