The purpose of this study was to evaluate the effectiveness of a computer-based program designed to increase percentage of correct match to sample discrimination tasks and generalization of the skills to a natural setting. Four students with moderate to severe intellectual disabilities participated. The dependent variables were: (a) the percentage of correct match to sample trials completed on the computer and (b) the percentage of items correctly selected in the natural setting of a local grocery store. Pre and post generalization testing included locating items presented and not presented during instruction. The independent variable was a multimedia computer program entitled Project SHOP that provided instruction through interactive practice activities incorporating a graduated response criterion. A multiple probe design across behaviors and replicated across four participants was used to evaluate experimental control. Results indicated that following intervention, the percentage of correct response in the community increased.