This study evaluated the use of an interrupted chain procedure to establish spontaneous mands (i.e., requests) in three preschool aged children with a diagnosis of autism spectrum disorder. All three participants attended a university based preschool center, which used the methodology of Applied Behavior Analysis. Additionally, all three participants used a high-tech speech-generating device that consisted of an iPad® tablet and application Proloquo2Go™. At this center all three children were exposed to the training procedure that used the interrupted chain procedure to first establish motivation and a five-second time delay with full physical prompts. The results indicated that all three participants acquired the ability to mand for a missing item needed to complete a play activity (i.e., puzzle) to the mastery criteria of three consecutive sessions at or above 80% independent and accurate responding. On average it required the participants five training sessions to reach this criterion. Additionally, maintenance data were collected, and the acquired repertoire was demonstrated to maintain for all three participants. Limitations of the research procedures, clinical implications, and suggestions for future research are also discussed.