During the four decades that research has been carried out on antipsychotic drugs, a variety of methods have been used to study the effects of these compounds on dopamine neurotransmission. An important issue in this research was to find an explanation for the difference between "typical" and "atypical" antipsychotic drugs. The hypothesis that the beneficial properties and the motor side effects of antipsychotic drugs result from their effects on different groups of dopamine neurons has received considerable attention. Numerous researchers have tried to discover regiospecific actions of antipsychotic drugs in mesolimbic and in mesocortical dopamine neurons. An overview of these research attempts is presented here. Electrophysiological studies showed a selective action of atypical antipsychotic drugs on A10 dopamine neurons. It was found that chronic treatment with these compounds induced a preferential depolarisation block of the A10 neurons that project to the mesolimbic areas. The model represents certain clinical features of antipsychotic drug use and offers a possible explanation for the lack of extrapyramidal side effects of atypical antipsychotic drugs. Dopamine neurons projecting from A10 to the frontal cortex are also considered as a possible site of action of atypical antipsychotic drugs. Microdialysis studies have shown that certain atypical antipsychotic drugs selectively enhance the release of dopamine in the prefrontal cortex when compared with typical antipsychotic drugs. The finding that repeated treatment with antipsychotic drugs increased dopamine D(2) receptor binding in the frontal cortex confirms the significance of this brain area. These properties might indeed explain certain beneficial effects of atypical antipsychotic drugs such as improvement of cognitive dysfunction. However the effects of typical and atypical antipsychotic drugs in the frontal cortex could not be fully differentiated, which illustrates the difficulty of localising clinical effects of antipsychotic drugs in terms of regional dopamine neurons. Recently new insights into the mechanism of action of typical and atypical antipsychotic drugs have been published. Clinical positron emission tomography (PET) studies have indicated that a moderate dopamine D(2) receptor occupancy, probably combined with a high dissociation rate, might provide the optimal clinical conditions for an antipsychotic drug, without inducing extrapyramidal side effects. Moreover the efficacy of benzamides as atypical antipsychotic drugs suggests that low to moderate dopamine D(2) blockade is probably the most important-if not the only-criterion that determines "atypicality". Interestingly these new insights are based on PET studies of the human basal ganglia and not on the comparison of different brain areas. Apparently, according to this concept an ideal antipsychotic drug need not to act on a particular type of dopamine neurons, as it is the moderate dopamine D(2) receptor occupancy that determines the desirable clinical effects. It is concluded that both beneficial actions and side effects, of antipsychotic drugs might be dose dependently localised in A9 as well as A10 dopamine neurons.