Memory Features in Simulated Students to Improve the Software Engineering Process and the Performance of Intelligent Tutoring Systems

This paper describes how memory features have been incorporated into the student modelling process of educational software to create simulated students. These simulated students are agents that simulate the way that students learn new facts from the domain being taught with respect to the amount of time that they can remember these facts. For this purpose, the student modelling process is based on principles of cognitive psychology. The resulting simulated students can be used to improve the software engineering process of the educational application and its performance. The software engineering process can be improved by the use of simulated students to evaluate educational software before it is delivered to real students. Moreover, the performance of the educational application can be improved since the simulated students can indicate how much new knowledge the real students may learn at a specific time and what needs to be reviewed. Thus, the system can plan the tutoring of new material and the revisions to be presented to students dynamically.