Agent-based participatory simulation activities for the emergence of complex social behaviours Stefano Cacciaguerra, Matteo Roffilli Department of Computer Science, University of Bologna Mura Anteo Zamboni 7, 40127 Bologna, Italy {scacciag, roffilli}@cs.unibo.it Abstract Nowadays, social organizations (at macro-level) can be represented as complex self-organizing sys- tems that emerge from the interaction of complicated social behaviours (at micro-level). Modern multi-agent systems can be employed to explore “artificial societies” by reproducing complicated social behaviours. Unfortunately, promoting interactions only among pre-set behavioural models may limit the capability to explore all possible evolution patterns. To tackle this issue, we aim at discovering emergent social behaviours through simulation, allowing human people to participate in the simulation environment, so that the range of possible behaviours is not pre-determined. In order to support this new approach, we propose a system architecture that is able to support an endless session level between a software agent and a human player (called participatory framework). In par- ticular, while network faults or human low reactivity do not allow the human being to control his agent, this system architecture adopts a virtual player mechanism (called ghost player) that takes control of the agent driven by the user. The advanced version of such a ghost player relies on sub- symbolic Machine Learning techniques for mimicking the strategy of the off-line human being. 1 Introduction Social organizations can be studied at many differ- ent levels of abstraction and analysis. Historically, in the analysis of organizational decision-making processes, a common strategy is to reduce a com- plex social activity to a single constrained optimisa- tion problem that is solved by means of a (macro- level) function. Nowadays, social organizations can be approached as complex self-organizing systems that emerge from the interaction of complicated social behaviours (at micro-level) (Lomi et al., Groningen 2003). This approach makes possible to explore the connection between the micro-level be- haviour of individuals and the macro-level patterns that emerge from the interaction of many individuals (Lomi et al., Notre Dame 2003). It is possible to effectively describe these behaviours as the actions of agents into an environment, where the agents are the individuals and the environment is the complex self-organizing system. We consider an agent as a computer system capable of independent actions in order to satisfy its planned objectives. In particular, to describe a complex self-organizing system we need several individuals, while to reproduce it, we need several agents. Along with this consideration, a multi-agent system can be successfully employed, in order to describe self-organizing systems. A multi- agent system is an environment that consists of a number of agents, which interact with one-another. Therefore, it is possible to reproduce social societies into a synthetic environment by creating “artificial societies”. To successfully mimic real societies, the multi-agent systems make the agents interact thanks to their ability to cooperate, coordinate, and negoti- ate (Stone et al., 2000). Nowadays, multi-agent sys- tems are used for educational purposes. For exam- ple, a multi-agent system could be used as a com- puter-based learning environment to teach students of social and economic schools a number of central issues when studying organizational and decision- making processes, and the respective representation of problems (Chen et al., 1993; Colella et al., 1998). These “artificial societies” create a quasi- experimental observation-generation environment where it is possible to conduct tests. Modern multi- agent systems can be employed to explore multiple phenomena from natural to social ones by involving different disciplines: art, biology, chemistry, phys- ics, computer science, earth science, games, mathe-