Findings suggest that cognitive therapy (CT), mindfulness-based stress reduction (MBSR), and behavior therapy (BT) for chronic pain produce improvements through changes in putative mechanisms. Evidence supporting this notion is largely based on findings showing significant associations between treatment mechanism variables and outcomes. An alternative view is that treatments may work by reducing or decoupling the impact of changes in mechanism variables on changes in outcomes. We examined the degree to which relationships between previous changes in potential treatment mechanisms and subsequent changes in outcomes changed as treatment progressed and vice versa. Cognitive therapy, MBSR, BT, and treatment as usual (TAU) were compared in people with chronic low back pain (N = 521). Eight individual sessions were administered with weekly assessments of putative treatment mechanisms and outcomes. Lagged analyses revealed mechanism × session number interactions and outcome × session number interactions, such that associations between mechanism and outcome variables were strong and significant in the first third of treatment, but weakened over time and became nonsignificant by the last third of treatment. These effects were similar across treatment conditions but did not emerge among people undergoing TAU. Results suggest that during the course of CT, MBSR, and BT, the links between changes in treatment mechanism variables became decoupled from subsequent changes in outcomes and vice versa. Thus, starting by midtreatment and continuing into late treatment, participants may have learned through participation in the treatments that episodes of maladaptive pain-related thoughts and/or spikes in pain need not have detrimental consequences on their subsequent experience.