The current study assessed whether four 15- to 17-year-old individuals diagnosed with autism would remain on-task for more intervals and complete tasks independently as a function of using self-reinforcement or teacher-delivered reinforcement. An adapted alternating-treatments design with teacher-delivered reinforcement, self-reinforcement, and a control condition was used. The participants followed a schedule on an iPod touch to complete a vocational task, a daily living task, and a leisure task. To promote independence, the proximity of the instructor was faded until the instructor was completely removed during self-reinforcement. Results of this study demonstrated high percentages of on-task behavior and schedule completion. Overall, during the self-reinforcement condition, proximity of the instructor was faded in fewer sessions than during the teacher-delivered reinforcement condition. Generalization of responding was observed in the community. The results of this study established a novel method for implementing self-reinforcement.