A basic problem in the discovery and development of novel drugs to be used in the treatment of neurological and psychiatric disorders is the absence of relevant in vitro or in vivo animal models that can yield results which can be extrapolated to man. Drug research now benefits from the fast development of functional imaging techniques such as positron emission tomography (PET) which trace radiolabelled molecules directly in the human brain. PET uses molecules that are labelled with short-lived radionuclides and injected intravenously into experimental animals, human volunteers or patients. The most frequent approach is to study how an unlabelled drug inhibits specific binding of a well characterised selective PET radioligand. The alternative direct approach is to radiolabel a new potential drug and to trace its uptake, anatomical distribution and binding in brain. Furthermore, the effects of a novel drug on physiological-biochemical parameters, such as glucose metabolism or blood flow, can also be assessed. The demonstration of quantitative relationships between drug binding in vivo and drug effects in patients is used to validate targets for drug action, to correlate pharmacological and physiological effects, and to optimise clinical treatment.