Top-Down and Bottom-Up Contributions to Memory Performance in OCD: A Multilevel Meta-Analysis With Clinical Implications.

Despite extensive coverage of a relationship between memory performance and executive function in the obsessive-compulsive disorder (OCD) literature, the relative contributions of specific aspects of executive control have remained elusive. We, therefore, extend our previous multilevel meta-analysis (Persson et al., 2021), where demand on executive function was the most significant determinant of memory deficits in OCD, and provide a finer-grained analysis of executive control via a segregation into top-down (attentional control, maintenance and updating, planning) and bottom-up (perceptual integration, perceptual salience) contributions. Our multilevel meta-analytic approach allowed us to accommodate the interdependency of 255 effect sizes from 131 studies, totaling 4,101 OCD patients. Results revealed that maintenance and updating (top-down) and perceptual integration (bottom-up) predicted memory performance generally, and specifically in those with clinical OCD. Exploratory analyses suggested that this effect may be somewhat different among subclinical OCD groups; however, these findings should be considered with conceptual and analytical caveats in mind. We explain these results via deficient sensory (perceptual integration) and working memory (maintenance and updating) gating mechanisms and propose a model to accommodate their expression in OC symptoms. In conclusion, our meta-analysis has expanded understanding of cognitive performance in OCD and identifies the possibility of untapped cognitive targets for intervention. General Scientific Summary: Deficits in executive function and memory are common in OCD. We identify and test two novel frameworks, top-down and bottom-up, to explain memory deficits in OCD using a multilevel meta-analysis. We found that maintenance and updating, as well as perceptual integration, predicted memory deficits in OCD, where in turn we explained these via deficient sensory and working memory (WM) gating mechanisms, respectively. Our meta-analytical results and explanations may inform future novel interventions. [ABSTRACT FROM AUTHOR]