Abstract
BACKGROUND: In safety-critical environments, human error is a leading cause of accidents, with the loss of situation awareness (SA) being a key contributing factor. Accurate SA assessment is essential for minimizing such risks and ensuring operational safety. Traditional SA measurement methods have limitations in dynamic real-world settings, while physiological signals, particularly EEG, offer a non-invasive, real-time alternative for continuous SA monitoring. However, the reliability of SA measurement based on physiological signals depends on the accuracy of SA labeling. OBJECTIVE: This study aims to design an effective SA measurement paradigm specific to high-speed train driving, investigate more accurate physiological signal-based SA labeling methods, and explore the relationships between SA levels and key physiological metrics based on the developed framework. METHODS: This study recruited 19 male high-speed train driver trainees and developed an SA measurement paradigm specific to high-speed train driving. A method combining subjective SA ratings and task performance was introduced to generate accurate SA labels. RESULTS: The results of statistical analysis confirmed the effectiveness of this paradigm in inducing SA level changes, revealing significant relationships between SA levels and key physiological metrics, including eye movement patterns, ECG features (e.g., heart rate variability), and EEG power spectral density across theta, alpha, and beta bands. CONCLUSIONS: This study supports the use of multimodal physiological signals for SA assessment and provides a theoretical foundation for future applications of SA monitoring in railway operations, contributing to enhanced operational safety.