Andreas Wilke: UCLA Department of Anthropology

When resources are distributed in patches animals must decide when to switch from a depleted patch. The optimal policy is given by the Marginal Value Theorem, which has successfully predicted animal behaviors, but as a mechanism it becomes problematic when each patch contains few discrete prey items. Biologists have proposed simple alternative decision mechanisms and calculated in which environments each works well. We tested whether the decision mechanisms evolved to direct animals when to leave a food patch also underlie human decision making in the same context, and whether humans in an internal-search task (e.g. information in memory) use the same mechanisms as in an external-search task (e.g. physical objects).