Reducing Verbal Redundancy in Multimedia Learning: An Undesired Desirable Difficulty?

Previous research on the redundancy principle in multimedia learning has shown that although exact correspondence between on-screen text and narration generally impairs learning, brief labels within an animation can improve learning. To clarify and extend the theoretical and practical implications of these results, the authors of the present research examined the extent to which varying degrees of correspondence between on-screen text and narration in a multimedia lesson affects recall and transfer. In 2 experiments, college students viewed an animated and narrated PowerPoint lesson about the life cycle of a star, with different participants experiencing different degrees of correspondence between on-screen text and narration. Consistent with the redundancy principle and the dual-channel theory of multimedia learning, both experiments demonstrated impairment for on-screen text identical to the narration. As an extension of previous research on the redundancy principle, however, both experiments also demonstrated an advantage for a small amount of discrepancy between narration and on-screen text. On the other hand, too much discrepancy resulted in learning just as poor as when the on-screen text was identical to the narration. Metacognitive judgments revealed that participants tended to prefer on-screen text identical to the narration, even though recall and transfer scores showed that on-screen text worded slightly differently than the narration was better for learning. Results indicate that despite learner preferences to the contrary, slight discrepancies between on-screen text and narration can be a desirable difficulty, suggesting an extension to the redundancy principle that is consistent with the desirable difficulties framework as well as the cognitive theory of multimedia learning. (Contains 1 figure, 4 tables and 1 footnote.)