Behavioral Aspects of Predation

Abstract

In this paper, I discuss four hypotheses concerning mechanisms which might be of general importance in the behavior of predators: hunting by searching image, hunting by expectation, area-restricted search, and “niche” hunting. In addition, I review briefly two approaches which have been taken in studying predator behavior at a more general level: experimental component analysis and optimal foraging theory. I make no attempt to review comprehensively the literature on behavior of predators.

The concept of hunting by searching image has been used in a great variety of senses, and to be of value it should be used in a more restricted manner. Searching-image formation in the strict sense of “learning to see” has been demonstrated in several laboratory experiments and perhaps in some seminatural field experiments with birds. The real problem is that it is difficult to distinguish in operational terms between “learning to see” and some other phenomena which result in the predator’s preferring one prey type over another. This is especially true in the field; thus the value of the concept of “learning to see” in field studies must be questioned. Gibb’s hypothesis of hunting by expectation still needs to be tested in a critical way. Although the hypothesis was put forward in the context of birds hunting for insect larvae in pine cones, essentially the same problem confronts any predator hunting for clumped prey, namely, when to leave one clump and go on to the next one. Some alternative hypotheses are discussed. Area-restricted search after capturing a prey item seems to be a fairly widespread phenomenon. At least some predators can learn to alter their searching path to improve their exploitation of particular dispersions of prey. The hypothesis of “niche” hunting is supported by some laboratory studies and indirectly by several pieces of field evidence. Experimental component analysis is an attempt to investigate and synthesize all the behavioral factors underlying the response of a predator to changes in prey density. This approach has been more concerned with determining the existence and effect of particular behavioral processes than with elucidating the mechanisms underlying each process. So far it has only been used in studying the behavior of predators in simple laboratory situations. Optimal foraging models have provided useful hypotheses for looking at particular mechanisms. More general models seem to make either very “weak” predictions about behavioral mechanisms (ones which could be produced by many other hypotheses) or ones which are not easily testable.

Financial support was provided by grant N.R.C. 67–6295.