Objective: To determine whether premorbid IQ mediates performance on neurocognitive tests in individuals diagnosed with cocaine use disorder (CUD). Method: Recently abstinent cocaine users (N = 113) completed measures sensitive to the effects of cocaine on cognition: Conners’ Continuous Performance Task–II (CPT–II), n-back working memory test, and Hopkins Verbal Learning Task–Revised (HVLT–R). Premorbid IQ was calculated using the Oklahoma Premorbid Intelligence Estimate, which integrates scores from the Wechsler Adult Intelligence Scale–III and demographic variables. Participants were grouped according to their premorbid IQ using commonly accepted classifications of ability level (above average [>110], average [90–109], and below average [<90]) and comparisons in neurocognitive performance were performed using one-way analysis of variance. Results: Significant differences were detected between groups on the HVLT–R including Trial 1 (p = .002), total word recall across the 3 list-learning trials (p < .001), and recall following a delay (p < .001). Significant differences were also detected on the N-back, including auditory and visual accuracy (p = .022 and p < .001, respectively) and mean and maximum block length (p < .001). Although significant differences were observed between the above average and average groups (mean effect size = .418 [Cohen’s d]), the magnitude of group differences was greatest between the average and below average groups (mean effect size = .716). Conclusions: These results raise questions as to whether the neurocognitive impairment observed in individuals diagnosed with CUD predated the onset of cocaine use or whether the impairments were caused by cocaine use. Because these impairments are potential risk factors for poor treatment outcomes, it is important to consider the need to modify treatment programs to account for lower premorbid IQ.
