Abstract
The purpose of this study is to investigate children's cognitive maps of school buildings. Sketch-map, placings of specific locations of floor plan, and projective convergence technique were employed to externalize children's cognitive maps. The subjects were second, third, and fifth graders. Some of the second graders' maps seem to be as accurate as the fifth graders' ones in the placings of specific locations. There were some charasteristic errors in cognitive maps represented by sketch-map and placings of specific locations. In one school building, the block slanting at 45 degree angle was distorted to be at right angle and pararell in about 30 children's cognitive maps. In the other school building, about 10 children's sketct-maps indicated rotated distortions. However, the children who showed some errors in sketch-maps and the placings of specific locations relatively, produced accurate cognitive maps from projective convergence technique.
