Emergence is one of the most interesting concepts in nature, where global patterns emerge as a result of simpler entities and tasks interacting among themselves. As a modeling and simulation researcher, emergence is particularly challenging to model, as one has to deduce the inner workings of complex models comprised of multiple smaller elements and understand how they interact to produce emergent behavior. This scenario involves two layers: a lower level of smaller entities interacting, which is modeled and simulated, and an upper layer where the interactions produce an emergent global behavior that can be studied and verified to understand the system as a whole.

The book discusses these concepts of emergence, complexity, and how computation can support them. The book particularly focuses on multiagent modeling, a technique that breaks away from traditional methods of differential equation models and allows more intricate details of models to be written for modeling complex models. The technique has been used in modeling economic systems, tissues in biology, and social networks across the globe.

Multiagent modeling allows smaller elements to be computed as agents with memory, functions, and interactions; when working with the agents, they can produce emergent behavior of the system. The book focuses on the modeling of Physarum polycephalum, which is a slime mold single-celled organism known for its curious properties and behavior.

The book presents a full account of research and will be extremely useful for young and early career researchers exploring the topic of both multiagent modeling and the slime mold. It starts with the initial few chapters discussing the concepts and the techniques of modeling. It then goes into more complex computations of how the networks in the mold are modeled, how data is approximated, and verification techniques. The agent model is discussed in great detail with illustrations and specifications of the mold and how it adapts over time; all of these details result in creating a complex agent-based model used to study biological behavior. By creating a virtual model, the authors are able to run experiments on how certain substrates will affect the growth of the mold and the networks formed, making the virtual model very similar to the real model. The authors discuss neural networks, spanning trees algorithms, and many techniques to introduce complexity in learning of the model.

The book finishes with the main research findings. Pattern formation, biological behavior, and adaptation of the virtual model allow researchers to study the organisms in depth without the need for lab experiments with multiple in vitro petri dishes. Applying the learning from logic circuits and emergent transport networks, the authors demonstrate how simple organisms can reflect on engineering methods to build more complex and reliable structures in everyday life. The book also discusses the limitations of the approach in terms of preliminary results, which could be extended for future research.

In conclusion, as a multiagent modeler, I found the book a useful tool for teaching students how agent-based modeling works, its complexities, and how it can be used to extend our knowledge of complex models to build better and reliable structures inspired by nature.