To Trust or Not to Trust? A Simulation-based Experimental Paradigm

The automated driving system is expected to enhance traffic safety and flow; however, the system will not be as effective if users do not accept it or do not utilize it appropriately. Appropriate acceptance and use of technology depends on attributes such as perceived risk, mental workload, self-confidence, and appropriate level of trust that matches system performance. An inappropriate level of trust in the technology, whether it is over-trust or undertrust, would negatively affect the benefits of that technology. Based on the literature, trust is a dynamic construct that consists of an initial or dispositional trust that is shaped before experiencing the system performance and a history-based trust that constantly changes with user experience of the system. This study first reviews the history of research on humans’ trust in automation and the factors that are correlated with trust. It also provides a brief overview of some previous models of trust in automation. Then, based on the gaps in the literature, a simulator-based experiment is proposed to further study the factors affecting initial or dispositioned trust and history-based trust. The results of this study are expected to help better understand drivers’ trust in automated vehicles and help enhance human-automation interaction models.自动驾驶系统（automated driving system）有望提升交通安全与通行效率；然而，若用户不接受该系统，或未合理使用，其效能将大打折扣。技术的合理接纳与应用，取决于感知风险、心理负荷（mental workload）、自信心，以及与系统性能相匹配的适度信任等属性。对技术的信任程度失当——无论是过度信任还是信任不足，都会对该技术的应用效益产生负面影响。基于现有研究文献，信任是一种动态建构：既包含在体验系统性能前便已形成的初始信任（或称特质性信任，dispositional trust），也包含随用户对系统的使用体验不断动态调整的基于经验的信任。本研究首先梳理了人类对自动化系统信任的研究脉络，以及与信任相关的各类影响因素，并简要概述了此前提出的若干自动化信任模型。随后，针对现有研究的空白，本研究提出一项基于模拟器的实验，以进一步探究影响初始（特质性）信任与基于经验的信任的各类因素。本研究的成果有望助力更深入地理解驾驶员对自动驾驶车辆的信任，并推动人机交互（human-automation interaction）模型的优化与完善。