Concrete vs. Abstract Problem Formats: A Disadvantage of Prior Knowledge

Abstract

Three experiments examine the effects of varying the relative concreteness of physics word problems on student performance.Previous studies have found that concrete representations benefit performance for relatively simple problems, whereas abstract representations are beneficial for more complexproblems. These findings are replicated in a physics context. More importantly,a significant disadvantage for concrete representations is identified for some questions. When a problem potentially elicits prior knowledge that is contrary to scientific knowledge, e.g. a scientific "misconceptio"n, it is found that the concrete representation invokes incorrect answers more frequently than abstract representations. In addition, an interaction is found between the final course grade of the student and the abstract vs. concrete problem format, with higher grade students performing disproportionally better on abstract problems. This is consistent with previous findings in reasoning and an explanation is provided that involves the cuing of familiar, automatic knowledge and skills vs. explicit deliberate processes.