Within-person prediction of differential responding to perseverative thought disengagement strategies

Project Summary/Abstract Perseverative (repetitive) thought, an umbrella term that encapsulates thoughts such as worry and rumination, is a major transdiagnostic symptom and mechanism of internalizing psychopathology. Perseverative thought varies in expression both between- and within-person; for example, in terms of its uncontrollability, valence, and abstractness. Although the uncontrollability of perseverative thought explains most of the variance in its relationship to adverse mental health outcomes, other features (e.g., valence, abstractness) have also been identified as maladaptive. Strategies for regulating perseverative episodes typically operate on one or more of these features; for example, aiming to increase concreteness or positive valence. Notably, there is little empirical guidance to suggest which kinds of strategies should be used given a particular set of thought features. Instead, “matching” decisions (which strategy to implement for which thoughts) are typically grounded in clinical experience or individual trial-and-error. In line with the NIMH Strategic Plan Objective 3.2, we will apply principles of precision medicine (“what works for whom?”) at the within-person level (“what works and when?”). Specifically, we will leverage mobile digital technologies and quantitative methods such as machine learning to test as well as generate within-person predictions regarding the efficacy of four evidence-based regulatory strategies (cognitive reappraisal; focused attention meditation; savoring; absorption practice) as a function of the features of the thought being regulated (negative valence; abstractness; verbal vs visual properties; temporal orientation; passiveness). Participants will be 200 community-dwelling adults who report elevated perseverative thought. In an ecological momentary intervention design, participants will be signaled four times per day for 20 days (80 signals total) to: 1) rate their current mood, impairment, and thought features; 2) complete one of the four active strategies or a self-monitoring control (counterbalanced; within-person design); and 3) re-rate their mood, impairment, and selected thought features. We will test several theory-driven hypotheses regarding differential immediate (same signal) and sustained (next signal) efficacy of thought strategy as a function of momentary thought features (rated pre-strategy). We will also use machine learning to examine the extent to which we can predict the efficacy of a particular strategy from thought feature ratings obtained immediately prior to completing the strategy. Finally, we will conduct preliminary exploratory analyses probing potential moderators at the between-person level. Findings from this work will have direct clinical utility and relevance for theoretical models of perseverative thought. This work will also provide a necessary foundation for the development of personalized just-in-time adaptive interventions that can tailor strategy recommendations on an individual basis in real-time.