Stefan G. Hofmann and Steven C. Hayes, Clinical Psychological Science 7(1):37–50, January 2019 (online 29 May 2018) — DOI-10.1177/2167702618772296
Author order. The brief gives “Hayes, S. C. & Hofmann, S. G.”. The
article, Crossref and PubMed all list Hofmann first, so first_author: is
Hofmann. The confusion is natural: the two also wrote the book
Process-Based CBT (2018) with Hayes first. Read from the author
manuscript in PubMed Central (PMC6350520, NIHMS956619), which carries no
journal page numbers.
Key takeaways
- The medical illness model behind the DSM assumes symptoms reflect latent disease entities. The authors argue that a syndrome-based research programme (roughly 1980–2010) “never led to the discovery of diseases”, and that comorbidity and heterogeneity made diagnosis feel “more like an empty ritual”.
- Process-based therapy (PBT) is “the contextually specific use of evidence-based processes linked to evidence-based procedures to help solve the problems and promote the prosperity of particular people”. Processes (the change mechanisms) are distinguished from procedures (the techniques).
- The test of an evidence-based intervention changes. A procedure enters PBT only if it “reliably produces gains and manipulates a process that mediates these gains”. If a model’s specified process does not hold up, the model “is incorrect and needs to be set aside” even if the intervention works.
- The predicted consequences are a decline of named therapies and broad schools, more mediation and moderation studies, diagnosis by functional analysis and idiographic networks, and the term CBT losing its distinctness.
Standing in the record
Filed on 2026-10-03 at the owner’s request, as the read statement of the
process-based, transdiagnostic view of how treatment works that the
psychopathology-and-treatment blurb names. Social science is tagged
because this is clinical psychology’s programme for its own science, and
philosophy of science because much of the argument is about what counts as
evidence for a model of change and about the field’s pre-analytic
assumptions. It carries no instruction for machine-learning practice. The
Anthology of the SOTA does not hold it (a grep of its literature.d,
notes.d and theory.d for “process-based”, “Hofmann” with “Hayes”, and the
DOI found nothing).
It bears on the batch:
- Salkovskis, Sighvatsson and Sigurdsson (LIT-538) start from the same Gordon Paul question and reach a similar diagnosis, that the field has studied efficacy and neglected mechanisms. They then make the opposite bet: one mechanism of change for all effective therapies, not a plurality of processes. Neither paper cites the other.
- Borsboom (LIT-547) gives the network theory these authors invoke (through Hofmann, Curtiss and McNally 2016) as the “analytic alternative to the latent disease model”.
- Engel (LIT-548) named the biopsychosocial model, and this paper’s foundational question asks for “core biopsychosocial processes”. Engel is not cited.
No instruction for machine-learning practice; nothing here belongs in the anthology.