Using Giving-Up Densities to Detect Search Images

Abstract

Search images may influence diet choice, a predator's functional response, coexistence of prey species, the stability of predator-prey interactions, and the effectiveness of species for the biological control of pests. Despite these ascribed significances for behavioral, population, and community ecology, most tests of the search image concept have involved operant conditioning in laboratory experiments. We use foraging theory to model the consequences of a forager's behavior if it were using a search image to detect two types of patchily distributed prey. This requires us to distinguish between search images that depend only on the forager's encounter with specific food types (passive) and search images that involve the forager's choice of a specific search mode (active). Both active and passive search images predict that the forager will appear to use a resource patch containing both food types less thoroughly than patches containing just one of the food types. The giving-up densities on the two foods in a mixed patch will be higher than a weighted average of the giving-up densities on the two foods in single-food patches. We tested this prediction on free-living fox squirrels using triplets of feeding patches containing sunflower seeds, granulated peanuts, and a mixture of both, respectively. We compared the amount of food remaining in the patch (giving-up density) containing both food items with the patches containing only sunflower and only peanuts. Our results provide no evidence for the presence of a search image in the foraging behavior of fox squirrels.