Is take-over time all that matters? The impact of visual-cognitive load on driver take-over quality after conditionally automated driving.

Currently, development of conditionally automated driving systems which control both lateral and longitudinal vehicle guidance is attracting a great deal of attention. The driver no longer needs to constantly monitor the roadway, but must still be able to resume vehicle control if necessary. The relaxed attention requirement might encourage engagement in non-driving related secondary tasks, and the resulting effect on driver take-over is unclear. The aim of this study was to examine how engagement in three different naturalistic secondary tasks (writing an email, reading a news text, watching a video clip) impacted take-over performance. A driving simulator study was conducted and data from a total of 79 participants (mean age 40 years, 35 females) were used to examine response times and take-over quality. Drivers had to resume vehicle control in four different non-critical scenarios while engaging in secondary tasks. A control group did not perform any secondary tasks. There was no influence of the drivers' engagement in secondary tasks on the time required to return their hands to the steering wheel, and there seemed to be only little if any influence on the time the drivers needed to intervene in vehicle control. Take-over quality, however, deteriorated for distracted drivers, with drivers reading a news text and drivers watching a video deviating on average approximately 8-9cm more from the lane center. These findings seem to indicate that establishing motor readiness may be carried out almost reflexively, but cognitive processing of the situation is impaired by driver distraction. This, in turn, appears to determine take-over quality. The present findings emphasize the importance to consider both response times and take-over quality for a comprehensive understanding of factors that influence driver take-over. Furthermore, a training effect in response times was found to be moderated by the drivers' prior experience with driver assistance systems. This shows that besides driver distraction, driver-related factors influencing take-over performance exist.