Distracted when Using Driving Automation: A Quantile Regression Analysis of Driver Glances Considering the Effects of Road Alignment and Driving Experience

Abstract

Background: SAE Level 2 driving automation, the state-of-the-art in commercial vehicles, requires drivers to monitor the environment to resume vehicle control when automation limits are surpassed. However, this type of driving automation was found to increase distraction engagement. Although experienced drivers were shown to better regulate their visual engagement in distracting activities when driving with and without automation, the moderating effects of road demands (e.g., road alignment) have yet to be explored in detail, in particular, for driving with automation. Objective: To better understand the combined effects of road alignment, driving automation, and driving experience, we investigated the effects of these factors on driver glances to a visual-manual distraction task. Method: We present a secondary analysis, using quantile regression, of two previously reported driving simulator experiments. A total of 32 participants’ data, 16 from each experiment, were utilized. Half of these participants were novice and the other half were experienced drivers. The first experiment focused on non-automated driving, while the second focused on driving with adaptive cruise control and lane keeping assistance systems combined (i.e., SAE Level 2). The analysis reported here focuses on drivers’ visual distraction engagement in two highway drives that were identical across the two experiments. Results: With driving automation, compared to experienced driver glances, the duration of novice driver glances to the distraction task was more variable, longer, and less sensitive to variations in road alignment. Implications: These findings suggest that, with driving automation, novice drivers are more at risk of inappropriate engagement in distractions and do not adapt to road demands as well as experienced drivers, and thus should be supported accordingly.