Representational and decorative pictures in science and mathematics tests: Do they make a difference?

Multimedia effects in learning receive much attention in research. In contrast, effects of different picture types in testing materials are hardly studied. The present computer-based experiment with n = 404 students investigates the effects of adding a representational picture (RP) or a decorative picture (DP) to text-based educational test items on students' performance, metacognition, test-taking motivation and item processing in the domains of science and mathematics. The data show that RPs enhanced students' performance, perception of ease and test-taking pleasure in both scientific and mathematics items. Furthermore, RPs increased time on task (TOT) in mathematics, but not in scientific items. DPs had no significant effect on students’ performance, test-taking pleasure or perceived ease, while DPs reduced TOT in mathematics items. Explanations for the process-related differences are discussed. The pattern of results indicates that the multimedia principle and the coherence principle from instructional psychology can be transferred to educational testing material.