Mental models and driver takeover in automated vehicles: a systematic review

The integration of Automated Driving Systems (ADS) in vehicles is reshaping the role of drivers, introducing new challenges in human-automation interaction. This systematic review synthesises research concerning the relationship between drivers’ mental models and their performance during transitions from automated to manual control in vehicles with ADS. Following PRISMA guidelines, 38 peer-reviewed studies were analysed. A thematic analysis revealed three key themes relating to drivers’: (1) understanding of ADS; (2) attentional processing; and (3) attitudes towards ADS. Both training-based and design-based interventions appear to improve drivers’ mental models and concomitant performance. Design solutions offering continuous and transparent feedback about the state of ADS show particular promise. However, an integrated approach combining both training and design interventions could yield optimal outcomes. This systematic review examined how drivers’ mental models influence their performance when transitioning from automated to manual vehicle control. A thematic analysis of 38 studies identified that interfaces with easily interpretable and continuous feedback about the system status and limitations appear most effective in supporting accurate mental models and enhancing take-over performance.