Learning by problem-posing has been suggested as an important way to understand problem structure in order to solve arithmetic word problems. Previously, we have developed a Triplet Structure Model for arithmetic word problems solved by one addition or subtraction. The model has been implemented in a software to support learning by problem-posing in arithmetical word problems, named Monsakun, which was developed to work on tablet computers. This software has been put in practical use in elementary schools as well as evaluated by teachers, and it has been confirmed that this system was useful for learning the problem-posing activity. In this research, we conducted experiments to validate the use of Triplet Structure Model in multiple languages. A total of 41 international students participated in the Monsakun practical use in English and Indonesian language. The log data were analyzed and compared to Monsakun experimental use in Japanese.

The purpose of this work is to propose a new instructional design (ID) approach which integrates software engineering (SE) into traditional ID approaches to become suitable for producing virtual learning platforms (VLPs), especially in designing and developing phases. We have made analyses of previous studies in regard to ID, SE, and VLP to specify various features of the proposed ID approach. It was found that the traditional ID approaches have some deficiencies in methodology for designing and developing e-learning and virtual learning products, especially VLPs. Based on these analyses, we have reached to a new suggestion of ID approach, which incorporate some SE techniques so as to resolve the deficiencies of the traditional ID process especially in designing and developing phases. Our suggestion would lead people who have interested in ID to develop a new generation of instructional products in the future.

In this study, we consider a nonzero-sum bargaining game with imperfect information in which two players exchange resources to achieve individual goals. One of the two players learns how to achieve the goal through repeated games using Q-learing. Experiment results show that a learning player can learn winning strategies against cooperative and non-cooperative players.

In this study, we consider an environment in which multiple agents share information and cooperatively search objects. We then observe the evolution of self-interested agents who do not act cooperatively or even act dishonestly to increase their own profits. Experiment results show that such self-interested agents get high profits when a large number of cooperative agents exist.

While most of us don't pay much attention to the placement of objects in our daily life, focusing on the meaning of the placement of the objects may help us to acquire new concepts and viewpoints. Our goal is to develop a system promoting a person's awareness through the placement of objects. In this article, we present the results of two preliminary experiments conducted before developing the system. First experiment shows that considering the relationship between objects and humans is important for understanding the intensions of the person who placed the objects. Second experiment reveals that focusing on the difference between the placement by oneself and that by others, and then trying to build a consensus between them are likely to promote the awareness.

The human process of a motor-skill development is a kind of potential change on weight selection on the categorized memory about the skill. Therefore, the subsequent training and empirical knowledge based on the past experience makes sense. In this study, we design the promoting environment for development of hula hoop control skill by way of applying the inter-trial relationship as a feedback.