Performance of Long-Evans rats repeatedly treated with a high dose of phencyclidine (PCP) was compared with that of controls on two tasks involving rule shifts. Fifteen hours after receiving 12 daily doses of PCP (10mg/kg) or saline, rats were tested in the first task where a fixed or variable goal had to be recognized among four maze-arm locations. After receiving the last of 33 injections, rats were subjected to a second task: they had to discriminate the relevant perceptual dimension in a pair of stimuli and choose the appropriate stimulus in each problem. Results from both tasks revealed no difference between controls and rats injected with PCP either during acquisition using a constant rule or during testing with rule shifts. Consequently, PCP appears to lack consistency in disturbing frontal cognitive functions in animal models of schizophrenia.