Chronic cocaine self-administration in rhesus monkeys: impact on associative learning, cognitive control, and working memory.

Cocaine users display a wide range of cognitive impairments. Because treatment outcome is dependent on baseline cognitive ability, it is clinically important to understand the underlying neurobiology of these deficits. Therefore, it is crucial to determine whether cocaine exposure by itself is an etiological factor and, if so, to determine the overall nature of cognitive deficits associated with cocaine use. This will help to guide therapeutic approaches that address cognitive components of cocaine use to improve treatment outcome. We used rhesus monkeys in a longitudinal study in which 14 animals were characterized before assignment to matched control (n = 6) and cocaine self-administration (n = 8) groups. Self-administration took place on 4 consecutive days/week over 9 months, with a maximum (and typical) daily cumulative intake of 3.0 mg/kg. Weekly cognitive assessments (total of 36) were conducted after a 72 h drug-free period. We used a stimulus discrimination task with reversal to evaluate associative learning and the cognitive control/flexibility needed to adapt to changes in reward contingencies. After extended self-administration, initial accuracy on the stimulus discrimination indicated intact associative learning. However, animals were impaired at maintaining high levels of accuracy needed to reach criterion and initiate the reversal. Increasing the reward contrast between stimuli permitted evaluation of reversal performance and revealed striking deficits in the cocaine group. Impairments in visual working memory were also observed using a delayed match-to-sample task. These results suggest a combination of generalized, possibly attentional, impairments, along with a more specific cognitive control impairment implicating orbitofrontal cortex dysfunction.