The cognitive effects of subchronic phencyclidine administration in rats are still unsettled in the literature. Possible causes of discrepancies are different drug treatment regimens and task parameters. The current experiment tested whether variations in procedures of the delayed T-maze alternation task result in performance differences following identical PCP treatments. Sixteen rats were trained on a continuous version of the T-maze task where they alternated between successive free-choice runs. Another sixteen were trained on a discrete trials version where each trial started with a forced run followed by a free choice test. After training, half of the rats submitted to each task version were treated daily for 14 days with i.p. injections of PCP (10 mg/kg) and the remaining half received a saline solution. At 48 h after the last injection, the subjects were tested for 10 days in their respective task version. Results showed that rats treated with PCP were impaired relative to controls in the continuous alternation task whereas performance of PCP and Saline groups did not differ in the discrete trials version. Cognitive control from prefrontal cortex and/or striatal response-related processes could have been damaged by PCP exposure. The systematic study of differences in tasks parameters may help reconcile discordant findings on PCP's functional outcomes.