A task annotation model for Sandbox Serious Games

SGs represent a great potential for education since they can join an instructional value with the appealing language of digital natives. Searching for methodologies to support effective authoring of serious games (SGs), we have abstracted a model consisting of a 3D geo-referenced virtual world (VW) where knowledge is implemented also through tasks that are disseminated across the environment. The model is designed to support authors to focus on defining the tasks´ content and annotation (e.g. relevance to various topics, difficulty, position in the VW, typology of supported learning styles, interrelation with other tasks and events, etc.), without being asked to do game scripting, which typically involves hardcoding events, actions, locations and contents of a specific game. Task annotation allows reusing tasks also in different games. Moreover, different game designers can exploit repositories of tasks and build educational games specifying a VW, and its missions as sequences of tasks that meet learning strategy criteria that can be expressed through the proposed framework. This model opens new perspectives to pedagogical experts and game designers to develop runtime game engines able to provide players with adaptive contents. The paper presents the user, task and learning strategy models in detail and an example of application.