A framework for deep learning for dynamic online discussion in distance education is illustrated in this paper. The foundation of the framework is based on three general processes: information, methods and cognition. A structure for dynamic discussions within the framework provides three types of online discussion; flexible peer, structured topic and collaborative task discussion. The framework was applied during two semesters of an online multimedia design for instruction in a graduate level course. The strategies for creating dynamic discussion serve to facilitate online interactions among diverse learners and assist in the design of assignments for effective interactions. Students build on the adoptive learning taking place through assignments designed to promote adaptive learning and challenge their cognitive abilities, resulting in deep learning. The proposed framework and the strategies for dynamic discussion provide an online learning environment in which students learn beyond the goal of the course.