Two experiments investigated the role of spatial prototypes in estimates of
location. In Experiment 1 (N = 144), children and adults learned the locat
ions of 20 objects in an open, square box designed to look Like a model hou
se. In two conditions, opaque lines or walls divided the house into four re
gions, and in the other condition, no boundaries were present. Following le
arning, the dots marking the locations were removed, and participants attem
pted to replace the objects. Children and adults overestimated distances be
tween target locations in different regions. Contrary to Huttenlocher, Hedg
es, and Duncan's hierarchical theory of spatial memory, none of the groups
displaced the objects toward the region centers. In Experiment 2 (N = 96),
boundaries were removed during testing to determine whether children and ad
ults were more likely to displace objects toward region centers when uncert
ainty about location increased. Again, all age groups overestimated distanc
es between target objects in different regions. In addition, adults and 11-
year-olds in the most salient boundary condition displaced objects toward t
he region centers. Discussion focuses on the implications of these results
for understanding how children and adults estimate location.