Language or syndrome? Investigating the origins of the general psychopathology factor through large language model embeddings
Psychiatry and Clinical Neurosciences · 2025
DOI: 10.1111/pcn.70010 ↗Kurz gefasst
Fachartikel in der Zeitschrift Psychiatry and Clinical Neurosciences, 2025. Verfasst von Hiroki Kojima und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Information Medicine, National Institute of Neuroscience, National Center of Neurology and Psychiatry, Tokyo, Japan. Hauptthemen laut MEDLINE: Mental Disorders, Language, Semantics, Psychopathology. Die Arbeit nennt 51 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: National Center of Neurology and Psychiatry, Japan Society for the Promotion of Science.
Abstract (Originalsprache)
Aims
Psychiatric diagnostic categories are challenged by comorbidity, symptom heterogeneity, and blurred nosological boundaries, sparking interest in trans‐diagnostic dimensions such as the general psychopathology factor (p‐factor). Yet it remains unclear whether p‐factor represents a valid psychopathological construct or merely echoes semantic overlap among questionnaire items. Leveraging recent advances in large language models (LLMs), we conduct an empirical investigation that aims to disentangle linguistic similarity from syndromic co‐occurrence in the p‐factor.
Methods
We analyzed 145 symptom items spanning eight diagnostic categories with two similarity metrics: (1) Semantic—cosine similarity of LLM embeddings; and (2) Syndromic—polychoric correlations of responses from 985 community participants. Exploratory factor analyses extracted latent structures, their alignment was quantified, and factor scores were tested against lifetime psychiatric diagnoses.
Results
Both similarity matrices converged on a dominant p‐factor with high cross‐diagnostic loadings. Loading patterns were strongly congruent, indicating that LLM‐derived semantics alone recapitulate much of the latent structure in human data. Crucially, syndromic p‐factor scores predicted individual diagnostic histories more accurately than semantic p‐factor scores, revealing clinically informative variance beyond language.
Conclusions
LLM embeddings reveal that the linguistic fabric of symptom items contains a shadow of the p‐factor, yet real‐world co‐occurrence embeds additional, diagnosis‐relevant information. Our findings position LLM‐driven text analytics as a new lens for probing psychiatric constructs while underscoring the complex interplay between linguistic semantics and symptom co‐occurrence in shaping our understanding of psychiatric constructs.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Information Medicine, National Institute of Neuroscience, National Center of Neurology and Psychiatry, Tokyo, Japan.
ORCID-Profile: Takafumi Soda, Yuichi Yamashita
Förderung
- National Center of Neurology and Psychiatry
- Japan Society for the Promotion of Science
Zitieren
APA
Kojima, H., Soda, T., & Yamashita, Y. (2025). Language or syndrome? Investigating the origins of the general psychopathology factor through large language model embeddings. Psychiatry and Clinical Neurosciences. https://doi.org/10.1111/pcn.70010
Vancouver
Kojima H, Soda T, Yamashita Y. Language or syndrome? Investigating the origins of the general psychopathology factor through large language model embeddings. Psychiatry and Clinical Neurosciences. 2025. doi:10.1111/pcn.70010
BibTeX
@article{Kojima2025,
author = {Kojima, Hiroki and Soda, Takafumi and Yamashita, Yuichi},
title = {Language or syndrome? Investigating the origins of the general psychopathology factor through large language model embeddings},
journal = {Psychiatry and Clinical Neurosciences},
year = {2025},
doi = {10.1111/pcn.70010}
}
EUVIMED