標題: Can arousing feedback rectify lapses in driving? Prediction from EEG power spectra
作者: Lin, Chin-Teng
Huang, Kuan-Chih
Chuang, Chun-Hsiang
Ko, Li-Wei
Jung, Tzyy-Ping
生物科技學系
電機工程學系
Department of Biological Science and Technology
Department of Electrical and Computer Engineering
公開日期: 1-十月-2013
摘要: Objective. This study explores the neurophysiological changes, measured using an electroencephalogram (EEG), in response to an arousing warning signal delivered to drowsy drivers, and predicts the efficacy of the feedback based on changes in the EEG. Approach. Eleven healthy subjects participated in sustained-attention driving experiments. The driving task required participants to maintain their cruising position and compensate for randomly induced lane deviations using the steering wheel, while their EEG and driving performance were continuously monitored. The arousing warning signal was delivered to participants who experienced momentary behavioral lapses, failing to respond rapidly to lane-departure events (specifically the reaction time exceeded three times the alert reaction time). Main results. The results of our previous studies revealed that arousing feedback immediately reversed deteriorating driving performance, which was accompanied by concurrent EEG theta-and alpha-power suppression in the bilateral occipital areas. This study further proposes a feedback efficacy assessment system to accurately estimate the efficacy of arousing warning signals delivered to drowsy participants by monitoring the changes in their EEG power spectra immediately thereafter. The classification accuracy was up 77.8% for determining the need for triggering additional warning signals. Significance. The findings of this study, in conjunction with previous studies on EEG correlates of behavioral lapses, might lead to a practical closed-loop system to predict, monitor and rectify behavioral lapses of human operators in attention-critical settings.
URI: http://dx.doi.org/10.1088/1741-2560/10/5/056024
http://hdl.handle.net/11536/22717
ISSN: 1741-2560
DOI: 10.1088/1741-2560/10/5/056024
期刊: JOURNAL OF NEURAL ENGINEERING
Volume: 10
Issue: 5
結束頁: 
顯示於類別:期刊論文


文件中的檔案:

  1. 000324862400024.pdf

若為 zip 檔案,請下載檔案解壓縮後,用瀏覽器開啟資料夾中的 index.html 瀏覽全文。