Imagine wanting to get on an airplane and having to choose between two pilots, one having flown for 30 hours and another having flown 100 hours but exclusively in a flight simulator, and you have vicariously experienced what this book is about: simulation, a method of scientific discovery whose proponents still have to fight battles on all fronts, for example, validity, correctness, and reuse.

Science at large is not only occupied with providing insight in nature, but erroneously seems to have a concomitant intrinsic obligation to compare any intelligent behavior to humans. To dispel this antiquated notion, the author discusses John Searle’s Chinese room thought experiment in chapter 1. But in contradistinction to Searle, who claims that a person not knowing Chinese but having unlimited access to instructions on how to behave as if he were a native speaker can never be said to really understand Chinese, the author believes that there are good reasons to assume that such a person can indeed be said to understand the language. Searle’s argument can be unhinged by pointing to the fact that it is the function of language in a given discourse and not the intentionality derived from a biological notion of understanding that enables programs to think [1].

Apart from being part of metaphysical arguments, simulation is an important part of scientific discovery. While mathematical models provide a top-down approach to inquiries and employ differential equations, simulation, especially if it is agent-based, can lead to surprising results when investigating complex systems by unlocking emergent properties that are dependent on the roles of each agent.

While chapter 1 lays the scientific and metaphysical groundwork for the book and delineates limitations and criticism of simulation, chapter 2 serves as an introduction to evolutionary computing, including genetic programming, algorithms, and interactive evolutionary computation.

Chapter 3 explains the Swarm modeling platform invented at the Santa Fe Institute and available for download under the open system licensing to test hypotheses faster, cheaper, better, and, above all, more consistently. Swarm is both a common language and an organizing framework for establishing agent-based modeling as a relevant tool for scientific investigations. Swarm consists of an object-oriented collection of active agents that interact with each other in a specific environment. Each such scenario can be based on an already built scene or on a newly (or to be) developed one. The author emphasizes that Swarm, with all of its potential for modeling phenomena, is but one simulation platform for agent-based modeling.

Chapter 4 uses genetic algorithms for evolutionary modeling. Simulation examples include sexual selection to explain the striking beauty of some animals, the prisoner’s dilemma, and an extended version of it: the iterative prisoner’s dilemma, which includes repeats of the experiment with the same participants. Chapter 5 is all about ants: trail simulation, ant colony optimization, and cooperative behavior among ants. Chapter 6 deals with the flocking behavior of birds and fish. Chapter 7 discusses cellular automata, including Game of Life, the Turing model, traffic simulation, and others. Chapter 8, the final chapter, discusses the importance of bottom-up simulation.

The book is very readable and contains great illustrations. Each chapter summarizes the problems addressed and the current state of the art, and eases into a detailed discussion on why agent-based modeling sheds new light on the topic at hand. The author performs a difficult task gracefully: he explains just enough for the reader to grasp the essence of a problem, while the bulk of the chapter is spent demonstrating the relevance of agent-based modeling in addressing it.

This is not an introductory book for agent-based modeling at large, even though the author recaps the essential parts of scientific theory at the beginning of each chapter for convenience. It is a collection of multidisciplinary overviews of how to employ agent-based modeling using Swarm as but one example of a modeling language dealing with artificial intelligence, artificial life, genetic algorithms, and other topics, and the prime issues and challenges these subject areas address, such as the traveling salesman problem and the prisoner’s dilemma.

In order to benefit from this book, the reader should be comfortable dealing with algorithms and should already have a theoretical and working knowledge of many of the topics covered. The author concludes: “This book would immensely contribute to research in the fields of complex system and artificial life.” It has certainly done that.