What is Stigma Attributed to? A Theory-Grounded, Expert-Annotated Interview Corpus for Demystifying Mental-Health Stigma

TL;DR

This paper introduces a theory-grounded, expert-annotated interview corpus to improve neural models' ability to detect and understand mental-health stigma, addressing limitations of existing social-media-based resources.

Contribution

It provides a novel, theory-informed dataset of human-chatbot interviews with expert annotations, enabling better computational detection and analysis of mental-health stigma.

Findings

Benchmarking state-of-the-art neural models on the dataset

Empirical analysis of challenges in stigma detection

Open availability of the corpus for future research

Abstract

Mental-health stigma remains a pervasive social problem that hampers treatment-seeking and recovery. Existing resources for training neural models to finely classify such stigma are limited, relying primarily on social-media or synthetic data without theoretical underpinnings. To remedy this gap, we present an expert-annotated, theory-informed corpus of human-chatbot interviews, comprising 4,141 snippets from 684 participants with documented socio-cultural backgrounds. Our experiments benchmark state-of-the-art neural models and empirically unpack the challenges of stigma detection. This dataset can facilitate research on computationally detecting, neutralizing, and counteracting mental-health stigma. Our corpus is openly available at https://github.com/HanMeng2004/Mental-Health-Stigma-Interview-Corpus.