You're the One Whom I'm Talking To: The Role of Contextual External Human-Machine Interfaces in Multi-Road User Conflict Scenarios

As autonomous vehicles (AVs) become more prevalent, mixed-traffic environments involving pedestrians, cyclists, and manual vehicle drivers pose significant challenges for ensuring safe and effective interactions. External Human-Machine Interfaces (eHMIs) have emerged as a solution, particularly context-based eHMIs, which provide specific information such as Whom, When, and Where, showing potential for improving communication in complex scenarios. However, their impact on road user behavior and safety in interactions involving multiple road users remains insufficiently explored. This study addresses this gap by examining how contextual eHMIs affect crossing performance and subjective feelings during multi-user conflict scenarios. Using a virtual reality-based multi-agent simulation, 42 participants were equally divided into three groups-pedestrians, cyclists, and manual vehicle drivers-to make crossing decisions during interactions with an AV. Our findings demonstrated that providing contextual information in AV-multi-road user interactions significantly e nhanced p articipants' crossing performance and improved their perceived safety, trust, and clarity. These findings highlight the potential of context-based eHMIs to facilitate safer and more intuitive interactions in mixed-traffic environments.

CCS Concepts: • Human-centered computing → Empirical studies in HCI.