UC San Diego

Advances in Autonomous Vehicle (AV) technology promise significant individual and societal benefits. However, unresolved questions about how to meet the informational needs of riders hinder real-world adoption. Complicating our ability to satisfy rider needs is the intuition that different people, with different goals, and in different driving contexts, may have different requirements for what constitutes a successful interaction. Unfortunately, most human-AV research and design today treats all people and all situations as if they are the same. As a result, it is crucial that we understand what information an AV should communicate to meet rider needs, and how these communications should change when aspects of the human-AV complex system change, including when communications fail. I argue that understanding the relationships between different aspects of the human-AV system can help us build better AV communications and improve interactions beyond one-size-fits-all approaches. I support this argument using three empirical studies. First, by exploring the novel application of an AI race car driving coach, I identify optimal communication strategies that enhance driving performance, confidence, and trust for learning in even the most extreme driving environments. Findings highlight the need for task-sensitive, modality-appropriate communications tuned to learner cognitive limits and goals. Next, I examine the effect of AV communication errors on rider trust, showing that an error's impact is dependent on the driving context within which it occurs. Results highlight the consequences of deploying faulty communication systems and emphasize the need for context-sensitive communications. Third, I explore individual differences in trust perceptions, using machine learning (ML) to illuminate personal factors predicting young adult trust in AVs. This study highlights the importance of tailoring designs to individual traits and concerns, with implications for personalized design. Together, this dissertation supports the necessity of transparent, adaptable, and personalized AV systems that cater to individual needs, goals, and contextual demands. By considering the complex system within which human-AV interactions occur, we can deliver valuable insights for designers, researchers, and policymakers. This dissertation also provides a concrete domain to study theories of human-machine joint action and situational awareness, and can be used to guide future human-AI interaction research.