Self-administration of cocaine by humans: a laboratory perspective.

Laboratory research evaluating the behavioural and physiological effects of cocaine has produced important basic information relevant to the prediction and control of cocaine abuse. Adaptation for human subjects of laboratory procedures originally developed with non-humans has provided the methodology for assessing the relationship between cocaine-taking and the self-reported effects of cocaine. This laboratory model for evaluating drug-taking has been adapted to include all the routes of administration by which cocaine is abused. Data collected in cocaine choice/self-administration studies indicate that there is a generally good correlation between cocaine self-administration and its stimulant-like 'positive' effects. There are, however, areas in which these two measures clearly diverge, and drug self-administration procedures provide a sensitive measure of cocaine's effects when it is taken by the routes and in the patterns people use outside the laboratory. Combining these measures with other behavioural evaluations provides useful information for understanding cocaine use and abuse. In addition, and importantly, these laboratory procedures are useful tools in the development and assessment of treatment interventions which might be used effectively.