Foraging behaviour in guinea pigs: further tests of the marginal value theorem

We studied the foraging behaviour of guinea pigs (Cavia porcellus) in laboratory environments with a single patch type. Six experiments were designed to test predictions of the marginal value theorem (MVT) for various foraging problems in a constant physical setting. In Part one (experiments I, I', and II) we used patches with resource depression. In experiment I there were two treatments, which differed in the function relating cumulative food gain to time in the patch. Experiment I' was a replicate of experiment I with greater differences in patch quality between treatments. In experiment II the treatments differed in the travel requirement between patches, while the patch gain function remained the same throughout. In experiments III, IV and V (Part two) there were patches with linear gain function and sudden exhaustion, and two treatments in each experiment. The treatments differed in prey encounter rate, maximal number of prey per patch, and travel time, respectively. In the six experiments the MVT predictions for prey per patch visit were qualitatively supported by the experimental results, and in most cases the quantitative fit was also good. Giving up times were longer than predicted. We conclude that the hypothesis of rate maximization, in spite of failing to predict some aspects of the results, provides a suitable framework for examining the foraging behaviour of this species.