ARTIFICIAL CREATIVITY: A SYNTHETIC APPROACH TO THE STUDY OF CREATIVE BEHAVIOUR ROB SAUNDERS AND JOHN S. GERO The Key Centre of Design Computing and Cognition Department of Architectural and Design Science University of Sydney, NSW, 2006, Australia e-mail: {rob, john}@arch.usyd.edu.au Abstract We present a novel approach to the computational study of creativity, called artificial creativity. Artificial creativity promotes the study of the creative behaviour of individuals and societies in artificial societies of agents. It is similar to the approach to that taken by Artificial Life researchers involved in developing computational models. We present a framework for developing artificial creativity systems as an adaptation of Liu’s dual generate-and-test model of creativity. An example implementation of an artificial creativity system is presented to illustrate the potential benefits of our new approach as a way of investigating the emergent nature of creativity in societies of communicating agents. Finally, we discuss some future research directions that are possible by extending the abilities of individuals and studying the emergent behaviour of societies. 1. INTRODUCTION The aim of artificial creativity is to gain a better understanding of creativity-as-it-is in the context of creativity-as-it-could-be. In other words, it is the comparative study of creativity as it is found in human societies against creativity as it may be found in artificial societies of agents that may follow quite different social conventions. In this way, the study of artificial creativity is similar to the study Artificial Life; both are synthetic approaches to understanding a complex, and ill-defined behavioural phenomenon, i.e. creativity and life respectively. The artificial creativity approach provides an opportunity for researchers to study the emergence of creative behaviour in controllable environments, affording a number of possible studies not possible in the real world. The parameters that control the behaviour of individuals can be experimented with to study the affect that they have on the emergence of social structures. In addition, the environment that the society of agents is situated in, e.g. economic conventions, can be adjusted to study the affects that external factors have on the creativity of individuals and societies. As with Artificial Life, one of the most interesting possibilities of artificial creativity is to be able to re-run history with different starting conditions to find out how products of creative individuals and the structures of creative societies might have differed. For example, by re-running an artificial creativity simulation with different communication policies we can simulate the affect that different communication technologies might have on the development and dissemination of creative ideas. Artificial creativity is compatible with previous approaches to studying creativity that have developed computational models of creative thinking by allowing them to