Previous studies have suggested that the effects of contingent (response dependent) and noncontingent (response independent) cocaine administration may differ, which could limit the generality and validity of laboratory studies that use only noncontingent administration. Therefore, two separate three-component multiple schedules of operant responding were used to examine the effects of both types of cocaine administration on the acquisition and performance of response sequences, in four rhesus monkeys. In one multiple schedule, responding under a fixed-ratio (FR) 60 schedule was followed by intravenous (i.v.) saline or cocaine (0.0032-0.32 mg/kg per infusion), whereas responding in the other two components (i.e. acquisition and performance) was followed by food presentation. In the second multiple schedule, the cocaine administration component consisted of a variable-time (VT) schedule that mimicked each subject's pattern of self-administration. When compared to saline administration, increasing infusion doses of cocaine decreased overall response rates comparably in both food-maintained components, irrespective of the cocaine contingency. The 0.1-0.32 mg/kg infusion doses also increased the percentage of errors in 2 of 4 subjects; however, these disruptions in accuracy were not differentially associated with the type of cocaine administration and generally occurred at doses that produced large rate-decreasing effects. Taken together, these data suggest that the effects of cocaine on complex operant behavior in monkeys may not differ substantially as a function of contingent or noncontingent administration.