Abstract Objectives The primary aim was to explore state- and trait-level effects and candidate mechanisms of four Mindfulness-Based Cognitive Therapy (MBCT) practices. Method One hundred sixty adults self-selected from the general population were randomized to one of four mindfulness practices: body scan, mindful movement, breath and body, and befriending. Study 1 explored state-level self-compassion, mindfulness, decentering (mechanisms), and pleasantness of thoughts, emotions, and body sensations at multiple time points using two single mindfulness sessions. Study 2 explored trait-level self-compassion, mindfulness, decentering, interoceptive awareness, attentional control (mechanisms), anxiety, depression, and psychological quality of life pre-post 2 weeks of daily practice. Results In study 1, state-level effects were demonstrated in all candidate mechanisms and outcomes within the whole sample across time points (d = 0.27 to 0.86), except for state decentering. After controlling for pre-scores and additional covariates, no between-group effects were found (p = 0.050 to 0.973). In study 2, trait-level effects were demonstrated in psychological quality of life and most candidate mechanisms within the whole sample (d = 0.26 to 0.64) but no between-group effects were found (p = 0.080 to 0.805). Within the whole sample, after controlling for pre-scores, changes in mindfulness, self-compassion, decentering, and interoceptive awareness (i.e. body listening) were associated with improvements in psychological quality of life (r = 0.23 to 0.40) and self-led mindfulness practice (r = 0.18 to 0.23). Conclusions Future research should test the generated hypotheses using well-designed, adequately powered, and theory-driven studies that address universal and specific mechanisms in different populations and contexts. Pre-registration This study is not pre-registered.
Springer Science and Business Media LLC