Tardive dyskinesia (TD) is a late side effect of long-term antipsychotic use in humans, and the vacuous chewing movement (VCM) model has been used routinely to study this movement disorder in rats. Recent receptor occupancy studies in humans and rats have found that antipsychotics given in doses which lead to moderate levels of D(2) receptor blockade can achieve optimal clinical response while minimizing the emergence of acute motor side effects. This suggests that clinicians may have been using inappropriately high doses of antipsychotics. A review of the existing VCM literature indicates that most animal studies have similarly employed antipsychotic doses that are high, i.e. doses that lead to near complete D(2) receptor saturation. To verify whether the incidence or severity of VCMs would decrease with lower antipsychotic doses, we conducted initial experiments with different doses of haloperidol (HAL) given either as repeated daily injections or as depot injections over the course of several weeks. Our results demonstrate that (1) the incidence of VCMs is significantly related to HAL dose, and (2) significant levels of VCMs only emerge when haloperidol is continually present. These findings are consistent with the possibility that total D(2) occupancy, as well as 'transience' of receptor occupation, may be important in the development of late-onset antipsychotic-induced dyskinetic syndromes.