Thematic Maps Improve Memory for Facts and Inferences: A Test of the Stimulus Order Hypothesis

Undergraduates studied a thematic map of colonial Ceylon and read an expository text containing facts related to, or not related to, the map theme. Stimulus order (map-text vs. text-map) was varied between subjects, and half of the learners in each stimulus order group received a verbal prime related to the map theme. Subjects in the map-first condition recalled more theme-related and unrelated text facts and made more correct inferences involving the theme displayed on the map. Verbal priming had no effect on memory for theme-related facts or inferences, but reduced recall of unrelated facts. These results were interpreted within the dual coding framework, where maps are encoded as structurally coherent images. Such images are computationally efficient and can be used to retrieve associated text facts and to make the comparative judgments required for inferencing.