The Magic School Bus dilemma: How fantasy affects children's learning from stories.

Although children's books often include fantasy, research suggests that children do not learn as well from fantastical stories as from realistic ones. The current studies investigated whether the type of fantasy matters, in effect testing two possible mechanisms for fantasy's interference. Across two studies, 110 5-year-olds were read different types of fantastical stories containing a problem and then were asked to solve an analogous problem in a real lab setting. Children who were read a minimally fantastical version of the story, in which the story occurred on another planet "that looked just like Earth," were no more likely to transfer the solution than children who heard a story that was slightly more fantastical in that the story occurred on another planet and that planet looked different from Earth (e.g., orange grass, a green sky). In contrast, significantly higher rates of learning were observed when the story contained those elements and two physically impossible events (e.g., walking through walls). Furthermore, this improvement was obtained only when the impossible events preceded, and not when they followed, the educational content. Although fantasy may sometimes detract from learning (as other research has shown), these new studies suggest that minimal fantasy does not and that particular types of fantasy may even increase learning. We propose that the mechanism for this may be that a small dose of impossible events induces deeper processing of the subsequent events in the story.