# Feel the Difference? A Comparative Analysis of Emotional Arcs in Real and LLM-Generated CBT Sessions

## TL;DR

This study compares emotional dynamics in real and AI-generated CBT sessions, revealing that synthetic dialogues lack the emotional variability and authenticity present in real therapy conversations, highlighting limitations of current LLMs.

## Contribution

Introduces the RealCBT dataset and provides the first detailed analysis of emotional arcs in real versus synthetic CBT dialogues using the Utterance Emotion Dynamics framework.

## Key findings

- Real sessions show greater emotional variability.
- Synthetic dialogues are fluent but less emotionally rich.
- Low emotional arc similarity between real and synthetic conversations.

## Abstract

Synthetic therapy dialogues generated by large language models (LLMs) are increasingly used in mental health NLP to simulate counseling scenarios, train models, and supplement limited real-world data. However, it remains unclear whether these synthetic conversations capture the nuanced emotional dynamics of real therapy. In this work, we introduce RealCBT, a dataset of authentic cognitive behavioral therapy (CBT) dialogues, and conduct the first comparative analysis of emotional arcs between real and LLM-generated CBT sessions. We adapt the Utterance Emotion Dynamics framework to analyze fine-grained affective trajectories across valence, arousal, and dominance dimensions. Our analysis spans both full dialogues and individual speaker roles (counselor and client), using real sessions from the RealCBT dataset and synthetic dialogues from the CACTUS dataset. We find that while synthetic dialogues are fluent and structurally coherent, they diverge from real conversations in key emotional properties: real sessions exhibit greater emotional variability, more emotion-laden language, and more authentic patterns of reactivity and regulation. Moreover, emotional arc similarity remains low across all pairings, with especially weak alignment between real and synthetic speakers. These findings underscore the limitations of current LLM-generated therapy data and highlight the importance of emotional fidelity in mental health applications. To support future research, our dataset RealCBT is released at https://gitlab.com/xiaoyi.wang/realcbt-dataset.