In this mixed methods study, we used a convergent parallel design to create and analyze a task that was implemented across four US institutions with 99 preservice teachers (PSTs) using one of two multi-media platforms: LessonSketch or GoAnimate. We analyzed PSTs' comic-based depictions or animations to examine the level of sophistication for teacher questioning, number of teacher utterances, use of visual information and format of classroom interaction, and features of conveyed time. PSTs who used GoAnimate included significantly more utterances than those who used LessonSketch, yet the level of questioning was statistically equivalent across both mediums. Artifacts created with GoAnimate tended to convey small group interactions, whereas LessonSketch depictions typically portrayed whole class interactions and included additional visual information. We discuss these findings and their specific implications for teacher educators to select and align use of such multi-media platforms for particular instructional aims.