Answers are grounded in this paper and cite their sources.
The Hierarchical Taxonomy of Psychopathology (HiTOP) is thought to be scientifically and clinically useful because it aligns with the natural structure of psychopathology. However, because HiTOP is based on the analyses of standardized symptom criteria and closed-ended survey items, it may just reflect the structure of available measures. This study tests this possibility by examining whether the structure of similarities in language used by online communities when talking about their mental health corresponds to HiTOP. For our analysis, we obtained 136,667 naturalistic descriptions of mental health from >72,380 unique users on 27 online Reddit communities (i.e., subreddits). We used text embeddings in large language models to represent common themes of posts in the subreddits, each dedicated to a different diagnosis or maladaptive trait. Semantic similarity across subreddit language was quantified with a cosine similarity matrix. We then fit exploratory factor models to this matrix and examined the hierarchical structure with a bass-ackwards approach. Results showed that the structure of psychopathology discourse closely corresponds to the HiTOP model, with factors resembling internalizing, detachment, thought disorder, substance use, antagonism, and somatoform spectra at the lowest level of the hierarchy. The higher order factors diverged from HiTOP, including a general factor of problem construal reflecting variation in attribution of problems to internal versus external sources. Overall, the emergence of HiTOP spectra from natural language offers unique evidence for its generalizability and clinical utility. (PsycInfo Database Record (c) 2026 APA, all rights reserved).
No takes yet. Share an insight, caveat, or question.
Ringwald et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: