The acute effects of 1,3-trimethylxanthine (caffeine) were assessed using an operant test battery (OTB) of complex food-reinforced tasks that are thought to depend upon relatively specific brain functions, such as motivation to work for food (progressive ratio, PR), learning (incremental repeated acquisition, IRA), color and position discrimination (conditioned position responding, CPR), time estimation (temporal response differentiation, TRD), and short-term memory and attention (delayed matching-to-sample, DMTS). Endpoints included response rates (RR), accuracies (ACC), and percent task completed (PTC). Caffeine sulfate (0.175-20.0 mg/kg, IV), given 15 min pretesting, produced significant dose-dependent decreases in TRD percent task completed and accuracy at doses > or = 5.6 mg/kg. Caffeine produced no systematic effects on either DMTS or PR responding, but low doses tended to enhance performance in both IRA and CPR tasks. Thus, in monkeys, performance of an operant task designed to model time estimation is more sensitive to the disruptive effects of caffeine than is performance of the other tasks in the OTB.