This paper describes an experiment designed to test an imagery-based craving management technique with a sample of adolescents diagnosed with substance-use disorders. Seventy adolescents between the ages of 14 and 18 (41 males) were recruited through two substance-abuse treatment programs. The experimental procedure involved stimulating craving using a previously tested imagery technique and then administering a newly developed procedure for reducing craving. Participants were randomly assigned to one of three conditions: an "interpersonal help" craving reduction condition; an "interpersonal distracter" craving reduction condition; and a control condition in which craving was stimulated and allowed to attenuate naturally, without intervention. Results indicated that following exposure to imagery-based drug cues, the "interpersonal distracter" condition effectively interrupted craving, compared to the control condition. The "interpersonal help condition" did not attenuate participant's craving response relative to the control condition. This study illustrates how experimental methods can be used to test and refine the efficacy of clinical interventions.