COCOA: CBT-based Conversational Counseling Agent using Memory Specialized in Cognitive Distortions and Dynamic Prompt

TL;DR

CoCoA is a conversational counseling agent that uses CBT techniques, memory management, and dynamic prompting to effectively identify and address cognitive distortions in mental health dialogues.

Contribution

The paper introduces a novel CBT-based counseling agent with a memory system and dynamic prompts, enhancing response appropriateness and insight extraction in mental health conversations.

Findings

CoCoA outperforms other models in counseling dialogue evaluation.

The memory system improves information management during therapy sessions.

Dynamic prompting enables flexible application of CBT techniques.

Abstract

The demand for conversational agents that provide mental health care is consistently increasing. In this work, we develop a psychological counseling agent, referred to as CoCoA, that applies Cognitive Behavioral Therapy (CBT) techniques to identify and address cognitive distortions inherent in the client's statements. Specifically, we construct a memory system to efficiently manage information necessary for counseling while extracting high-level insights about the client from their utterances. Additionally, to ensure that the counseling agent generates appropriate responses, we introduce dynamic prompting to flexibly apply CBT techniques and facilitate the appropriate retrieval of information. We conducted dialogues between CoCoA and characters from Character.ai, creating a dataset for evaluation. Then, we asked GPT to evaluate the constructed counseling dataset, and our model demonstrated a statistically significant difference from other models.