Studies on the dependence-related behavioural effects of nicotine in animals and humans have been compared to assess the extent of inter-species consistency in relation to the face, construct and predictive validity of the animal models. The major areas covered in this selective review are neuroadaptations occurring as a consequence of repeated exposures to nicotine (i.e. tolerance, sensitization and withdrawal) and the stimulus properties of nicotine (positively reinforcing, discriminative stimulus and aversive stimulus effects). Most of these phenomena were demonstrated first in animal subjects, often many years before comparable data for humans became available, although this was not invariably the case. It is argued that the overall pattern of results for the neuroadaptations is very similar across species. More specifically, it is proposed that repeated exposure to nicotine produces a change from a relatively aversive initial overall profile to one in which tolerance has developed to the aversive effects and the positive effects on mood have been enhanced or sensitized; at the same time a withdrawal syndrome develops that can now be assessed by a variety of measures in animal and human subjects. With regard to stimulus properties, the nicotine discriminative stimulus in animal and human subjects shows a particularly close correspondence, which may reflect the functional equivalence of the procedures in the different species. The self-administration of pure nicotine, as contrasted with tobacco smoke, has been firmly established in animal subjects for some time but there are only a small number of strictly comparable studies in humans. Both differences and similarities in findings across species may be seen; however, care must be taken to identify the possible confounding influence of procedural variation between species, rather than the species per se, as the determining factor. Overall, the animal studies in this area have shown remarkably good predictive validity, coupled with rather more variable levels of face and construct validity.