The self-administration paradigm is an experimental model of drug dependence in which the reinforcing properties of drugs can be directly assessed. This paradigm avoids the possible confounding influence of nonpharmacologic factors which may contribute to drug taking in the nonlaboratory environment. When animals serve as subjects, social and cultural factors unique to humans may also be eliminated as confounding influences. Most drugs of abuse are self-administered by animals and humans under such conditions. Until 1981, laboratory studies of nicotine self-administration suggested that nicotine, in its own right, was only a marginally effective reinforcer. As will be shown in the present review, a study by Goldberg and his co-workers in 1981  demonstrated clearly that nicotine could serve as a highly efficacious reinforcer in laboratory animals. There are several parameters which can function to substantially strengthen the behavior which leads to nicotine ingestion. These include the following: (1) intermittent availability of nicotine, (2) intermittent presentation of nicotine-paired stimuli, and (3) concurrent schedules of food reinforcement. Initial findings from a human IV nicotine self-administration study were consistent with those from the animal studies. Together these results confirm that nicotine can function to control behavior by serving as a reinforcer for animals and humans. The results also suggest that commonly used tobacco products function as ideal nicotine delivery systems for controlling behavior since they provide discrete nicotine-paired stimuli and lend themselves to intermittent nicotine delivery.