Examining Driver Situation Awareness in the Takeover Process of Conditionally Automated Driving With the Effect of Age

Background: Previous research has shown that drivers from different age groups demonstrated different Situation Awareness (SA) levels in the takeover process of conditional automated driving, where drivers need to collect surrounding information, perceive hazardous events, and make correct actions. Objective: To further explore the reason for this age-related difference, we investigated the SA from two different measures, Hazard Perception Time and scores from delayed SAGAT questions. The Hazard Perception time reflects both cognitive processes and ocular movements. The delayed SAGAT test was completed after each scenario, revealing how drivers perceived and comprehended information. Method: This study recruited drivers from three age groups, young, middle-aged, and older drivers. Especially, this study recruited old-old drivers (75+ years old), who have more severe cognitive impairments compared to young-old drivers (65 - 75 years old). Each driver went through 12 driving scenarios with different road types (highway straight, highway curved, city straight, and city curved) and driver types (manual only, autopilot only, and autopilot with non-driving-related Tasks). Results: The result showed that older drivers had significantly higher Hazard Perception Time and statistically equivalent SAGAT scores compared to the other two age groups. Also, curved roads led to significantly higher Hazard Perception Time for older drivers. Conclusion: Older drivers' decreased SA mainly came from the delayed ocular movement, which was moving their gaze toward hazardous events in the current experiment. Researchers should be more mindful regarding the slowed ocular movement of older drivers when designing takeover systems.