Use of analogy in learning scientific concepts

Four experiments compared learning of scientific concepts as expressed in either traditional literal form or through an analogy. Comprehension of basic-level details and inferential implications was measured through multiple-choice testing. In Experiment 1, literal or analogical renditions were presented in textual form only. In Experiment 2, text was accompanied by a dynamic video. In Experiment 3, the video and text literal rendition was compared with a text-only analogical rendition. In Experiment 4, subjects read only about a familiar domain. Subjects consistently answered basic-level questions most accurately when concepts were expressed literally, but answered inferential questions most accurately when concepts were expressed analogically. Analysis of individual differences (Experiment 2) indicated that this interaction strongly characterized the conceptual learning of science novices. The results are discussed within the framework of schema induction.