347

Measurina Behaviour: An Introductory Guide. Paul Martin and Patrick Bateson,

Cambridge University, Cambridge, 1986, 200 pp. ISBN 32368-l Price: 620.00 zyxwvutsrqponm
hardback
6.95 paperback.

As any science matures, its practioners turn their attention to explicit

statements of its formal methodology. Within behavioural biology, T.C.

Schneirla’s 1950 classic “The relationship between observation and

experimentation in the field study of behavior” has long been an

influential contribution in this regard. More recently, several fairly

high-level texts have roamed this terrain. In the present book, Martin and

Bateson have prepared a brief and clear introduction to the principles and

methods of quantitative studies of behaviour, directed toward undergraduate

and graduate students. Within a framework of the steps involved in

behavioural research, they begin with a consideration of general issues,

research design, and preliminaries: the importance of measurement, Niko

Tinbergen’s programmatic four problems of causation, ontogeny, function, and

evolution, the choice of species and ievel of analysis, establishing

ethograms, the problems of anthropomorphism and observer effects, and

experimental designs, with attention to such matters as confounding and

correlated observations. There is an approving presentation of the

discussion by Lee Machlis and co-workers of the “pooling fallacy” with

respect to repeated observations from a small number of animals. Martin

and Bateson conclude that “repeated measurements from the same subject

should be averaged to give a single data point for that subject and the

sample size (n_) should be equal to the number of subjects, not the number of

measurements” (p. 291. However, levels of analysis, discussed earlier by

the authors, are relevant to this conclusion: behavioural research focuses

on actions, not animals, and appropriate use of nested variance analysis,

mentioned by Martin and Bateson, enables a partitioning of these levels.

The literal invocation of this conclusion could be unfortunate both

intellectually and ethically: intellectually, individual differences, which

are increasingly being realized as an important research topic, might be

348

ignored, and ethically, unnecessarily large numbers of animals might be

employed in experiments.

The book then reviews aspects of data recording, including observations

on behavioural rhythms, the reliability and validity of measures, and the

controversy over the utility of one-zero sampling. While a detailed

presentation of multivariate statistics is beyond the scope of the book, the

necessity for these tools to tackle the inherently multidimensional problems

of behavioural biology, and indeed of biology generally, is clear.

The latter chapters turn to field studies, social behaviour, data

analysis, and miscellaneous issues such as experimental protocols and data

presentation. Field researchers will certainly benefit from the advice

offered, as will students of social systems unraveling individuals,

relationships, and group dynamics. There is a good discussion of

statistical issues including exploratory and confirmatory investigation, the

importance of correlations and homogeneous populations, and sources of

error. For all workers who tend to become preoccupied with a particular OC-

level, the distinction of statistical significance and the size of an effect

is essential. There is a presentation of attempts to measure the duration

of bouts of activity. Given that the temporal patterning of behaviour is

recognized as the output of a number of stochastic generators, such attempts

appear quixotic. It is more enlightening to focus on the properties of the

underlying generating process rather an arbitrary measure which is

inevitably subject to error. Finally, the attention paid to the continuing

difficulty of uniform ethical guidelines is entirely appropriate in an

intellectual community with rising concern for animal welfare. This

difficulty stems from the diversity of national patterns on animal ethics,

from the strong anti-vivisectionist tradition in Britian through the North

American position to the less restrained activity of other countries.

Each chapter usefully concludes with annotated references. There is an

index and appendices which present guidelines for the use of animals in

research, a schematic for an electronic beeper for time sampling, and a

table of SI units. The text is light on illustrations and on examples in

order to focus on general issues. Martin and Bateson have pulled together,

or make reference to, an impressive array of material. They deliberately

avoid problems belonging to related disciplines, such as physiology,

genetics, and ecology, as well as specific issues in behavioural biology or

the philosophy of science per se. The space dedicated to electronic data

processing reflects the impact which technology has on science. The brisk

and friendly style avoids didactic aridity. Overall, the book serves well

as a primer for its intended audience.

Patrick Colgan
Biology and Psychology
Queen’s University at Kingston
Ontario, K7L 3N6 CANADA

Altruism andAq;gression : Biologicpl___and __$c_cL@> Oripins. Carolyn Zahn-

Waxier, E. Mark Cummings and Ronald Iannotti (Eds), Cambridge University
Press, Cambridge, London and New York, 1986; pp.337. ISBN 0 521 26834 6.
Price E30..00 - $39.50.

This book is the outcome of a conference held in 1982 at The National

Institute of Mental Health, Bethesda, Maryland. The editors claim to have
brought together an interdisciplinary group of scientists to “discuss the
interconnections of altruism and aggression within and across species”. The
expressed aims of the gathering were to discuss 1) genetic transmission of
affiliative and aggressive behaviours together with their biological
correlates, 2) environmental factors that influence prosocial and antisocial
behaviours and 3) internal processes that regulate the expression of selfish
or kind behavior.

The book contains several interesting, well-written chapters along

with effective Introduction and Conclusion Sections by the editors and a
useful subject index. Jaak Panksepp supplies a wide-ranging review on “The
Psychobiology of Prosocial Behaviors” and Robert Cairns “An Evolutionary and
Developmental Perspective on Aggressive Patterns”. Youniss then looks at
“The Development of Reciprocity Through Friendship” and Strayer and Noel at
“The Prosocial and Anti-social Functions of Preschool Aggression” - both
studies being exclusively devoted to school children. Ervin Staub considers
the determinants (motives, the self and the environment) and development of
altruism and aggression. Cummings et al give an account of “Early
Organization of Altruism and Aggression : Developmental Patterns and
Individual Differences” in children. Seymour and Norma Feshbach review
aggression and altruism in relation to the child’s personality structure and
dynamics. Again from the human point of view, Grusec and Dix consider “The
Socialization of Prosocial Behavior” examining in particular, the likely
reasons for discrepancies between psychological theory’s claims concerning
the most effective ways of training children to adapt to the values and
demands of society and what parents actually &. They speculate on the