Childhood adversity and impaired reward processing: A meta-analysis

Childhood adversity (CA) is associated with increased risk of psychopathology, and reward processing (RP) may be one of the underlying mechanisms. However, evidence on impaired RP in childhood adversity is theoretically and methodologically heterogeneous. To provide a quantitative overview of studies on the relation between childhood adversity and RP assessed at the behavioral and subjective levels, and identify differences between studies that influence the effect size. Twenty-seven studies (overall N = 6801) were included. Peer-reviewed publications describing empirical studies on the relation between CA and behavioral and self-report measures of RP in humans were identified through systematic searches in six bibliographic databases. Effect sizes (r) were pooled using random-effects models. The potential moderator role of RP dimension, type of RP assessment, type of childhood adversity assessment, and age were examined. Results indicated a small, but consistent association between CA and impaired RP (r = 0.12; 95% CI: 0.07, 0.16), with medium heterogeneity (I2 = 62.43). The effect size was significantly larger (i.e., medium-sized) in studies that focused on reward learning rather than reward valuation and reward responsiveness; used cognitive tasks rather than self-report assessments of RP; and relied on official records rather than subjective reports of CA. There was evidence of publication bias, but overall effect size remained significant after imputation. These results suggest that multidimensional RP impairments (e.g., deficits in reward learning, biased reward valuation) are a consistent marker of CA, and may represent mechanisms underlying the increased risk of psychopathology.