Based on the hypothesis that drugs of abuse increase locomotor activity through mechanisms related to reinforcement, i.e. the mesolimbic dopamine (DA) system, ethanol-induced hyperactivity might provide a screening model to investigate the effect of ethanol on reward pathways. In the present study, ethanol had bidirectional effects on locomotion in mice: hyperactivity at low doses (2-3 g/kg) and sedation at high doses (4-5 g/kg). Such high doses induced a loss of righting reflex (LRR). The stimulant effect of ethanol was blocked by the D2/D3 antagonists, haloperidol (0.2 mg/kg) and tiapride (30-60 mg/kg), and by the D1 antagonist, SCH 23390 (0.03 mg/kg) whereas the non selective DA antagonist, clozapine decreased ethanol-induced hyperactivity at a dose (1 mg/kg) which also decreased activity in control animals. Unlike haloperidol and clozapine which potentiated LRR induced by ethanol, the selective DA antagonists, tiapride and SCH 23390, had no effect. Pretreatment with the D2/D3 agonist, quinpirole (0.1-0.3 mg/kg), reduced hyperactivity induced by ethanol presumably by stimulation of pre-synaptic receptors but did not change LRR. The D1 full agonist, SKF 81297 which produced hyperactivity by itself and the D1 partial agonist, SKF 38393, did not specifically affect ethanol-induced activities. The results indicate that activation of D1 and D2/D3 DA receptors is implicated in ethanol-induced hyperactivity whereas other mechanisms might mediate the sedative effects of ethanol. Tiapride and haloperidol, both used in the management of alcohol dependence, might exert beneficial effects by counteracting the reinforcing effects of ethanol. Tiapride's lack of interaction with the depressant effects of ethanol may account for its better tolerance in alcoholic patients.