In plain language
Most mental health research follows a “protocol-for-syndrome” model: take a diagnosis from the DSM, then test a fixed treatment package against it. This paper argues that model has two big problems. First, treatment protocols bundle together many change processes, some of which are irrelevant to a given individual. Second, DSM syndromes are too crude to capture the nuances of an individual person’s suffering and well-being.
As an alternative, the authors lay out process-based therapy (PBT), which focuses on the specific processes of change that matter for a particular client. They present a slightly revised version of the extended evolutionary metamodel (EEMM) — a shared language for describing change processes in terms of the evolutionary drivers of variation, selection, and retention, organized across psychological dimensions (such as cognition, affect, self, motivation, attention, and overt behavior) and levels (from biology and physiology through psychology to social relationships and culture). They then tested whether people and artificial intelligence could actually use this common language, training AI models (based on distilBERT) to classify therapeutic content into EEMM categories.
Human raters classified therapy-relevant questionnaire content with good reliability, and the fine-tuned AI models generalized well to unseen material — the dimensions model mislabeled only 9% of labels and the levels model only 5% on test data. The tool was released as a free web app so the community can use and improve it. Combined with network theory and new analytic methods, this framework gives therapists from different traditions a common ground for communicating, cooperating, and tailoring therapy to the individual — and, the authors suggest, opens the door to a potential unification of psychotherapy.
Key findings
- Critiques the dominant “protocol-for-syndrome” model on two grounds: protocols include change processes irrelevant to many individuals, and DSM syndromes fail to capture the nuances of individual suffering and well-being.
- Presents a slightly revised extended evolutionary metamodel (EEMM) as a consistent common language for change processes, built on variation, selection, and retention across dimensions (cognition, affect, self, motivation, attention, overt behavior, plus context) and levels (biology/physiology, psychology, social/culture).
- Human raters used the EEMM to classify therapeutic processes reliably: KR-20 reliabilities ranged from .73 (context) to .90 (overt behavior), with cross-rater-group correlations for levels up to r = .91.
- Fine-tuned distilBERT AI models accurately classified therapeutic content into EEMM categories, mislabeling only 9% of dimension labels and 5% of level labels on unseen test data, and were deployed as a free public web app.
- Shows how network theory and new analytics can help therapists customize therapy to the needs of individual clients rather than diagnostic categories.
- Argues that PBT supports therapeutic diversity while creating common ground across schools of therapy, enhancing communication, cooperation, and comparison — and potentially enabling the unification of psychotherapy.
How to cite
APA
Ciarrochi, J., Hernández, C., Hill, D., Ong, C., Gloster, A. T., Levin, M. E., Yap, K., Fraser, M. I., Sahdra, B. K., Hofmann, S. G., & Hayes, S. C. (2024). Process-based therapy: A common ground for understanding and utilizing therapeutic practices. Journal of Psychotherapy Integration, 34(3), 265–290. https://doi.org/10.1037/int0000348
BibTeX
@article{ciarrochi2024process,
title = {Process-based therapy: A common ground for understanding and utilizing therapeutic practices},
author = {Ciarrochi, Joseph and Hern{\'a}ndez, Crist{\'o}bal and Hill, Diana and Ong, Clarissa and Gloster, Andrew T. and Levin, Michael E. and Yap, Keong and Fraser, Madeleine I. and Sahdra, Baljinder K. and Hofmann, Stefan G. and Hayes, Steven C.},
journal = {Journal of Psychotherapy Integration},
year = {2024},
volume = {34},
number = {3},
pages = {265--290},
doi = {10.1037/int0000348}
}
Related work
- All publications by Joseph Ciarrochi (searchable, with free PDFs)
- Process-Based Therapy & Idionomic Analysis
Author: Joseph Ciarrochi (ORCID 0000-0003-0471-8100). Free copy hosted with permission for scholarly use. Please cite the published version via the DOI above.