Purpose <p>This is a systematic review investigating factors that influence children’s learning of speech-generating devices (SGDs) for manding with a single icon. The review addressed three questions: (1) the demographic characteristics of participants involved in SGD studies, (2) the effectiveness of different teaching strategies, and (3) the impact of intervention dosage on the acquisition of manding responses using Tau-U.</p> Methods <p>This review focused on peer-reviewed journal articles that studied Mand learning with SGDs in children with autism, employing a single-case design and behavioral teaching procedures. ProQuest, the Education Resources Information Center (ERIC), and PsycINFO were accessed in September 2023 for this review. Through a comprehensive analysis of 14 peer-reviewed studies from these databases, the progress of 42 participants was compared across different instructional methods. The review contrasted three major instructional methods with time delays: least-to-most prompting, most-to-least prompting, and full physical prompting.</p> Results <p>Participants’ demographics aligned with statistics from the Centers for Disease Control and Prevention. The settings of the studies indicated a potential literature gap in teaching manding in naturally occurring settings. The Tau-U scores suggested similar effectiveness between the most-to-least and least-to-most prompting procedures, but peer mediation was recommended.</p> Conclusion <p>No linear relationship was found between the effectiveness of the treatment and treatment dosage based on Tau-U scores. Several research trends were identified. This study noted an uneven distribution of intervention methods across the studies, and the total number of participants was limited. The findings could inform practitioners on teaching manding with a single icon from an SGD and suggest directions for future research on teaching the use of SGDs for manding in children with autism.</p>