The allocation of human participants to resources was studied by observing the population dynamics of people interacting in real time within a common virtual world. Resources were distributed in two spatially separated pools with varying relative reinforcement rates (50-50, 65-35, or 80-20). We manipulated whether the participants could see each other and the distribution of the resources. When the participants could see each other but not the resources, the richer pool was underutilized. When the participants could see the resources but not each other, the richer pool was overutilized. In conjunction with prior experiments that correlated the visibility of agents and resources (Goldstone & Ashpole, 2004), these results indicate that participants' foraging decisions are influenced by both forager and resource information. The results suggest that the presence of a crowd at a resource is a deterring, rather than an attractive, factor. Both fast and slow oscillations in the harvesting rates of the pools across time were revealed by Fourier analyses. The slow waves of crowd migration were most prevalent when the resources were invisible, whereas the fast cycles were most prevalent when the resources were visible and the participants were invisible.
How do groups of people allocate themselves to resources? Do people, like many other animal species (Fretwell & Lucas, 1970), distribute themselves so as to cover resources in an approximately optimal fashion? In this work, we were particularly interested in the role of knowledge in influencing group foraging behavior. We manipulated two types of information that animals might use in determining where to allocate their foraging time. These sources of information are the distributions of resources and of other foragers.
Undermatching is empirically found more often than overmatching. In Kennedy and Gray's (1993) meta-analysis, values of s less than 1 were found for 44 out of 52 experiments, and the mean value of s was 0.7. If we restrict our attention to the relatively few laboratory studies in which group foraging in humans has been explored, we also find consistent evidence for undermatching (Goldstone & Ashpole, 2004; Kraft & Baum, 2001; Madden, Peden, & Yamaguchi, 2002; Sokolowski, Tonneau, & Freixa i Baqué, 1999).
Forager and Resource Information
Several empirical inquiries have explored the influence of information limits on foraging behavior. Baum and Kraft (1998) found that the presence of visual barriers that prevented pigeons from simultaneously seeing two patches had no effect on undermatching. Similarly, some studies have shown that animals make little use of other animals' foraging successes in allocating their own foraging time to patches (Valone & Giraldeau, 1993). However, other studies have shown that animals use both their personal foraging histories and vicariously obtained information from observing other foragers' successes and failures to shape their strategies and that the use of vicarious information increases with the difficulty of obtaining accurate personal information (Templeton & Giraldeau, 1996). Such vicarious information is useful in group foraging situations because one's conspecifics can act as scouts for assessing patches that one has not personally visited. Assuming that an organism uses information about the foraging success or failure of its competitors, it is clear how this information should be used. Patches where many successful foraging events have happened should be visited relatively often.
It is less clear how information regarding the sheer number of competitors should be used to guide foraging strategies. On the one hand, an animal may be attracted toward patches occupied by its conspecifics. An animal can use the prevalence of conspecifics in a patch as information that the patch is highly productive. In accord with this hypothesis, field experiments on migratory birds have shown that the presence of birds attracts other birds to the region (Pöysä, Ehnberg, Sjöberg, & Nummi, 1998; Stamps, 1988). …