Peter Todd

The Michigan State University
Cognitive Science Program presents

Indiana University, Bloomington

Seeking and finding:
Strategies for searching in space and in mind

December,10, 2007 - 5:30 pm
Room 116, Natural Science Building

Humans and other animals often encounter resources in patches, such as berries on bushes or information on websites. Facing such clumpy resources, we must decide when a patch has been sufficiently depleted that it is worthwhile seeking a new one. The optimal strategy is to leave a patch when the instantaneous rate of resource return it provides falls below the long-term mean return rate of the whole environment. But individual animals often lack the knowledge about the environment that is necessary for this optimal approach. Instead, the actual mechanisms by which they make patch switching decisions are likely to be simpler heuristics, which can perform very well when used in environments with appropriate structure. Do humans use similar heuristics for searching among patches, and adjust them to the environments they face? We tested this by using one domain analogous to food search (a fishing task) and another domain that represented search for information from memory (an anagram task), and found that similar rules were used in both. We have also found evidence, in the form of priming between two such tasks, that a more general search process may underlie behavior in both of these and other domains of search. This raises questions about when we can expect humans and other organisms to display ecological rationality--the adaptive use of decision mechansims matched to the structure of information in their task environments.