ODRA: Synthesizing Cognitive Behavioral Therapy Sessions with Structured Chain-Of-Thought and Dynamic Patient Resistance

2026-08-05

AI Summary

This work addresses the limitation of existing virtual patients, which often lack clinical fidelity due to excessive compliance and fail to balance the structured requirements of cognitive behavioral therapy (CBT) with dynamic patient resistance. To overcome this, the authors propose the ODRA framework, which uniquely integrates a Beck CBT–guided structured chain-of-thought with an adjustable patient resistance mechanism. By leveraging language model fine-tuning and behavior alignment techniques, ODRA generates high-fidelity CBT dialogues. Experimental results demonstrate that the approach significantly outperforms baseline methods across both automated metrics and expert evaluations. Licensed therapists expressed a preference for ODRA-generated interactions on 12 out of 13 clinical dimensions, and the system exhibited superior therapeutic robustness and clinical authenticity with both cooperative and resistant virtual patients.

Abstract

Synthetic generation of Cognitive Behavioral Therapy (CBT) sessions is challenged by two competing demands: adhering to strict therapeutic structure while modeling the resistant, unpredictable behavior of real patients. Existing script-based methods fail to capture dynamic therapeutic interactions, while multi-agent approaches struggle to adhere to CBT's sequential structure; both suffer from sycophancy, producing overly compliant patients that misrepresent real clinical settings. In this work we introduce ODRA, a novel framework for synthesizing therapy dialogues through a Chain-of-Thought (CoT) strategy grounded in foundational CBT guidelines (Beck, 2020). ODRA further incorporates a resistance orchestrator to solve patient sycophancy, which employs steering techniques to elicit behaviors aligned with their resistance level. Automated and expert evaluations show that ODRA significantly outperforms existing methods across therapeutic skills, CBT alignment, and patient behavioral fidelity, with licensed psychologists preferring ODRA sessions across 12 of 13 clinical metrics. Furthermore, models fine-tuned on our dataset demonstrate superior therapeutic performance against both cooperative and resistant patients, validating that explicit resistance modeling in synthetic training data directly translates to downstream clinical robustness.

Problem

• Cognitive Behavioral Therapy
• patient resistance
• synthetic dialogue generation
• therapeutic structure
• behavioral fidelity

Innovation

• Chain-of-Thought
• patient resistance
• cognitive behavioral therapy
• synthetic dialogue generation
• therapeutic alignment