The current study employed an adapted alternating treatments design with reversal and multiple probe across participants components to compare the effects of traditional video priming and simultaneous video modeling on the acquisition of play skills in two children diagnosed with autism. Generalization was programmed across play sets, instructors, and settings. Overall, both video modeling procedures proved to be effective in teaching and producing maintenance of play skills. For one participant, these procedures appeared to be equally effective in terms of acquisition of the main dependent variable, scripted play actions. For another participant, scripted play actions were acquired more quickly in the simultaneous condition. (Contains 2 tables and 3 figures.)