Study type: Secondary analysis of clinical experience-sampling data.
In plain language
Psychological flexibility is generally associated with better well-being. But does each flexibility process relate to mood in the same way for each person, and does that relationship stay stable through treatment? Yee and colleagues examined those questions using repeated daily-life assessments from 103 psychiatric patients who received acceptance and commitment therapy (ACT) in Switzerland.
Participants reported mood and six psychological flexibility processes during two assessment weeks, around the beginning and end of treatment. The researchers estimated relationships within individual patients using idiographic time-series models, then compared Bayesian models that treated these relationships as shared or allowed them to vary across people and assessment periods.
The aggregate results were positive: greater flexibility was associated with better mood. The individual analyses added a different level of information. Associations varied in strength and sometimes direction, differed across flexibility processes, and could change between the two assessment periods. A process relevant to one patient’s mood could be less relevant, unrelated, or differently related for another patient.
Key points
- Individual variation improved model performance. A Bayesian random-effects model predicted the observed process–mood estimates better than a fixed-effect model, supporting non-random heterogeneity.
- Variation differed by process. Present-moment awareness, values, and committed action showed especially clear additional person-level variation relative to acceptance, the reference process. A small contrast with acceptance did not imply that all people shared the same association.
- Relationships could change over time. Forest plots of committed action illustrated stable, strengthening, weakening, and occasionally negative associations across the two assessment periods, even when pooled estimates appeared similar.
- Personalization requires ongoing information. The findings support examining a person’s own process–outcome patterns and reassessing them over time, rather than assuming a population-average relationship remains valid for that person.
What the paper does and does not establish
- These were associations in an existing treatment dataset. The study did not establish that increasing a specific flexibility process causes improved mood, or that using the models improves clinical outcomes.
- There was no sampled no-treatment control, and treatment was not randomized or standardized for this analysis. Time, treatment, client characteristics, and contextual changes could contribute to changes in the associations.
- Some individual series were short, limiting precision. The authors describe subgroup patterns visually and identify formal subgroup tests and time-lagged analyses as future work.
How to cite
APA 7
Yee, H. F., Fraser, M. I., Ciarrochi, J., Sahdra, B., Yap, K., Hayes, S. C., Hernández, C., & Gloster, A. T. (2026). Beyond the average: Idionomic evidence of individual-specific variability in the psychological flexibility-mood relationship. Journal of Contextual Behavioral Science, 40, Article 100993. https://doi.org/10.1016/j.jcbs.2026.100993
BibTeX
@article{yee2026beyond,
title = {Beyond the average: Idionomic evidence of individual-specific variability in the psychological flexibility-mood relationship},
author = {Yee, Hannah F. and Fraser, Madeleine I. and Ciarrochi, Joseph and Sahdra, Baljinder and Yap, Keong and Hayes, Steven C. and Hernández, Cristóbal and Gloster, Andrew T.},
journal = {Journal of Contextual Behavioral Science},
year = {2026},
volume = {40},
pages = {100993},
doi = {10.1016/j.jcbs.2026.100993}
}
Related work
- Heterogeneity in psychological flexibility and well-being
- One size fits none: A call for an idionomic revolution
- Time-series machine learning for individual treatment planning
- All publications by Joseph Ciarrochi
Summary based on the full published article supplied by the author. The local PDF is the publisher version, published online 10 March 2026. Summary prepared from the full paper on 29 September 2026. Please cite the published article linked above.