UC Berkeley

Spatial memory is critical to the survival of all mobile animals. Using spatial memory animals find and keep valuable resources such as shelter, food and companionship. In a series of three studies, I explored spatial strategies and the use of multiple landmarks from comparative, computational and developmental perspectives.

In the first study, I examined the spatial strategies and landmark use of a diurnal mammalian species, the fox squirrel, Sciurus niger. Free-ranging fox squirrels showed a large degree of flexibility both in their use of landmarks and in their use of spatial strategies. Furthermore, their use of landmarks appeared to be sensitive both to changes in the test apparatus and the season in which they were tested. In the second study, I further explored fox squirrel spatial strategies from a computational modeling perspective. Given the squirrels' use of the majority strategy, we specifically investigated the suitability of using a Bayesian model to characterize the squirrels' spatial decisions. The resulting model provides a unifying framework for illustrating different strategies of landmark integration, and a tool for investigating the circumstances under which particular landmarks are used. I show that the best fitting models changed depending on the season at testing and the details of the task. These analyses support my previous claims that squirrels adopt flexible strategies in landmark use.

In the third study, I investigated whether human children develop similar flexible spatial strategies as the fox squirrels and what types of allocentric landmarks they might be using to construct such spatial strategies. In particular, we were interested in the initial development of such strategies when children first begin to use non-coincident, allocentric landmarks. Thus, all studies were conducted using 3 and 4 year old children since it is at this age that children begin to robustly use non-coincident, allocentric landmarks. I found that 3 year old children rely on a less flexible, hierarchical strategy based on their preferred landmark, but that by four years of age they begin to develop more flexible spatial strategies, similar to those of adult free-ranging fox squirrels.

In sum, both behavioral data and the results of Bayesian modeling support the conclusion that a free-ranging diurnal, mammal uses a flexible, spatial strategy to orient in its environment and that environmental factors may change the expression of such a strategy depending both on the season and properties of the experimental apparatus. At the age of four, human children have begun to develop similar flexible strategies, not long after they have begun to use non-coincident, allocentric landmarks. These results, taken together, suggest that both humans and animals have developed similar strategies to deal with spatial problems and that these strategies develop along with use of more distal landmarks in human children.