This study focuses on modeling perspective taking behavior during collaborative explanation activities and the process of searching for relevant knowledge for constructive interaction. A laboratory experiment was conducted where dyads learned about concepts and engaged in a reasoning task using concept maps through discussions. We collected leaners conversational data and gaze using two eye trackers to examine perspective taking behavior. The results showed that leaners who considered others’ diverse perspective expeditiously reflected upon their own knowledge and used the relevant knowledge for explaining the phenomenon to their partners. To further model the process of the perspective taking behavior and knowledge retrieval that occurred in the experiment, we developed a model using the cognitive architecture, Adaptive Control of Thought Rational (ACT-R). We further explained the mechanisms by implementing a model where perspective taking increases the opportunity to search and reconsidering the use common knowledge owing to the discrepancy triggered by diverse perspectives. This paper contributes to model-based research and provides knowledge on capturing the relationships of perspective taking and relevant knowledge retrieval in collaborative learning.
CITATION STYLE
Hayashi, Y., & Shimojo, S. (2022). Modeling Perspective Taking and Knowledge Use in Collaborative Explanation: Investigation by Laboratory Experiment and Computer Simulation Using ACT-R. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 13355 LNCS, pp. 647–652). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-11644-5_62
Mendeley helps you to discover research relevant for your work.