The cognitive effects of long-term cannabis use are insufficiently understood. Most studies concur that cognitive deficits persist at least several days after stopping heavy cannabis use. But studies differ on whether such deficits persist long term or whether they are correlated with increasing duration of lifetime cannabis use. The authors administered neuropsychological tests to 77 current heavy cannabis users who had smoked cannabis at least 5000 times in their lives, and to 87 control subjects who had smoked no more than 50 times in their lives. The heavy smokers showed deficits on memory of word lists on Days 0, 1, and 7 of a supervised abstinence period. By Day 28, however, few significant differences were found between users and controls on the test measures, and there were few significant associations between total lifetime cannabis consumption and test performance. Although these findings may be affected by residual confounding, as in all retrospective studies, they suggest that cannabis-associated cognitive deficits are reversible and related to recent cannabis exposure rather than irreversible and related to cumulative lifetime use.