Mental models in human AI interaction need clearer study methods
[MM/AI] Mental Models in Human-AI Interaction: Methods and Challenges in the Generative and Agentic AI Era (Workshop)
Human-Computer Interaction
Summary
People use mental models to understand how computers and AI systems work, but these mental models are often mixed up with similar ideas and studied in many different ways. The authors point out that new AI systems, which act independently for users and are hard to understand, make it even trickier to know how people form these mental models. They organized a workshop to rethink how mental models should be studied and understood in the age of advanced AI. This event brought experts together to share ideas and identify tough questions about mental models in AI.
What this means in practice
- •For human-computer interaction teams: Design new ways to study and clarify how users think about AI systems with complex independent behavior.
- •For product design teams: Develop user interfaces that better account for how people mentally represent AI acting on their behalf across applications.
A position paper. It proposes an approach and reports no results.
Authors
Téo Sanchez, Bhada Yun, Prerna Ravi, Laura Schütz, Anna Neumann, Robin Shing Moon Chan, April Yi Wang, Qiaosi Wang, Sumit Asthana
Abstract
The mental model construct is widely used in HCI to refer to the knowledge structure people hold in order to reason about and interact with computing systems. Yet it is often operationalized intuitively: the construct is often used interchangeably with related concepts (e.g., folk theories, sensemaking) and methods of studying it (e.g., through elicitation) are many and diverse, with each method resting on distinct assumptions about what counts as a mental model. Generative and agentic AI systems may further complicate mental model formation and elicitation as such systems are opaque by design and increasingly act on users' behalf across files, applications, and on the web. Together, these challenges may hinder the commensurability of research on people's mental models of AI systems. The MM/AI workshop calls for a critical reassessment of how we understand and study mental models in human-AI interaction research. It aims to foster theoretical and methodological exchange on mental models in human-AI interaction, identify open challenges, and develop directions for future research. We invite short papers on users' or stakeholders' mental models of AI systems, particularly contributions that reflect on the conceptual and methodological foundations of the construct. The half-day workshop combines lightning talks, hands-on elicitation exercises, and structured discussions on key questions concerning the future of the mental model for human-AI interaction research.