Accumulating evidence suggests that antipsychotics (APs) that lead to sustained blockade of dopamine D 2 receptors are more likely to induce acute extrapyramidal side effects (EPS) compared to APs that only occupy D 2 receptors transiently. It is unclear, however, whether a similar relationship exists for long-term AP-induced motoric side effects like tardive dyskinesia (TD). The objective of this study was to ascertain whether transient (via daily subcutaneous (s.c.) injections) vs continuous (via osmotic minipump) AP-induced D 2 receptor occupancy differentially affects the development of haloperidol-induced vacuous chewing movements (VCMs), an animal model of TD. Six groups of 12 rats received 0.1, 0.25, or 1 mg/kg of haloperidol or vehicle (n ¼ 36) via osmotic minipump (to provide within-day sustained) or daily s.c. injection (within-day transient) for 8 weeks. VCMs were measured on a weekly basis and D 2 occupancy levels were measured in vivo using [ 3 H]-raclopride at the end of the experiment. Minipump-treated rats developed HAL dose-dependent D 2 occupancies of 0.1 mg/kg/day (57%), 0.25 mg/kg/day (70%), and 1 mg/kg/day (88%). S.C.-treated rats also developed HAL dosedependent D 2 occupancies of 0.1 mg/kg/day (83% peak, 3% trough), 0.25 mg/kg/day (89% peak, 0% trough), and 1 mg/kg/day (94% peak, 17% trough). A total of 43% of rats given 0.25 and 1 mg/kg/day of HAL via minipump developed high VCMs compared to only 8% of the rats given the same doses via daily s.c. injections. The 0.1 mg/kg dose did not give rise to VCMs beyond vehicle levels regardless of the route of administration. These findings support the contention that D 2 occupancy levels induced by chronic HAL must be high and sustained through the day before significant risk of VCMs, and perhaps also TD, emerges.