Substance abuse and criminal thinking: testing the countervailing, mediation, and specificity hypotheses.

The purpose of this study was to determine (a) which of 2 dimensions of criminal thinking (proactive and/or reactive) correlates with prior substance abuse; (b) whether criminal thinking mediates the relationship between prior substance abuse and recidivism; (c) if a direct relationship exists between specific drugs of abuse and specific criminal thinking styles. First, the reconstructed Proactive (Prc) and Reactive (Rrc) Criminal Thinking scores from the Psychological Inventory of Criminal Thinking Styles (PICTS; Walters, 1995) were correlated with a dichotomous measure of prior substance abuse and a continuous measure of the number of substances abused in a sample of 2877 male federal prisoners (age: M = 34.96, SD = 9.89, range = 18-84; race: 63.6% Black, 17.3% White, 17.6% Hispanic, 1.4% other). The results indicated that only the Rrc score correlated significantly with prior substance abuse when the effect of the alternative measure (Prc in the case of Rrc and Rrc in the case of the Prc) was controlled through partial correlations. Second, reactive criminal thinking was found to mediate the relationship between a history of prior substance abuse and subsequent recidivism in a subsample of 1101 inmates who were released from prison during a 1- to 76-month follow-up. Third, both specific (alcohol with cutoff; marijuana with cognitive indolence) and global (heroin, cocaine, and amphetamine with cutoff, cognitive indolence, and discontinuity) drug-criminal thinking correlations were obtained. These results suggest that reactive criminal thinking plays a potentially important role in the drug-crime relationship.