Educational system based on simulation and intelligent conversation

Abstract

This paper explores the use of intelligent conversation and simulation to increase learning outcomes of science topics, i.e. Physics in our case. The underlying theory is that educational 3D interactive simulations can enhance learning by, first of all, adding a playful element to the learning process which boosts motivation, and, second, by creating an opportunity to "learn by doing" when enabling the learner to explore the simulated phenomena by themselves. Furthermore, conversational agents can enhance and strengthen the simulation-based learning environment by playing the role of guide or tutor, e.g. offering hints when the student is struggling, as well as evaluator, which is needed to provide appropriate feedback and guidance. Subsequently, a prototype of an educational system that combines a Physics simulation and a conversational agent was developed and then evaluated. In order to evaluate the effectiveness of the proposed educational system to induce learning gains, we conducted a randomized controlled trial experiment. In particular, we compared the following experimental conditions: using a simulation alone followed by a full system that included both the simulation and a conversational agent versus the conversational agent alone followed by the full system. In both conditions, students were exposed to the same content.