The curious effect of using drawings in conditional reasoning problems

College level students were given two paper and pencil conditional reasoning tests in varying order. The first test consisted of three problems presented entirely verbally. The second test was identical to the first except that two of the three problems made use of line drawings as concrete referents. Performance on the problems with drawing was significantly worse than for purely verbal presentation. In addition, interaction effects between the two forms of presentation were observed over the one-week interval between the two tests.