Abstract
Children distort angular figures so that the constituent angles are nearer 90° than they should be. This could be due to a perpendicular bias or a bisection bias, or to both. A study is reported which was designed to establish whether a perpendicular bias would appear independently of bisection. Twenty four-year-old children were tested on two types of angular figure: (i) a baseline with another line joining at the end at 45°, 90°, or 135° and (ii) a baseline with another line joining at the middle at 45° or 90°. Perpendicular errors were obtained both for ‘end’ and for ‘middle’ figures, but overall more strongly for ‘middle’ figures. However, while 90° ‘middle’ figures were copied more accurately than 45°/135° figures, this effect was only obtained for vertical and horizontal presentations of ‘end’ figures, and was reversed for oblique presentations. Also, for ‘end’ figures, directional errors varied with subtended-line orientation, whereas for ‘middle’ figures they varied with baseline orientation. It is concluded that although errors towards the perpendicular do occur with single-angle figures, angle equalisation may take place when there are two adjacent angles in the figure. One interpretation of the differing orientation effects is that in ‘middle’ figures strong internal relational forces produce a distortion that varies with the angle at which the figure is viewed, whereas in ‘end’ figures the absence of relational forces within the figure leads to a stronger influence from external cues.
Get full access to this article
View all access options for this article.