Partial preferences and foraging

Abstract

The maximization of the rate of energy gain in the standard model of prey choice requires that a prey type is either always accepted or always rejected. In contrast to this, most published data indicate that a prey type is sometimes accepted and sometimes rejected, i.e. partial preferences occur. Various ways in which partial preferences can be optimal are reviewed. A common feature of many of these explanations is that the standard model is too simple: states that are equivalent in it are different in a more appropriate model. Averaging over these states produces the appearance of partial preferences. A model in which the optimal action is not always chosen is explored. The idea that costly deviations from optimality should be rare is combined with a psychological choice principle to yield a general framework for suboptimal choice. When reward rate is maximized subject to the limitations of the choice principle, partial preferences are observed. When the probability of starvation is minimized these preferences depend on the animal's reserves and the time until the end of the foraging period.