Dataset aligns counseling talks with cognitive behavioral therapy methods
Steering the Compass: Aligning Dynamic Psychological Counseling Conversations with Cognitive Behavioral Therapy Strategies
Computation and Language
Summary
Counseling that uses Cognitive Behavioral Therapy (CBT) depends on therapists choosing the right approach during conversations. Existing research often ignores how therapists adjust their strategies based on what the client is thinking or feeling in the moment. The authors created a large set of counseling conversations called StratCBT, where each counselor response matches one of eight CBT strategies. They show their dataset helps computers generate counseling replies that better reflect professional CBT approaches, tested with simulated clients to mimic real counseling.
What this means in practice
- •For mental health chatbot developers: Create counseling chatbots that select and use CBT strategies effectively during sessions, improving conversation quality and therapy support.$Commercial implications: Enables products delivering automated CBT counseling with strategy-aware responses useful for mental health app providers.
- •For clinical training platforms: Develop training tools that simulate therapy sessions with strategy-tagged counselor replies to help therapists practice CBT techniques.
Authors
Zimu Wang, Yiwen Jiang, Xiangyu Zhao, Yaling Shen, Jiahe Liu, Stephanie Fong, Maxmartwell H Cheng, Guilherme C Oliveira, Anh Nguyen, Robert Desimone, Barnaby Nelson, Dominic Dwyer, Zongyuan Ge
Abstract
Recent advancements in large language models have revolutionized the field of psychological counseling, especially in the context of Cognitive Behavioral Therapy (CBT). While the success of CBT relies heavily on dynamic decision-making informed by the client's real-time mental state, this aspect has often been overlooked in current research, limiting both flexibility and therapeutic outcomes. In this paper, we introduce StratCBT, a dataset specifically designed for psychological counseling conversations with CBT Strategies, consisting of 9,688 sessions and around 256K utterances, with each counselor's response aligned with one of eight distinct strategies. The creation of StratCBT involves modeling clients based on their negative thoughts and generating high-quality counseling conversations through self-chat, incorporating realistic sessions as guidance, thereby significantly surpassing existing datasets in both general counseling and CBT-specific skills. We conduct extensive experiments to demonstrate the effectiveness of strategy-aligned generation and evaluate its efficacy in delivering professional and effective counseling with LLM-simulated clients to reflect real-world scenarios. The dataset can be obtained from https://github.com/zimuwangnlp/StratCBT.