Understanding Mental States in Active and Autonomous Driving with EEG

TL;DR

This study provides the first EEG-based comparison of driver mental states—cognitive load, fatigue, valence, and arousal—between active and autonomous driving using a simulated setup with three complexity levels. A Transformer-based classifier, trained under LOSO cross-validation, reveals a notable distribution shift in neural patterns between modes, driven by motor engagement and attentional demands. Self-supervised pretraining on public EEG datasets enhances transferability, while domain-shift analyses (UMAP) and cross-scenario training further quantify mode-specific differences. The findings underscore the need for scenario-specific data to design robust driver-monitoring systems for autonomous vehicles. Overall, EEG-based mental-state detection remains feasible in autonomous driving, with distinct neural signatures across driving modes guiding tailored monitoring approaches.

Abstract

Understanding how driver mental states differ between active and autonomous driving is critical for designing safe human-vehicle interfaces. This paper presents the first EEG-based comparison of cognitive load, fatigue, valence, and arousal across the two driving modes. Using data from 31 participants performing identical tasks in both scenarios of three different complexity levels, we analyze temporal patterns, task-complexity effects, and channel-wise activation differences. Our findings show that although both modes evoke similar trends across complexity levels, the intensity of mental states and the underlying neural activation differ substantially, indicating a clear distribution shift between active and autonomous driving. Transfer-learning experiments confirm that models trained on active driving data generalize poorly to autonomous driving and vice versa. We attribute this distribution shift primarily to differences in motor engagement and attentional demands between the two driving modes, which lead to distinct spatial and temporal EEG activation patterns. Although autonomous driving results in lower overall cortical activation, participants continue to exhibit measurable fluctuations in cognitive load, fatigue, valence, and arousal associated with readiness to intervene, task-evoked emotional responses, and monotony-related passive fatigue. These results emphasize the need for scenario-specific data and models when developing next-generation driver monitoring systems for autonomous vehicles.

Figures (13)

• Figure 1: The overall pipeline of our study.
• Figure 2: EEG device used in our study and the electrode location
• Figure 3: Data collection using the driving simulator.
• Figure 4: Self-reporting scale of cognitive load, fatigue and affect showing (a) Cognitive load and fatigue ranging from very low to very high (b) valence scale representing emotional pleasantness from very negative to very positive, and (c) arousal scale representing emotional activation from very low (calm) to very high (excited).
• Figure 5: The Wizard of Oz setup used in our study: a hidden human operator controls the vehicle while participants experience simulated autonomous driving.