Animated Pedagogical Agents as a non-restrictive approach to Gaming The System

Intelligent Tutoring Systems have a big potential supporting the learning process because they provide personalized assistance. However, many times the effectiveness of these systems is reduced because of some student’s behaviors. One of these cases is the Gaming The System, when the student try to explore the tutor features in order to have a faster progress in the problem solving. In order to minimize Gaming The System and improve student’s experience, this paper presents a non-restrictive approach to the behavior. We developed Animated Pedagogical Agents with different kinds of personality. The agents react when the learner “Game The System” by help seeking and show that they recognize the student’s inappropriate behavior. The two different approaches show that the tutor can influence the way how the student face the experience through its actions and, in this way, discourage this practice. Even if the results of a research with 41 students from a private school are not statistically significant, they reveal that the agents with different personalities have a different impact in the learner’s behavior.