Analysis of Learning Activities with Automated Auxiliary Problem Presentation for Breaking Learner Impasses in Physics Error-based Simulations

Abstract

Error-based Simulation (EBS) is a framework that supports learning by visualizing errors. EBS systems generate and visualize simulations of assumed phenomena from learner answers. By letting the learner observe the results, the system shows how wrong answers result in wrong phenomena. The system thus supports trial-and-error processes until a correct response results in correct phenomena. EBS systems effectively support trial-and-error in learning of physical mechanics. Despite this, learners may become stuck because the EBS system does not provide the correct answers to problems. Stepwise presentation of auxiliary problems can be useful for resolving such impasses. Auxiliary problems are simplifications of the original problem containing fewer targeted learning concepts, a scaffolding approach leading toward the original problem. Therefore, we developed a learning support system that adaptively presents auxiliary problems by determining from error trends the concepts that the learner does not yet understand. We also evaluated the developed system through experimentation. Using system logs from those experiments, we analyzed in detail the learning activities related to the auxiliary problem presentation function in EBS. The results suggested that adaptively presenting auxiliary problems to learners can effectively resolve impasses.