Abstract
The current project maps the internal structure of a rectangular frame through goodness ratings of one or more probe targets placed at different positions and orientations in its interior (Palmer, 1991). In Experiment 1, participants rated the goodness-of-fit for a single, small, circular dot placed at each of 7 x 11 positions within a rectangle. The highest ratings occurred when it was at the center, decreasing as a function of distance from it. In addition to this “center bias” toward global balance, elevated goodness ratings were found for positions along the vertical and horizontal symmetry axes of the frame and also along the local symmetry axes that bisect the corners. Experiment 2 showed that the corner effects were, in fact, due to local symmetries along the corner bisectors rather than to global diagonals. Experiments 3 and 4 used a single more structured probe target - a small isosceles triangle - to map the orientational structure of the rectangle in more detail. The results confirmed both the importance of the center and the symmetry structure of the whole configuration, but also showed a directional “facing bias” to have the triangle point into (rather than out of) the rectangle. Further experiments investigated the influence of a second dot in the same rectangle. The pattern of ratings as a function of position changed globally and dramatically, but symmetries and relational structure were still the most important factors, accounting for almost 70% of the variance. These results may be relevant to the empirical study of aesthetic preference, such as the “center bias” and the “facing bias” reported by Palmer and Gardner (VSS-06, VSS-07).