Explaining anxiety prediction in psychotherapy transcripts: The role of patient linguistic features and theoretical constructs
Our findings highlight a trade-off between performance and interpretability. Unigram-bigram models maximized predictive accuracy, whereas theory-driven constructs provided clinically meaningful insights into core psychological processes. Identifying predictive linguistic features, especially those linked to psychological theory, may guide future research on feedback systems and clinical applications by providing interpretable and theory-aligned insights.
Authors
- Jana Schaffrath (ORCID: https://orcid.org/0000-0003-0106-0243)
- Christopher Lalk (ORCID: https://orcid.org/0000-0002-1793-3507)
- Steffen T. Eberhardt (ORCID: https://orcid.org/0000-0002-9900-4671)
- Julian Rubel (ORCID: https://orcid.org/0000-0002-9625-6611)
- Wolfgang Lutz (ORCID: https://orcid.org/0000-0002-5141-3847)
- Tobias Steinbrenner
- Anke Haberkamp
- Kim Targan
Institutions
- Trier University of Applied Sciences (DE)
- Witten/Herdecke University (DE)
- Osnabrück University (DE)
- Hochschule Osnabrück (DE)
- Universität Trier (DE)
Publication Details
- Journal
- Behaviour Research and Therapy
- Published
- 2025-09-09
- DOI
- https://doi.org/10.1016/j.brat.2025.104857
- Primary Topic
- Mental Health via Writing
- Type
- article
- Field-Weighted Citation Impact
- 0.00