Abstract:

Education or training through online games is one of the most effective strategies to achieve positive results in a short time, with engagement and satisfaction of learners. This is particularly true when the contents of an e-learning course are hard to translate into theoretical terms, and concern practices and patterns of flexible behaviors. Think about training courses designed to develop practical skills regarding sales techniques or the relationship between customer and retailer, or about digital products created to help children cope with different forms of disease by applying theoretical information and practical suggestions.

In all these situations, disguising the course contents under the surface of a linear videogame (with sequential or tree structure), with high-quality graphics and multimedia effects is not enough. Both for adults and children, learning objects should consist of plausible situations that reproduce reality, actively engage learners and let them immediately see the positive or negative effects of their decisions. This is possible only if e-learning products, apart from combining training and game, succeed in “simulating” the context where to make choices, by telling a “story” and reconstructing its main features in a valid way. Therefore, instructional design needs to focus more on the logic of the game than on its layout, in order to reflect its underlying complexity, make it more challenging, and avoid obvious or predictable situations.

The focus on the “intelligence” of the game led the authors to refine a design method for digital edugames and integrate neural algorithms in the calculation engine. In particular, they developed some learning objects with a “brain” composed of MultiLayer Perceptron (MLP) Artificial Neural Networks. Thanks to this mathematical brain, edugames can model and simulate the different variables in play, by proposing contexts that are plausible (from a logical point of view) and stimulating, and where learners can make choices and learn in a significant way.

The first example of the NeuralEduGaming method concerns an app that simulates a sequential sales process, from the analysis of the products to the negotiation with the buyer. The player has to analyze information and unexpected events, make an offer to sell the products and strategically manage the interactions with the other characters in the story.

The second example is a “neural” videogame reproducing the mechanism of interaction between a retailer and a customer in a natural and organic cosmetics store. The game reproduces the logic of subjective observation with multiple options and unpredictable scenarios.
The last example consists of an app for children suffering from a serious hereditary disease, where players take care of an avatar, by performing a series of different actions (e.g. eat, drink, play sports, go to school, etc.), with the aim of keeping the physiological values under control. This is an “infinite” circular simulation, totally focused on an open reticular mechanism, whose main objective is to maintain a balance among the different clinical factors in competition.

Keywords:

Edugame, Artificial Neural Networks, MultiLayer Perceptron, learning object, game, learning, instructional design.