Applying Social Norms to Implicit Negotiation among Non-Player Characters in Serious Games

Believable Non Player Characters (NPCs), i.e. arti-ficial characters simulating rational entities, are a great additionto videogames, no matter if used for entertainment or seriousreasons. Especially NPCs that represent people in realistic set-tings need to show plausible behaviors; to this end, one of mainissues to be tackled is coordination with other participants, eitherother NPCs or human players, when performing everyday taskssuch as crossing doors, queuing at an office, picking the firstfree object up from a set, and so on. Much of this coordinationhappens silently and is driven by social norms that may varyaccording to culture and context. In this paper, we propose anapproach to represent social norms in autonomous agents andenable implicit coordination driven by observations of others’behavior. Our approach does not use central coordinators or acoordination protocol, but rather let each agent take its owndecision so to support more realistic interactions with humanplayers. A software architecture and initial experimental resultsare presented and discussed.