Spatial representation of maps.

Two experiments tested the hypothesis that the cognitive representation of spatial relations on a map is primarily a visually keyed process. In Experiment 1, undergraduates studied a labeled map where features were presented either as words alone, as mimetic drawings of the referent, or as geometric symbols. Cued recall data showed that discrete features are best located when they are semantically congruent with the label referent, a fact providing considerable support for the image position. Recognition testing required a set of introspective judgements ordinally related to the interfeature metric on the map. The judgment X distance estimates were highly correlated for all groups, intimating that a high correspondence exists between image distances and the original stimulus metric. Experiment 2 had undergraduates view either a full-scale or a three-quarter reduction of a reference map. Subjects in the reduced version were able to recognize significantly more interfeature comparisons, suggesting a predictable relation between visual and imaginal acuity. The judgement X distance results were essentially the same as in Experiment 1, and the two groups failed to differ in terms of cued recall. Data from both experiments imply that spatial relations are cognitively represented in at least a partially isomorphic fashion, which seems to possess quasi-pictorial qualities.