Adaptive Questioning as Decision Intelligence: A Progressive Disclosure Framework for Uncertainty Driven Human AI Interaction

Decision-making under uncertainty requires individuals to acquire relevant information while managing cognitive effort and time constraints. Conventional decision support systems rely heavily on static forms or rigid conversational interfaces, which often overwhelm users and fail to adapt to evolving decision states. This paper proposes a Progressive Disclosure Adaptive Questioning Interface (PDAQI) as a novel human–AI interaction technique that operationalizes adaptive questioning as a decision intelligence mechanism. The interface dynamically reveals questions based on real-time belief updates and uncertainty reduction, thereby aligning information elicitation with decision relevance. Grounded in Bayesian decision theory and human-centered design principles, the study develops a formal framework, algorithms, and interface logic for adaptive questioning. An analogy-based case study demonstrates that the proposed interface reduces cognitive load, minimizes redundant questioning, and improves decision confidence compared to static and linear interfaces. The paper contributes a new class of decision intelligence interfaces that integrate decision optimization with human-centered interaction design.