PSYCHE: A Multi-faceted Patient Simulation Framework for Evaluation of Psychiatric Assessment Conversational Agents
This addresses the need for clinically relevant and ethical evaluation methods for conversational agents in psychiatry, though it is incremental as it builds on existing simulation approaches.
The authors tackled the lack of standardized benchmarking for psychiatric assessment conversational agents (PACAs) by proposing PSYCHE, a framework for simulating psychiatric patients to evaluate PACAs, validated with 10 board-certified psychiatrists.
Recent advances in large language models (LLMs) have accelerated the development of conversational agents capable of generating human-like responses. Since psychiatric assessments typically involve complex conversational interactions between psychiatrists and patients, there is growing interest in developing LLM-based psychiatric assessment conversational agents (PACAs) that aim to simulate the role of psychiatrists in clinical evaluations. However, standardized methods for benchmarking the clinical appropriateness of PACAs' interaction with patients still remain underexplored. Here, we propose PSYCHE, a novel framework designed to enable the 1) clinically relevant, 2) ethically safe, 3) cost-efficient, and 4) quantitative evaluation of PACAs. This is achieved by simulating psychiatric patients based on a multi-faceted psychiatric construct that defines the simulated patients' profiles, histories, and behaviors, which PACAs are expected to assess. We validate the effectiveness of PSYCHE through a study with 10 board-certified psychiatrists, supported by an in-depth analysis of the simulated patient utterances.