Investigating the situational dynamics of visual information sampling in lateral vehicle control : Subjective vs. objective estimates of spare visual capacity

Abstract

Drivers continually adapt their information sampling behavior to changing traffic conditions for safe driving. Scientists have studied this sampling behavior for decades; however, the literature on how drivers adapt their visual information sampling in response to observed driving dynamics is still incomplete, especially concerning what might be considered safe adaptation from an external perspective. While occlusion methods are commonly employed to study drivers’ visual information sampling, the variability in self-selected occlusion times and their relationship to actual driving performance has yet to be fully understood. In a driving simulator study with 30 participants, we analyzed and compared the situational dynamics influencing visual information sampling and performance in an occluded lane-keeping task. The findings underscore the significant influence of speed, lane position, time-to-line-crossing at the start of occlusion, and steering during occlusion on spare visual capacity in lane-keeping. Although the participants were able to make slight adjustments to their visual sampling based on these variables, their occlusion time choices appeared to be stable and primarily driven by individual preferences, unrelated to their driving experience or general lateral control instability under occlusion. In contrast, drivers’ general instability in lateral control under single-occlusion driving emerged as the strongest predictor of lane crossing during continuous, intermittently occluded driving. These insights contribute to the understanding of information sampling dynamics and spare visual capacity in lateral vehicle control, potentially guiding the development of personalized and contextually intelligent driver attention monitoring and warning systems.