How Internal and External Risks Affect the Relationships Between Trust and Driver Behavior in Automated Driving Systems

Abstract

Automated driving systems (ADSs) allow vehicles to engage in self-driving under specific conditions. Along with the potential safety benefits, the increase in productivity through non-driving-related tasks (NDRTs) is often cited as a motivation behind the adoption of ADSs. Although advances have been made in understanding both the promotion of ADS trust and its impact on NDRT performance, the influence of risk remains largely understudied. To fill this gap, we conducted a within-subjects experiment with 37 licensed drivers using a simulator. Internal risk was manipulated by ADS reliability and external risk by visibility, producing a 2 (ADS reliability) × 2 (visibility) design. The results indicate that high reliability increases ADS trust and further enhances the positive impact of ADS trust on NDRT performance, while low visibility reduces the negative impact of ADS trust on driver monitoring. Results also suggest that trust increases over time if the system is reliable and that visibility did not have a significant impact on ADS trust. These findings are important for the design of intelligent ADSs that can respond to drivers' trusting behaviors.

Measuring and Understanding Trust Calibrations for Automated Systems: A Survey of the State-Of-The-Art and Future Directions

“…Screening of dangerous objects/subjects [4, 14, 15, 19, 36, 51, 54-56, 85, 94, 107, 117], detection of system malfunctions [89,90], crime prevention [10], recidivism prediction [132], watch a video of a house search [66,83] Transportation Responding to take-over requests [2, 6, 7, 47, 61, 68-70, 77, 84, 98], collusion avoidance [8], managing (air) traffic [31,118], pedestrians interaction with AV [48], observe AVs [64,119], drive in driving simulator [87,96,136] Military Screening tasks [17,33,44,63,81,82,127,130,135,139,140], gathering of information [58], mission planning [92], human-AI collaboration for search and destroy missions [116] Production Improving production [73,133,143], disassembly [5], moving objects [34,35,45], demand forecasting [39], harvesting [113], quality checks [142] Gaming trust game [3,23], collaboration game [24], flanker task…”

The Cognitive Effects of Machine Learning Aid in Domain-Specific and Domain-General Tasks

“…The breadth of applications of machine learning (ML) intelligent systems in predictive decision-making scenarios is substantial, and users are interfacing with these assistant systems in a multitude of ways. This human-ML collaboration is seen in a variety of fields, from medical diagnosis and treatment [1,2,3], to automated driving systems [4,5], to threat detection [6]. ML assistance has also been utilized in aiding visual search and predicting the presence of targets in modalities such as baggage screens [7,8,9] and simulated combat images [10].…”