Abstract

Although stimulant drugs have been available and abused for centuries, it is only recently that procedures for evaluating their liability for abuse have been developed. This methodological development relies on several converging assessment procedures, combining to reveal the resultant profile of effects. The first of these, drug self-administration, is based on the principle that those drugs which can serve as reinforcers, maintaining behavior leading to their delivery, have a high liability for abuse. In addition, the more traditional approach to measuring abuse liability has been through assessing self-reported effects in which standardized questionnaires are used. A variant of this, a drug 'liking' scale has also been used. A third focus of measurement has been the accuracy with which subjects can identify the drug they have been administered, with some laboratories developing procedures for training subjects to discriminate specific substances. The profile of effects obtained with a specific compound can then be compared to the profile obtained with prototypic stimulants such as amphetamine or cocaine. The extent to which these profiles are similar would predict relative liability for abuse. This paper describes the data collected for amphetamine and concludes that the most accurate assessment of abuse liability is obtained when an unknown drug is compared to a drug with abuse liability using as many of these measures as possible.