The objective of the present work was to identify electroencephalographic (EEG) components in order to distinguish between braking and accelerating intention in simulated car driving. To do so, we collected high-density EEG data from thirty participants while they were driving in a car simulator. The EEG was separated into independent components that were clustered across participants according to their scalp map topographies. For each component, time-frequency activity related to braking and acceleration events was determined through wavelet analysis, and the cortical generators were estimated through minimum norm source localisation. Comparisons of the time-frequency patterns of power and phase activations revealed that theta power synchronisation distinguishes braking from acceleration events 800 ms before the action and that phase-locked activity increases for braking 800 ms before foot movement in the theta-alpha frequency range. In addition, source reconstruction showed that the dorso-mesial part of the premotor cortex plays a key role in preparation of foot movement. Overall, the results illustrate that dorso-mesial premotor areas are involved in movement preparation while driving, and that low-frequency EEG rhythms could be exploited to predict drivers' intention to brake or accelerate.
Keywords: Cluster analysis; Driving; High-density EEG; ICA; Movement preparation; Time-frequency analysis.
Copyright © 2018 Elsevier B.V. All rights reserved.