Skip to main content
eScholarship
Open Access Publications from the University of California

Learning from Errors in Collaborative Learning: An Experimental Study of Erroneous Worked-Out Examples

Creative Commons 'BY' version 4.0 license
Abstract

Collaborative learning facilitates cognitive activities such as knowledge elaboration through interaction. However, when extraneous cognitive load is high, the cognitive resources available for engaging in learning processes may be reduced. Erroneous worked-out examples (EWoEs), which present solutions containing misconceptions, can induce cognitive conflict and facilitate knowledge revision. This study examined the effectiveness of EWoEs and their underlying mechanisms in collaborative learning. To enable adaptive example presentation based on learners' knowledge, we employed a knowledge generation model based on the Adaptive Control of Thought–Rational (ACT-R). We used drawing performance as an indicator of knowledge elaboration and the Interactive–Constructive–Active–Passive (ICAP) framework as an indicator of learning processes. EWoEs enhanced drawing performance and the co-occurrence of Constructive and Interactive modes. Moreover, such co-occurrences were associated with improvements in drawing performance. These results suggest that EWoEs facilitate learning processes involving knowledge elaboration triggered by misconceptions.