Reviewed by
Gönenç Yücel
Delft University of Technology
Complex dynamic phenomena have long been the subject of inquiry for researchers in various domains. Having its roots in these various domains such as biology, computer sciences, economics and sociology, complexity developed as a new strand of research during the last three decades. Mainly the emergence of the strand is due to the recognition of similar dynamic complex phenomena demonstrated by different systems in distinct domains, which lead to the conceptualization of a general type of system as the main subject of research, i.e. complex adaptive systems. Although it is possible to encounter a variety of characterizations of such systems, very roughly they are systems composed of many autonomous and adaptive agents, which interact with each other. Complexity research aims to explore characteristics of these systems in general that yield to such complex phenomena, and also benefit from the insights developed in understanding the dynamics of specific complex adaptive systems. Systems having the characteristics of complex adaptive systems can be identified in various domains, and the complexity perspective is inspiring a variety of research projects that seem to have very few in common. The book edited by Shan and Yang aims to provide a sketch of this variety by providing a mixed sample of studies on complex adaptive systems. The aim of the book is stated to be covering the state of the art in this highly evolving area and claimed to be an indispensible state of the art reference in the field by the editors, which seems to be a quite enthusiastic one considering the span of the complexity perspective and research on complex adaptive systems. At this point it is also worth mentioning that it is misfortunate that the title of the book gives the impression to the reader that complex adaptive systems is a tool or method, rather than a subject being studied.
The book is one of the two volumes that handle the subject, and this volume mainly focuses on the techniques and applications. The techniques section of the book is composed of four chapters. The chapter by Chiong introduces an agent-based economic where transactions are performed according to an iterated prisoner's dilemma (Agent Strategies in Economy Market). Proving an overview of game theoretic strategies that competed in Axelrod's competitions, Chiong studies the competing strategies in the market environment. In the second chapter, De Luca and Quattrociocchi introduce their study where they model context-aware interactions in autonomous systems using an agent-based model (Dynamic Contexts and Concepts as a Set of State Variations Under Emerging Functions: A Logical Model for Evolving Ontologies and Autopoietic Multi-Agent Systems). The third chapter of the techniques section, which seems to lack a coherent organization, is devoted to cellular automata (Cellular Automata, by Terry Bossomaier). Different from the former two, the chapter by Bossomaier is an overview of the approach, rather than a discussion based on a specific application. The final chapter of the section by Negrello and Huelse introduces an application of dynamical systems theory to the analysis of the structures and functions on recurrent neural networks (Adaptive Neurodynamics).
The second section of the book is devoted to the applications on complex adaptive systems. The chapter by Ratna et al. re-studies the research of Coleman on diffusion of new medication among doctors (Innovation Diffusion Among Heterogeneous Agents: Exploring Complexity with Agent-Based Modelling). Using the same dataset, the authors aim to explore sources of complexity in the diffusion process not addressed in the original work or its former re-analyses. Mainly they focus on the heterogeneity in terms of the degree of predisposition to knowledge. Satterfield discusses an agent-based model used to explore the linguistic evolution, more specifically the emergence of linguistic structures as a consequence of two different and interacting social groups, i.e. slaves and slave owners (Unique Applications of Multi-Agent Models in Uncovering Language Learning Processes). The chapter by Turini et al. covers the issue of social belief dynamics (The Intelligence of Rumors: A Cross-Methodological Approach to Social Belief Dynamics). The chapter lays the atomic components relevant to the spread of reputation, i.e. memetic, epistemic and pragmatic. Following that three model-based study is used to verify these three different components. The chapter by Reschke and Kraus, which proposes to focus on evolutionary processes of change and their implications for strategic planning and related issues of organization, gives the impression of a discussion rather than an application (Strategic Management, Evolutionary Economics, and Complex Adaptive Systems). The ninth chapter of the applications section stands a bit like an outlier among other chapters in the sense that it is the only one not focusing on social systems (Building Complex Adaptive Systems: On Engineering Self-Organizing Multi-Agent Systems, by Sudeikat and Renz). Sudeikat and Renz provide a new perspective from the viewpoint of distributed software systems domain. In their chapter they aim to examine the relation between complex systems and self-organizing multi-agent systems. The chapter by Outkin et al. introduces a big modeling study conducted in the Los Alamos National Laboratory (FinSim: A Framework for Modeling Financial System Interdependencies). The agent-based study discussed in the chapter aims to study the probable impacts of disruptions in communication and energy systems on the financial system. The chapter focuses more on the specifications of the model, and reserving very limited discussion on the experiments and outcomes. Holloman handles the issue of military transformation as an instance of large-scale organizational change and proposes that insights developed in the domain of CAS may contribute significantly (Complex Adaptive Systems Theory and Military Transformation). The chapter provides a review of transformation needs and former efforts in the US military, followed by a brief survey of social change theories as well as CAS theory. The final chapter of the book explores the use of evolutionary game theory to model the dynamics of adaptive opponent strategies for a large population of players (Insights into the Impact of Social Networks on Evolutionary Games, by Sycara et al.). Sycara et al. mainly focus on the propagation of information in the social networks and its effects in their chapter.
Consequently, the edited book is a compilation of diverse works focusing on different issues, but having a sort of complex adaptive systems perspective in common. However, it is hard to qualify the book as an indispensible reference about the issue of complex adaptive systems. The distinction between the techniques and the applications sections of the book is not that clear, since majority of the chapters in the former one also qualify for introduction of a specific application. In short, the first section fails to live up to expectations for giving an overview of techniques used in the field. The second section is composed of a set of interesting works. However, almost all of these studies (except one outlier) seem to focus on social/organizational systems. Hence, it is hard to conclude that the section covers the multiplicity of application domains where it is possible to find the influences of complexity and complex adaptive systems perspectives.
Return to Contents
of this issue
© Copyright Journal of Artificial Societies and Social Simulation, 2009