The Reward Probability Index: Design and Validation of a Scale Measuring Access to Environmental Reward

Abstract Behavioral models of depression implicate decreased response-contingent positive reinforcement (RCPR) as critical toward the development and maintenance of depression (Lewinsohn, 1974). Given the absence of a psychometrically sound self-report measure of RCPR, the Reward Probability Index (RPI) was developed to measure access to environmental reward and to approximate actual RCPR. In Study 1 ( n = 269), exploratory factor analysis supported a 20-item two-factor model (Reward Probability, Environmental Suppressors) with strong internal consistency (α = .90). In Study 2 ( n = 281), confirmatory factor analysis supported this two-factor structure and convergent validity was established through strong correlations between the RPI and measures of activity, avoidance, reinforcement, and depression ( r = .65 to .81). Discriminant validity was supported via smaller correlations between the RPI and measures of social support and somatic anxiety ( r = – .29 to – .40). Two-week test–retest reliability was strong ( r = .69). In Study 3 ( n = 33), controlling for depression symptoms, hierarchical regression supported the incremental validity of the RPI in predicting daily diary reports of environmental reward. The RPI represents a parsimonious, reliable, and valid measure that may facilitate understanding of the etiology of depression and its relationship to overt behaviors.