BACKGROUND

Rumination has been identified as a potential mechanism of therapeutic change, particularly in directive and focused psychotherapies for depression. Previous research has predominantly focused on either trait-like individual differences or state-like changes in rumination without integrating these aspects. In the current study, we propose a computational approach to investigating whether rumination serves as a compensatory or a resilience mechanism by integrating trait-like and state-like effects.

METHODS

Rumination and depressive symptoms were assessed (in  = 100) pretreatment and repeatedly throughout treatment. Mixed-level models were used to examine whether pretreatment trait-like rumination interacted with a time-variant variable of in-treatment state-like changes in rumination to predict subsequent changes in treatment outcomes. These models were used to determine whether individuals with higher or lower pretreatment trait-like levels of rumination benefited more from state-like reductions in rumination, thus contrasting the compensatory and resilience theoretical frameworks.

RESULTS

As hypothesized, the findings supported the compensatory framework; individuals with higher pretreatment trait-like levels of rumination benefited most from greater state-like reductions in rumination during treatment, as evidenced by greater subsequent symptom reduction ( = .04).

CONCLUSIONS

The findings refine our understanding of rumination as an individual-specific mechanism of therapeutic change, dependent on an individual's trait-like levels of rumination. The proposed computational approach enabled an empirical comparison of the 2 main theoretical frameworks of treatment personalization, compensatory and resilience, offering new insights into mechanisms that drive therapeutic change. Future studies could leverage the paradigm proposed here to examine for which patients and in what contexts mechanisms of change function as compensatory versus resilience mechanisms.