Distinguishing Types of Superficial Similarities: Different Effects on the Access and Use of Earlier Problems

When solving a problem, people often access and make use of an earlier problem. A common view is that superficial similarities may affect which earlier problem is accessed, but they have little or no effect on how that earlier problem is used. The reported experiments provide evidence against this view. Subjects learned four probability principles illustrated by word problems. Test problems varied in their similarity to the study problems in three ways: story lines, objects, and correspondence of objects' roles (i.e., whether similar objects filled similar roles). The superficial similarity of object correspondences had a large effect on use (Experiment 1), although it sometimes had little or no effect on access (Experiment 2). Experiment 3 showed that two superficial similarities, story lines and object correspondences, differentially affect access and use.s These results suggest a more complex role of superficial similarity in problem solving and the need for distinguishing types of superficial similarities.