When do diagrams enhance learning? A framework for designing relevant representations

Abstract

Although many studies demonstrate large learning gains when instruction includes diagrams, diagrams do not always lead to improved outcomes. How can instructional designers know whether a given diagram will enhance learning? We have developed a framework of three factors that influence the effectiveness of a diagram in a particular learning situation: the learning objective, the design of the visual representation and the cognitive processing of the learner. In a randomized-design study conducted in a college chemistry class, we found that instruction that included diagrams created with this framework led to enhanced performance on open-ended transfer items compared to traditional instruction, particularly for low-performing students. We propose that a concept-based cognitive theory of multimedia learning that includes a conceptual working memory component may explain why the efficacy of diagrams depends heavily on the prior knowledge of the learner as well as the conceptual information available in the representation.