Development of Learning Support System for Acquisition of Convincing Argument Methods

We developed and evaluated a learning support system for learning logical reasoning. Focusing on the ability of convincing argument, we formulated a method of improving the convincing power of deductive and abductive arguments, then defined seven educational models based on the method called “Argument Unit.” We implemented a system in which learners assemble the argument structure of a given document with nodes of proposition and links of their relationship. The system then compares the structure with Argument Units and gives appropriate feedback to learners if the structure is incomplete. We conducted an experiment for evaluating the effectiveness of learning with the system. Nine university students learned a set of ways of making a convincing argument and took pre-/post-test. Both tests included 5 learning tasks. Post-test also included 5 transfer tasks. As a result, in learning tasks, scores were significantly higher in the post-test than in the pre-test. On the other hand, in transfer tasks, scores were also significantly higher in the post-test than in the pre-test. This result indicated a certain degree of learning effect in the system-based learning immediately after learning. However, the results of the task-by-task analysis suggest that the effectiveness of this system varies greatly depending on the type of argument units handled in tasks. In upcoming research, we plan to analyze whether the learning effect is maintained after a certain period of time has elapsed after learning with the system in this experiment.