The effect of withdrawal states on drug reinforcement and drug self-administration has received little systematic attention from drug abuse researchers. This is a critical omission because it is generally assumed that humans will seek drugs to alleviate unpleasant or relapsing withdrawal symptoms, thus maintaining their state of physical dependence or relapsing after a period of abstinence. This research will examine the following issues: (a) the pattern of self-administration when only low doses of opiates are available, (b) the effect of opioid antagonist-precipitated withdrawal on opiate self-administration, and (c) the relationship between self-administration behavior and subjective drug effects. This research should be useful in the development of better methods to predict abuse liability of drugs, because it combines the two primary strategies of abuse liability assessment, self-administration and subjective effects testing, in a single study. In addition to addressing these important pharmacological-behavioral interactions, this research may ultimately result in more effective treatment methods for drug abuse. Pilot testing of residential subjects with a history of opioid abuse has been completed, and the main study should begin in 1991.