We compared antipsychotic drugs (haloperidol, chlorpromazine and clozapine) and sigma antagonists (remoxipride, cinuperone, alpha-(4-fluorophenyl)-4-(-fluoro-2-pyrimidinyl)-1-piperazine butanol (BMY 14802) and rimcazole) in the radio-ligand binding and behavioural experiments in rodents. A good correlation was established between the affinity of compounds at dopamine2-receptors in the striatum and their ability to block apomorphine-, amphetamine- and quipazine-induced behavioural effects in rodents. By contrast, no correlation was found between the behavioural effects of these drugs and their affinity at dopamine1-5-HT2- and sigma receptors. The rank order of potency among the studied antipsychotic drugs in the behavioural tests and at dopamine2-receptors was following: haloperidol >> chlorpromazine > or = clozapine. The effectiveness of chlorpromazine and clozapine was nearly similar against apomorphine-induced aggressiveness and yawning, whereas at 5-HT2-receptors clozapine was more active than chlorpromazine. The weak activity of sigma antagonists at dopamine2 receptors could be a possible reason why these compounds were less effective in the behavioural studies compared to antipsychotic drugs. However, the antagonism of remoxipride against apomorphine-induced stereotypy and aggressiveness is not related to its activity at sigma receptors, because the other sigma antagonists did not block these effects of apomorphine. It is probable that remoxipride exerts its action through blocking of dopamine2 receptors. In conclusion, the present study revealed only weak activity of sigma antagonists in the behavioural models widely used to study the antipsychotic drugs. Therefore, the antipsychotic activity of sigma antagonists is doubtful.