Separated and Integrated Formats of Information Presentation in Multimedia Learning and the Expertise Reversal Effect

• In Event: Learning With Static and Dynamic Visualizations

Abstract

Objectives and purposes: Within two experiments, we examined the influence of text format in multimedia learning environments. The main objective was to show that the modality effect is actually based on a split-attention effect. Therefore, principles known from research with static images have been transferred to multimedia learning with dynamic animations.
Theoretical framework: Recent research increases doubts about the validity of the modality effect in multimedia learning from an empirical and a theoretical perspective. A recent study by Schüler, Scheiter, Rummer and Gerjets (2012) indicates, based on eye tracking analyses, that learners with visual only animations focus on one aspect of an animation, (e.g., the text) while the other information source is out of focus. Thus, it is likely that a genuine split-attention effect leads to worse performance of learners in animations explained by visual text.
Method: In a first experiment, animations with corresponding visually presented text (either in a separated or integrated format) were compared to animations with auditorially presented text. In addition, a high and low amount of text and a low amount of text were varied. A subsequent experiment replicated and extended the first experiment. As a quasi-experimental second factor, participants’ prior knowledge (low vs. high) was controlled.
Data sources and materials: In both experiments, learning outcomes were assessed as pre- and post-tests using a standardized test. In addition, cognitive load was measured as a dependent variable using the NASA-TLX questionnaire (Hart & Staveland, 1988). The learning material was presented as an animation about data transfer and routing algorithms. The animation was presented either in an auditory format or an integrated visual format (text within the animation), or in separated format (text at the bottom of the animation).
Results: Results from the first experiment revealed that, with the separated format, learners reported the highest cognitive load and also showed worse learning performance (η²=.16). Participants in the condition with the integrated format and the auditorially presented text performed comparably well in the knowledge post-test. Prior knowledge had a significant impact as a covariate (η²=.16). Results from the second experiment replicated findings from the first experiment on a descriptive level, revealing that the animation using the separated format led to the worst learning performance among learners with low prior knowledge. Alternatively, learners with high prior knowledge benefitted from the separated format. Nevertheless, all three animations did not differ significantly regarding knowledge acquisition (p=0.72).
Scientific or scholarly significance of the studies: Results indicate that differences in knowledge acquisition from animations with text or speech result from a split attention effect rather than from a difference in processing different modalities. This is in line with findings from eye-tracking research (e.g., Schüler et al., 2012). Although the second experiment as described here failed to replicate the expertise-reversal effect, outcomes support the assumption that the modality principle is more than questionable from a theoretical and an empirical point of view.

Authors

• Joerg Zumbach, University of Salzburg

• Stephanie Moser, University of Salzburg

• Ines Deibl, University of Salzburg