A battery of 12 neuropsychological tests were administered on two occasions to 15 chronic PCP users who reduced or eliminated use of PCP over a 4-week period. A comparison sample of 15 non-PCP drug users who did not differ in age, sex, education, and ethnic composition also were tested at the two time periods. Impairment, initially higher for PCP users, decreased significantly after reduction in use of PCP. A nonsignificant increase in impairment was found for non-PCP drug users. Analysis of each variable revealed that substantial improvement occurred on the acquisition, recall, and delayed recall scores of the Randt Memory Test. Improvement also was noted for some individuals on Trails B and Digit Symbol tests.