Prospective, double-blind, randomized trials comparing atypical antipsychotic drugs (APDs) to typical APDs, such as haloperidol, indicate that atypical APDs provide a modest benefit to cognitive function in schizophrenia. However, the validity of this inference has been contested by suggestions that the cognitive improvements observed with atypical APDs reflect practice effects associated with repeated testing on the same neuropsychological instruments, or an avoidance of a deleterious effect of haloperidol on cognitive function that might be dose related. These alternate hypotheses were assessed by meta-analyses that 1) examined the relationship between cognitive change and dose of haloperidol within the control arms of prospective atypical vs. typical APD clinical trials; and 2) compared the magnitude of change observed within the haloperidol arms of these studies to estimated practice effects for several commonly used neuropsychological measures. The results indicate that overall cognitive performance improves while on haloperidol. Studies that used a low dose of haloperidol (<10 mg) did not yield larger effect sizes for overall cognitive function or specific neuropsychological measures than studies that used a high dose (>10 mg), although doses greater than 24 mg appear to have deleterious effects. For two of the six neuropsychological tests examined (digit symbol substitution and verbal fluency) the magnitude of change observed was significantly less than practice effects. The results indicate that although haloperidol may cause deleterious effects at very high doses, or in specific cognitive domains, these effects are not likely to explain the broader range of cognitive improvements observed with atypical APDs.