Automatic Versus Evaluative Components of Cue Reactivity

Background: \- Relapse to drug abuse often happens in the presence of stimuli that are associated with previous drug use, also known as cues. Drug-taking behavior appears to be partly controlled by such cues. Some research suggests that cue-induced craving states are responsible for drug use and relapse, but other research suggests that cues can control drug taking without conscious craving or even awareness. Researchers are interested in using functional magnetic resonance imaging (fMRI) to study how individuals respond to drug-related cues and how these cues affect craving on a conscious or unconscious level. Objectives: \- To determine how drug-related stimuli (cues) affect thinking, information processing, and body reactions in current drug users. Eligibility: \- Individuals between 18 and 50 years of age who are dependent on either tobacco or crack-cocaine, or are healthy volunteers who are not dependent on either drug. Design: * The study will require two sessions, a training session and a testing session, that will take place within 48 hours of each other. Between the two test sessions, participants may not consume alcohol, tobacco, or illegal drugs, and must restrict caffeine consumption. * During the first session, participants will complete questionnaires about tobacco/cocaine use and craving, and will be trained on the tasks to be performed in the MRI scanner. * During the second session, participants will perform concentration tasks and look at pictures, some of which will be tobacco/cocaine related. Body reactions such as heart rate, pupil dilation, and sweating will be measured during this session. Some participants will have actual MRI scanning, while others will have mock MRI scanning. * After the MRI session, participants will complete questionnaires about craving and responses to the scan.