Book Reviews

Abstract

Houston, A.I. & McNamara, J.M. Models of Adaptive Behaviour. Pp. 338 . Cambridge University Press , Cambridge ( 1999 ). £55·00 (hardback), ISBN 0-521-38480-X ; £20·95 (paperback), ISBN 0-521-65539-0. Despite much criticism (Gould & Lewontin 1979; Pierce & Ollason 1987), the idea that animals optimize their behaviour in order to maximize their lifetime reproductive success is one that is powerful and useful for predicting the kinds of decisions animals make when foraging, avoiding predators and deciding clutch size. This book demonstrates how a variety of mathematical models can be constructed to determine the most advantageous strategy amongst a range of possible behaviours. Specifically it addresses the problem of how the state of the animal will affect its decision making. It seems intuitive that a starving animal will take bigger risks with regard to predation than a nearly satiated one, based on the probable consequences for fitness of neglecting a food source. This book demonstrates how these intuitive hypotheses can be formalized and exact optima obtained. The first chapter contains the briefest of introductions to the ecological problem and the philosophy of the approach. Alternative approaches are referred to but not discussed. The subsequent chapters introduce the concepts of state, decisions and fitness consequences firstly for a single decision and then for dynamic models with multiple decisions. The explanations combine clarity with rigour and the methods are illustrated with a worked example of a single day foraging strategy. Chapters 4, 5 and 6 deal with state-dependent optimal foraging theory, including the best strategies for dealing with uncertainty and the conflict between the need to obtain food and avoid predation. These chapters are very comprehensive, including much of the authors’ own work in the field, and show how constraints of time and space may be added to the models. Copious references are made to the source publications for these models and to related theoretical work. The optima are amply illustrated with line diagrams indicating the shapes of the fitness functions. The approach is unashamedly adaptationist although reference is made to other approaches, for example models that are constrained by perceptive and learning abilities (see Kacelnik & Bateson 1996 and papers in Dukas 1998). The last four chapters deal with more advanced topics. Game theory is discussed both in terms of static games, such as the Hawk-Dove game, which are similar to the models in the classic book of Maynard Smith (1982). The second half of the topic deals with dynamic models and demonstrates how the Evolutionary Stable Strategy for such games need not coincide with the Nash equilibrium for static games. Life history strategy (Chapters 8 and 10) and animal routines over daily and annual cycles (Chapter 9) necessarily involve complex mathematics including vector and matrix algebra and dynamic programming. The authors discuss how the ecological scenarios can be expressed in these formalisms but leave to others the explanation of the standard methods used. This is sensible because there are a number of computer packages that will perform the necessary calculations. This book will be of interest to anyone who wants to build or study behavioural ecology models. The mathematical knowledge assumed is not too great and any special conventions used (such as the use of E(x) to represent the mean of a distribution in (x) are explained when first met and there is plenty of clear but formal explanation. Many of the models used can be easily adapted to the kinds of real problems which will be encountered when seeking to explain the adaptive significance of animal decision making. The book will also be of interest to biologists who are not practitioners of modelling but who wish to understand the implications of models (who may wish to skip some of the mathematical details, especially the parts that are confined to the numerous appendices.) However the book is less well suited to those wishing to understand the reasons for animal behaviour; adaptationism is accepted a priori, with little explanation or refutation of alternative approaches, and there are few real biological examples. It should be regarded as a more detailed addition to books, such as Stephens & Krebs (1986) or Maynard Smith (1982), rather than as a replacement for them.