Agent-based computational modeling of emergent collective intelligence

Abstract

Collective Intelligence is a form of intelligence which emerges out of collaboration and coordination of many individual agents. A group of actors performing simple behaviours and interacting with fellow group members & the environment often produce global behaviours which seems intelligent. Understanding the emergence of intelligent collective behaviours in social systems, such as norms & conventions, higher level organizations, collective wisdom and evolution of culture from simple and predictable local interactions; has been an important research question since decades. Agent-based modeling of complex social behaviours by simulating social units as agents and modeling their interactions; provides a new generative approach to understanding the dynamics of emergence of collective intelligence behaviours. In this paper, we have presented an analytical account of nature, form and dynamics of collective intelligence, followed by some of our experimental work on evolution of collective intelligence. The paper concludes with a short discussion of the results and relevant implications for designing systems for achieving desired collective intelligence. © 2009 Springer Berlin Heidelberg.