명상도서관
F190. Mindfulness based stress reduction's influence on brain-computer interface control
자세히보기
- 자료유형학술지논문
- 저자명Stieger, James R.,Cline, Christopher C.,Coogan, Christopher,Rohit Murakonda, Bhavani Sai,Sherman, Samantha J.,Huynh, Andy,Hammond, Desirae,Breshears, Kit,Boyle, Taylor,Kreitzer, Mary J.,Engel, Stephen A.,He, Bin
- 학회/출판사/기관명Elsevier Science B.V., Amsterdam.
- 출판년도2018
- 언어영어
- 학술지명/학위논문주기CLINICAL NEUROPHYSIOLOGY -SHANNON-
- 발행사항Vol.129No.S1[2018]_x000D_
- ISBN/ISSN1388-2457
- 소개/요약Introduction Sensorimotor rhythm (SMR) based brain-computer interfaces (BCIs) can be used to control an external device through motor imagery. SMRs are oscillations that can be measured by electroencephalography (EEG) over the sensorimotor cortex. The ability to imagine movements and modulate SMRs is variable across individuals. Previous work has suggested that long-term mind-body awareness training (MBAT) offers merits in improving the performance of SMR based BCIs. However, whether such improved BCI performance results from a natural affinity for mental training or directly from MBAT is not clear. The aim of this study is to test the hypothesis that short-term MBAT can exert an influence over BCI control. Methods To assess the effects of MBAT on BCI control, 28 subjects attended an 8-week mindfulness based stress reduction (MBSR) class. BCI learning was compared between these MBAT subjects and 21 wait-listed controls, across several days of SMR based BCI training sessions. Subjects performed motor imagery of hand movements, while EEG rhythm at 12 Hz (3 Hz band, over C3/C4) was extracted and used to decode subjects’ intention to move a computer cursor. Subjects were instructed to imagine left (right) hand movement to move the cursor left (right), movement in both hands to move the cursor up, and a voluntary rest to move the cursor down. In separate blocks of trials, subjects attempted to steer the cursor to a target that required left/right (LR) movement only, up/down (UD) only, and combined 2D movement (2D). Percent valid correct (PVC) calculated as the number of hits divided by the total number of non-timeout trials was used to assess accuracy. Results Overall, MBAT subjects demonstrated improvements in accuracy across training sessions in the UD and 2D paradigms with increases in performance of 13% (p < 0.01) and 9% (p = 0.04), respectively. Smaller, not statistically significant training effects were seen in controls. However, the difference between groups was not significant (13% vs. 7% in UD (p = 0.17) and 9% vs. 7% in 2D (p = 0.67)). Conclusion MBAT subjects demonstrated mildly enhanced learning in our study, suggesting MBAT may be able to improve SMR based BCI training. Our past work showed an advantage for subjects with years of MBAT training; the present work extends this finding to 2 months of training. Work is ongoing to increase the sample size in the present study and examine whether cognitive, behavioral, or electrophysiological measures can explain the observed results. While works so far have focused on basic tasks such as computer cursor movement, it will be interesting to test whether MBAT can enhance learning of more complicated BCI tasks.
TOP
사전신청
오시는길
명상챗봇
