Communicating Internal and External Risks in Human-Robot Interaction

Abstract

Robots are expected to play an increasing role in users' lives, assisting with everyday tasks and complex decision-making, which creates the need to communicate risks -including probabilities and consequences of hazards-and to introduce robots as effective risk communicators to ensure safety. My research investigates how robots can convey two types of risks: external risks to inform users of relevant hazards, presenting the robot as a conduit for risk communication, and internal risks to transparently convey risks related to the robot itself. By leveraging the robot's embodiment and agency, my work addresses important and novel ways to communicate risk through robot affordances and their effects on user perception, decision-making, and affective responses. This dissertation offers empirical insights, implications, and design guidelines for creating robotic systems that support intuitive, human-centred risk communication.