metadata
task_categories:
- text-classification
language:
- en
tags:
- psychological defense mechanism
- supportive conversations
- mental health
- AI for mental health
pretty_name: PsyDefConv
size_categories:
- 1K<n<10K
license: cc-by-nc-4.0
PsyDefConv
PsyDefConv is the first conversational dataset annotating Defense Mechanism Rating Scale (DMRS) defense levels at the utterance level, built on top of ESConv.
📄 Paper: You Never Know a Person, You Only Know Their Defenses: Detecting Levels of Psychological Defense Mechanisms in Supportive Conversations (ACL 2026)
Overview
- 200 dialogues / 2,336 seeker utterances labeled with defense levels
- 9 labels: 7 DMRS levels +
No Defense(0) +Needs More Information(8) - Double-blind annotation, Cohen's κ = 0.639 (substantial)
- 80/20 stratified split (train / test)
Label Scheme
| Label | Name |
|---|---|
| 0 | No Defense |
| 1 | Action |
| 2 | Major Image-Distorting |
| 3 | Disavowal |
| 4 | Minor Image-Distorting |
| 5 | Neurotic |
| 6 | Obsessional |
| 7 | High-Adaptive |
| 8 | Needs More Information |
See the paper for full definitions and examples.
Citation
@inproceedings{na-etal-2026-psydefconv,
title = "You Never Know a Person, You Only Know Their Defenses:
Detecting Levels of Psychological Defense Mechanisms
in Supportive Conversations",
author = "Na, Hongbin and Wang, Zimu and Chen, Zhaoming and
Zhou, Peilin and Hua, Yining and Zhou, Grace Ziqi and
Zhang, Haiyang and Shen, Tao and Wang, Wei and
Torous, John and Ji, Shaoxiong and Chen, Ling",
booktitle = "Findings of the Association for Computational
Linguistics: ACL 2026",
month = jul,
year = "2026",
address = "San Diego, USA",
publisher = "Association for Computational Linguistics",
}
Please also cite ESConv (Liu et al., 2021).