But first let me see: Comparison of light cues after a take-over request in automated driving

With conditional automated driving, drivers can engage in non-driving-related tasks, but they must remain capable of taking over control if the vehicle reaches a system boundary. For a safe take-over, the driver must have sufficient situation awareness. Previous research indicated that drivers often show dysfunctional gaze behavior, an indirect measure of situation awareness. Providing structured guidance for drivers’ gaze may mitigate dysfunctional gaze behavior. This approach could enhance perception, improve situation awareness, and increase safety. To identify the best method for guiding the gaze during a two-step take-over, we conducted a 4 x 2 within-subject experiment using a static driving simulator. We compared four light cue variants and two different sequences. The four variants of light cues were designed using attention-attracting features like size, movement, and blinking. Cues targeted the rear-view mirror, the side mirror, the road ahead, and the speedometer. We assessed gaze behavior, user experience, and driving behavior for N = 26 participants. Two variants were more effective in guiding the gaze: the variant using a big onset and the variant using growing movement. The big onset variant seems inappropriate because of the bad user experience. We propose an adapted variant that combines features of both light cue variants, which may deliver a positive user experience while remaining effective. This adapted variant will be tested in a future study. Furthermore, the light cues’ overall effectiveness regarding driving behavior, gaze behavior, and user experience should be evaluated and compared with that of other support methods.