Optimal Question Selection from a Large Question Bank for Clinical Field Recovery in Conversational Psychiatric Intake

2026-04-23

AI Summary

This study addresses the challenge of efficiently eliciting critical clinical information during initial psychiatric interviews, particularly when patients provide vague, brief, or evasive responses. The authors present the first large-scale, structured benchmark for question selection in psychiatric intake, comprising 655 clinical questions and synthetically generated patient cases that incorporate a novel dimension of patient behavioral difficulty. Evaluating diverse questioning strategies across 300 simulated interviews, the experiments demonstrate that large language model (LLM)-guided adaptive questioning significantly outperforms both random selection and standard questionnaire approaches in recovering clinical information. This advantage is especially pronounced when patients exhibit guarded and terse communication styles, underscoring the pivotal role of question sequencing and adaptivity in effective clinical information gathering.

Technology Category

• Natural Language Processing: Question Answering
• Cognitive Modeling & Cognitive Systems: Adaptive Behavior
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Search and Retrieval-Augmented AI: Web question answering
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Psychiatric intake is a sequential, high-stakes information-gathering process in which clinicians must decide what to ask, in what order, and how to interpret incomplete or ambiguous responses under limited time. Despite growing interest in conversational AI for healthcare, there is still limited infrastructure for conversational AI in this application. Accordingly, we formulate this task as a question-selection problem with clinically grounded questions, known target information, and controllable patient difficulty. We also introduce a task-specific question-selection benchmark based on a bank of 655 clinician-authored intake questions and corresponding synthetic patient vignettes with 5 different behavioral conditions. In our evaluation, we compare random questioning, a clinical psychiatric intake form baseline, and an LLM-guided adaptive policy across 300 interview sessions spanning four patients and five behavioral conditions. Across the benchmark, the clinically ordered fixed form substantially outperforms random questioning, and the LLM-guided policy achieves the strongest overall recovery. The advantage of adaptation grows sharply under patient behavior that is less amenable to field recovery, especially under guarded-concise conditions. These findings suggest that performance in conversational clinical systems depends not only on language understanding after information is disclosed, but also on whether the system reaches the right topics within a limited interaction budget. More broadly, the benchmark provides a controlled framework for studying how clinical structure and adaptive follow-up contribute to information recovery in interactive clinical machine learning.

Problem

• question selection
• clinical information recovery
• conversational AI
• psychiatric intake
• adaptive questioning

Innovation

• question selection
• conversational AI
• clinical intake
• adaptive questioning
• information recovery