This article describes the design and building process of a Learning Virtual Object that uses the Multiagent Systems theory to promote the thinking about advantages of cooperative work while solving problems. The software context is the environment care and for its execution users establish the cooperative conditions of three different robot groups responsible for collecting, classifying and storing three types of recyclable materials. Throughout the execution, the user evaluates if the established cooperative conditions are accurate to the correct task development. The software has been designed using Eclipse Kepler as IDE, Processing version 2.1 and the library AI for 2D Games and G4P to manage the robots states and the user interface respectively. The implementation results prove the software accuracy to learn how to cooperate in daily basis tasks through the cooperative robots programming, besides a conceptual change about cooperative concept that is evident in users, the team works the advantages and the distribute cognition principles for the cooperative tasks development.