Effects of Prior Knowledge on Metacognitive Scaffolding in Simulation-Based Learning

• In Event: Learner Support in Digital Learning Environments

Abstract

Objectives and Purpose
The goal of this experiment was to investigate the impact of prior knowledge and metacognitive scaffolding during self-regulated learning from simulations. We analyzed the impact of metacognitive scaffolding by means of prompting on learning outcomes and cognitive load when learning from an interactive simulation.
Theoretical Framework
Providing metacognitive support when learning with interactive media such as hypermedia (e.g., Bannert, Hildebrand, & Mengelkamp, 2009) or simulations (Moser, Zumbach, & Deibl, 2017) have been shown as effective scaffolding techniques. Nevertheless, it remains unclear how learners with different levels of expertise benefit from metacognitive support, either direct by training or indirect by prompting. Thus, this experiment aims on investigating a possible expertise-reversal effect assuming that learners with a low level or prior knowledge benefit from metacognitive prompting and training, while learners with high prior knowledge do not.
Method
Overall, 123 university students participated in this 2x2x2 factorial experiment. The level of expertise was varied by giving one group of participants a training within the domain of the simulation. Another independent variable was the provision of a training in metacognitive strategies prior to working with the simulation (with vs. without). The third factor was the provision of metacognitive prompting during working with the simulation (with vs. without). Participants were randomly assigned to one of the eight conditions.
Data Source and Material
A pre- and post-test assessing knowledge acquisition, metacognitive behavior, and cognitive load were applied. The learning environment was a simulation about natural selection. We hypothesized that students with scaffolding outperform the groups without scaffolding in knowledge acquisition. We also expected advantages of the metacognitive training. In addition, an expertise-reversal-effect was expected, assuming that learners with low prior knowledge benefit from metacognitive support while learners with high prior knowledge do not benefit and report a higher cognitive load.
Results
Results of an ANCOVA and descriptive data reveal that indeed, extraneous and intrinsic cognitive load are higher for learners with high prior knowledge when receiving additional metacognitive support (η2 = .58). Nevertheless, this does not affect learning performance. With regard to metacognitive scaffolding, we find a significant effect showing that prompting improves learning outcomes (η2 = .13): Metacognitive training had no significant effect on learning outcomes.
Scientific Significance of the Study
Results suggest that metacognitive support is not always beneficial but could rather increase cognitive load and, thus, might hinder learning. Especially in combination with high prior knowledge, it does not really contribute significantly to learning performance. Nevertheless, indirect metacognitive support led to improved learning with the simulation while direct support had hardly any impact here.

References
Bannert, M., Hildebrand, M. & Mengelkamp, C. (2009). Effects of a metacognitive support device in learning environments. Computers in Human Behavior, 25, 829-835.
Moser, S., Zumbach, J., & Deibl, I. (2017). The Effect of Metacognitive Training and Prompting on Learning Success in Simulation-based Physics Learning. Science Education, 101, 944-967.

Authors

• Viola Maria Geiger, University of Salzburg

• Joerg Zumbach, University of Salzburg

• Ines Deibl, University of Salzburg