What do misestimations and asymmetries in spatial judgment indicate about spatial representation?

When people are asked to judge the distance between 2 points, they may prod uce systematic over- or underestimations. Their judgments may also show asy mmetries, as when, for example, people estimate the distance from their hou se to a mailbox as different from the distance from the mailbox to their ho use. It has been argued that such errors show that spatial representations are fundamentally nonmetric. In 3 experiments, however, the authors show th at these effects can be explained by a category-adjustment (CA) model of sp atial coding, in which coding is hierarchical (i.e., occurs at more than on e level of measurement, with estimates based on combination across levels). In this model, coding at each level is uncertain but not distorted. Experi ment 1 shows that, in a carefully controlled experimental setting, the CA m odel can be used to predict under- or overestimations of distances with res pect to objective standards. Experiments 2 and 3 show that, when people lea rn maps, the CA model correctly predicts patterns of asymmetries in estimat ion.