Impacts of system accuracy and environment status on human trust in intelligent vehicles: a simulated driving study with eye-tracking technique

Trust plays a key role in the acceptance and use of intelligent vehicles. This study aimed to examine the impacts of system accuracy and environment status on human trust in intelligent vehicles based on eye-tracking technique. A two-factor (3×2) mixed experimental design was employed. Three system accuracy levels (low, medium, and high) and two environment status (night-time and daytime) were examined. Participants were required to complete simulated driving tasks under the experimental conditions, during which data on a set of perception, behavioral, and eye-movement measures were collected. The results showed that system accuracy had a significant effect on compliance errors. There were significant differences in trust perception, compliance rate, compliance errors, pupil diameter, and number of saccades in different environment status. System accuracy and environment status had significant interaction effects on trust perception. Our findings contribute to the literature by highlighting the important role of system accuracy and environment status in shaping human trust towards intelligent driving, and provide practical implications for the design of reliable and acceptable intelligent vehicles.