Distraction mitigation through driver monitoring systems: a scoping review

Driver monitoring systems (DMS) are increasingly recommended or required by national and regional governing bodies. We present a scoping review of DMS-based real-time interventions for distracted driving, aiming to identify relevant research trends and gaps, and to help guide intervention design and implementation. Five databases (ACM Digital Library, Scopus, IEEE Xplore, TRID, SAE Mobilus) and a reference tracking platform (Research Rabbit) were systematically searched in May 2024. Primary research assessing DMS-based distraction interventions in a driving context was included ( n = 28). Although there is vast literature on the effects of driver distraction, countermeasures, and driver state detection algorithms, DMS-based interventions have received less attention from the research community. Warnings were evaluated the most ( n = 23), followed by advanced driver assistance systems ( n = 6), adaptive user interfaces (n = 2), and sociable agents ( n = 1). Overall, interventions helped reduce the impact of distractions on driving and gaze behaviors and distraction engagement. However, some studies reported undesirable effects, like increased distraction engagement ( n = 5), potentially due to over-trust in driving automation, DMS inaccuracy, or driver confusion. Interventions targeted mostly visual-manual ( n = 14) and visual (n = 5) distractions based on off-road glance information, which may not capture look-but-fail-to-see events or cognitive distraction. Seven studies investigated interventions in an automated driving context, but it remains unclear whether interventions found effective for manual driving can be adapted for automated driving. We summarize research trends in a taxonomy of distraction mitigation strategies and consolidate our findings with a previous scoping review on DMS-based drowsiness interventions to inform general DMS research.