This volume examines all aspects of using agent or individual-based simulation. This approach represents systems as individual elements having their own set of differing states and internal processes. The interactions between elements in the simulation represent interactions in the target systems. What makes this "social" is that it can represent an observed society. Social systems include all those systems where the components have individual agency but also interact with each other. This includes human societies and groups, but also increasingly socio-technical systems where the internet-based devices form the substrate for interaction. These systems are central to our lives, but are among the most complex known. This poses particular problems for those who wish to understand them. The complexity often makes analytic approaches infeasible but, on the other hand, natural language approaches are also inadequate for relating intricate cause and effect. This is why individual and agent-based computational approaches hold out the possibility of new and deeper understanding of such systems. This handbook marks the maturation of this new field. It brings together summaries of the best thinking and practices in this area from leading researchers in the field and constitutes a reference point for standards against which future methodological advances can be judged. This second edition adds new chapters on different modelling purposes and applying software engineering methods to simulation development. Revised existing content will keep the book up-to-date with recent developments. This volume will help those new to the field avoid "reinventing the wheel" each time, and give them a solid and wide grounding in the essential issues. It will also help those already in the field by providing accessible overviews of current thought. The material is divided into four sections: Introduction, Methodology, Mechanisms, and Applications. Each chapter starts with a very brief section called 'Why read this chapter?' followed by an abstract, which summarizes the content of the chapter. Each chapter also ends with a section on 'Further Reading'. Whilst sometimes covering technical aspects, this second edition of Simulating Social Complexity is designed to be accessible to a wide range of researchers, including both those from the social sciences as well as those with a more formal background. It will be of use as a standard reference text in the field and also be suitable for graduate level courses.

This book explores the view that normative behaviour is part of a complex of social mechanisms, processes and narratives that are constantly shifting. From this perspective, norms are not a kind of self-contained social object or fact, but rather an interplay of many things that we label as norms when we ‘take a snapshot’ of them at a particular instant. Further, this book pursues the hypothesis that considering the dynamic aspects of these phenomena sheds new light on them. The sort of issues that this perspective opens to exploration include: Of what is this complex we call a "social norm" composed of? How do new social norms emerge and what kind of circumstances might facilitate such an appearance? How context-specific are the norms and patterns of normative behaviour that arise? How do the cognitive and the social aspects of norms interact over time? How do expectations, beliefs and individual rationality interact with social norm complexes to effect behaviour? How does our social embeddedness relate to social constraint upon behaviour? How might the socio-cognitive complexes that we call norms be usefully researched?

This book explores the view that normative behaviour is part of a complex of social mechanisms, processes and narratives that are constantly shifting. From this perspective, norms are not a kind of self-contained social object or fact, but rather an interplay of many things that we label as norms when we 'take a snapshot' of them at a particular instant. Further, this book pursues the hypothesis that considering the dynamic aspects of these phenomena sheds new light on them. The sort of issues that this perspective opens to exploration include: Of what is this complex we call a "social norm" composed of? How do new social norms emerge and what kind of circumstances might facilitate such an appearance? How context-specific are the norms and patterns of normative behaviour that arise? How do the cognitive and the social aspects of norms interact over time? How do expectations, beliefs and individual rationality interact with social norm complexes to effect behaviour? How does our social embeddedness relate to social constraint upon behaviour? How might the socio-cognitive complexes that we call norms be usefully researched?

Socially situated planning provides one mechanism for improving the social awareness ofagents. Obviously this work isin the preliminary stages and many of the limitation and the relationship to other work could not be addressed in such a short chapter. The chief limitation, of course, is the strong commitment to de?ning social reasoning solely atthe meta-level, which restricts the subtlety of social behavior. Nonetheless, our experience in some real-world military simulation applications suggest that the approach, even in its preliminary state, is adequate to model some social interactions, and certainly extends the sta- of-the art found in traditional training simulation systems. Acknowledgments This research was funded by the Army Research Institute under contract TAPC-ARI-BR References [1] J. Gratch. Emile: Marshalling passions in training and education. In Proceedings of the Fourth International Conference on Autonomous Agents, pages 325-332, New York, 2000. ACM Press. [2] J. Gratch and R. Hill. Continous planning and collaboration for command and control in joint synthetic battlespaces. In Proceedings of the 8th Conference on Computer Generated Forces and Behavioral Representation, Orlando, FL, 1999. [3] B. Grosz and S. Kraus. Collaborative plans for complex group action. Arti?cial Intelli gence, 86(2):269-357, 1996. [4] A. Ortony, G. L. Clore, and A. Collins. The Cognitive Structure of Emotions. Cambridge University Press, 1988. [5] R.W.PewandA.S.Mavor,editors. Modeling Human and Organizational Behavior. National Academy Press, Washington D.C., 1998.

This volume presents revised versions of the papers presented at the 4th International Workshop on Multi-agent Based Simulation (MABS 2003), a workshop federated with the2ndInternationalJointConferenceonAutonomousAgentsandMulti-agentSystems (AAMAS 2003), which was held in Melbourne, Australia, in July 2003. In addition to the papers presented at the workshop, three additional papers have been included in this volume (Robertson, Noto et al., and Marietto et al.). Multiagent Based Simulation (MABS) is a vibrant interdisciplinary area which brings together researchers active within the agent-based social simulation community (ABSS) and the multiagent systems community (MAS). These two communities have different, indeed somewhat divergent, goals. The focus of ABSS is on simulating and synthesizing social behaviors in order to understand observed social systems (human, animal and even electronic) via the development and testing of new models and c- cepts. MAS focuses instead on the solution of hard engineering problems related to the construction, deployment and ef?cient operation of multiagent-based systems.

Socially situated planning provides one mechanism for improving the social awareness ofagents. Obviously this work isin the preliminary stages and many of the limitation and the relationship to other work could not be addressed in such a short chapter. The chief limitation, of course, is the strong commitment to de?ning social reasoning solely atthe meta-level, which restricts the subtlety of social behavior. Nonetheless, our experience in some real-world military simulation applications suggest that the approach, even in its preliminary state, is adequate to model some social interactions, and certainly extends the sta- of-the art found in traditional training simulation systems. Acknowledgments This research was funded by the Army Research Institute under contract TAPC-ARI-BR References [1] J. Gratch. Emile: Marshalling passions in training and education. In Proceedings of the Fourth International Conference on Autonomous Agents, pages 325-332, New York, 2000. ACM Press. [2] J. Gratch and R. Hill. Continous planning and collaboration for command and control in joint synthetic battlespaces. In Proceedings of the 8th Conference on Computer Generated Forces and Behavioral Representation, Orlando, FL, 1999. [3] B. Grosz and S. Kraus. Collaborative plans for complex group action. Arti?cial Intelli gence, 86(2):269-357, 1996. [4] A. Ortony, G. L. Clore, and A. Collins. The Cognitive Structure of Emotions. Cambridge University Press, 1988. [5] R.W.PewandA.S.Mavor,editors. Modeling Human and Organizational Behavior. National Academy Press, Washington D.C., 1998.

Social systems are among the most complex known. This poses particular problems for those who wish to understand them. The complexity often makes analytic approaches infeasible and natural language approaches inadequate for relating intricate cause and effect. However, individual- and agent-based computational approaches hold out the possibility of new and deeper understanding of such systems.

Simulating Social Complexity examines all aspects of using agent- or individual-based simulation. This approach represents systems as individual elements having each their own set of differing states and internal processes. The interactions between elements in the simulation represent interactions in the target systems. What makes these elements "social" is that they are usefully interpretable as interacting elements of an observed society. In this, the focus is on human society, but can be extended to include social animals or artificial agents where such work enhances our understanding of human society.

The phenomena of interest then result (emerge) from the dynamics of the interaction of social actors in an essential way and are usually not easily simplifiable by, for example, considering only representative actors.

The introduction of accessible agent-based modelling allows the representation of social complexity in a more natural and direct manner than previous techniques. In particular, it is no longer necessary to distort a model with the introduction of overly strong assumptions simply in order to obtain analytic tractability. This makes agent-based modelling relatively accessible to a range of scientists. The outcomes of such models can be displayed and animated in ways that also make them more interpretable by experts and stakeholders.

This handbook is intended to help in the process of maturation of this new field. It brings together, through the collaborative effort of many leading researchers, summaries of the best thinking and practice in this area and constitutes a reference point for standards against which future methodological advances are judged.

This book will help those entering into the field to avoid "reinventing the wheel" each time, but it will also help those already in the field by providing accessible overviews of current thought. The material is divided into four sections: Introductory, Methodology, Mechanisms, and Applications. Each chapter starts with a very brief section called 'Why read this chapter?' followed by an abstract, which summarizes the content of the chapter. Each chapter also ends with a section of 'Further Reading' briefly describing three to eight items that a newcomer might read next.

"

This text compiles research from a vibrant social simulation community of researchers who have developed unique and innovative approaches to social simulation.