Understanding Mental Models of Generative Conversational Search and The Effect of Interface Transparency

2025-06-04

AI Summary

In generative conversational search, users’ mental models are often overly abstract, hindering explanation of concrete search behaviors and inducing trust miscalibration. Method: We conducted a controlled experiment with 16 participants performing task-oriented search across four levels of interface transparency. Using mental model elicitation and systematic coding analysis, we characterized— for the first time—the structural properties of users’ mental models in this context. Contribution/Results: We demonstrate that interface transparency significantly affects both mental model accuracy and trust calibration. Based on these findings, we propose a novel “hybrid web-dialogue search” paradigm that enhances system explainability and user controllability. Our work provides both theoretical foundations and actionable design principles for developing interpretable and trustworthy generative search interfaces.

Technology Category

• Machine Learning: Transparent, Interpretable, Explainable ML
• Humans and AI: Intelligent User Interfaces
• Computer Vision: Interpretability, Explainability, and Transparency

Application Category

• User Modeling, Personalization and Recommendation: Accountability, Transparency, and Ethics for personalization
• Responsible Web: Algorithmic accountability and transparency on the web
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

The experience and adoption of conversational search is tied to the accuracy and completeness of users' mental models -- their internal frameworks for understanding and predicting system behaviour. Thus, understanding these models can reveal areas for design interventions. Transparency is one such intervention which can improve system interpretability and enable mental model alignment. While past research has explored mental models of search engines, those of generative conversational search remain underexplored, even while the popularity of these systems soars. To address this, we conducted a study with 16 participants, who performed 4 search tasks using 4 conversational interfaces of varying transparency levels. Our analysis revealed that most user mental models were too abstract to support users in explaining individual search instances. These results suggest that 1) mental models may pose a barrier to appropriate trust in conversational search, and 2) hybrid web-conversational search is a promising novel direction for future search interface design.

Problem

• Understanding user mental models in generative conversational search
• Exploring effects of interface transparency on system interpretability
• Addressing barriers to trust in conversational search interfaces

Innovation

• Studied mental models in conversational search
• Tested interfaces with varying transparency levels
• Proposed hybrid web-conversational search design