Frequently, by the time car drivers perceive the danger of a crash while driving, they may not be able to move to avoid it. This reaction is a protection reflex that occurs when danger is sensed. During this reaction period, it is difficult to take action in order to avoid danger by manually controlling a vehicle. If this emergency state could be detected by an electroencephalogram (EEG), it would be possible to avoid danger automatically. In this study, the possibility to detect emergency states during car driving from the phase-locking index (PLI) of EEG was investigated. Five healthy subjects drove in a virtual reality (VR) environment with obstacles suddenly appearing in front of the car. The subjects were requested to push down a switch to stop the car immediately when they found an obstacle, to avoid crashing. VR scenes were visually presented to subjects via head-mounted displays. Two kinds of distance conditions between the car and obstacle were tested: a“far” distance, with approximately 1 s between the appearance of and collision with the obstacle, in which it was relatively easy to avoid crashing by braking; and a “near” distance with approximately 0.3 s between obstacle appearance and collision, in which crashing was more difficult to avoid. A 50-channel EEG was measured, and both the PLI and averaged evoked potentials were calculated over the span of the trials. In four out of five subjects, the phase-locking EEG was observed during the negative evoked potential at 100 ms after the obstacle appeared, and this occurred when the distance between the obstacle and the car was short, i. e. the near condition. As this phase-locking EEG response appeared immediately after the appearance of the obstacle, it is expected that the PLI of EEG can be used as an adequate indicator of an emergency state of the brain.
|ジャーナル||Transactions of Japanese Society for Medical and Biological Engineering|
|出版ステータス||Published - 2020|
ASJC Scopus subject areas
- Biomedical Engineering