Learning in a community of practice : Complete vs. incomplete information

This paper is a contribution to the modelling of interactions within social networks. We use an agent-based model to explore the learning process, in a particular form of social network: a community of practice. It is based on an empirical study, made in a research centre in France. We shed light on the importance of two parameters in learning in communities : the availability of the community members, and the information they have about their neighbourhood. The first parameter represents their willingness and engagement in the development of their community, whereas the second one gives us an idea on agents' ability to build a correct shared representation of the competencies of the other agents. We lead two sets of simulations, simulations with complete information and simulations with incomplete information.