Over the last few weeks decades research in psychiatry has undergone a considerable efflorescence. Experimental projects in fields as different as neurobiology or psychopathology, neuro-anatomy or epidemiology, or even cognitive psychology are of course subject to precise conditions which may derive from ethical guidelines as well as different legal systems. These legislative contexts therefore represent a parallel set of restrictions in the path of research activities and may in some cases curb their development. Restrictions of an ethical or deontological type are direct heirs of necessity. They emerged on the eve of the second world war when it became apparent that precisely defined conditions in which research could be conducted in humans were required, and derived for the most part from the recommandations of the Nuremberg Code on the one hand, and the Helsinki declaration on the other. The last version of the Code of Medical Deontology applicable to medical practice in France, and published by decree in September 1995, devoted an article to the question of experimentation. Each country has ifs own specific legislation which, in France, was explicitly formulated with the publication of the Loi Huriet in December 1988. The European Union is currently attempting to produce a homologous version of the various legislative documents and recommendations should soon be forthcoming which will be applicable to all member countries of the Community. Another area of limitation is less clearly formulated and involves the technical and occasionally methodological framework within research projects take place. Technical demands are highly variable depending on the field of investigation and chiefly revolve around a dogged hunt for statistically significant results (statistical significance sometimes seems to dispense with the need to determine the real meaning of results 5); in some cases they too may curb inventiveness. This last type restriction is particularly apparent in drug trials where imperatives of industrial development as viewed by the sponsor may not always coincide with the investigators' desire to mark therapeutic progress. The increasing rigidity of strategies for assessing new substances which has become obvious over the last few years may in the end risk penalizing the discovery of innovative treatments. Promoting a return to clinical practice in the setting of these research projects would indisputably provide some novel solutions. Although many of the restrictions which currently stand in the way of psychiatric research are, so to speak, natural limitations and difficult to debate, excessive formalism in some areas should be spoken out against. Much thought has been given to this subject, a trend which is likely to inject some dynamism into psychiatric research at the dawn of the third millennium just around the corner.