An Analysis and Detection of the Opportunities for Learning Activities within Automatically Generated Educational Games

Abstract

We proposed a design method to automatically generate an educational game by substituting the cards of an existing card game for new cards that have practice problems written on them. However, several games in which most players rarely solved problems were created by the method. Therefore, this study examines the reasons why they rarely solved problems. We conducted an experiment that asked test subjects to play useful and useless educational games and their reactions were videotaped. The video recordings and interviews suggest that few subjects solved problem in playing the educational games had no opportunities for solving problems to predict the future game state that would result from each choice to make a favorable choice. On the other hand, all subjects solved problems in playing the educational game had opportunities for the prediction. On the basis of the results, this study also organizes the rules that provides the prediction, and develops a system that detects useless educational games. The results of experimental evaluations of the system suggest that the detection system can detect useless educational game by the detection of the rules for the prediction.