Abstract
When a row of five adjacent rectangles is suspended in midair and illuminated with a spotlight, a dramatic compression in the range of perceived grays results. This compression can be reduced by increasing articulation, or the number of surfaces in the spotlight (Gilchrist et al, 1999), which according to Katz (1935) improves lightness constancy. Based on our prior findings that spatially separate fields of illumination can group together, we asked whether adding articulation to one field can reduce compression in another. A 25-patch Mondrian was added to the scene. The Mondrian and the five rectangles were equally illuminated and coplanar, but separated laterally by 45 cm. Two groups (Mondrian present vs. absent) made Munsell matches for the five rectangles. Presence of the Mondrian produced significantly less compression, but there was a loss of efficiency. The reduction of compression was only one third what would have been obtained had the two displays been merged into a single display. Four additional experiments showed that efficiency of grouping (measured as a decrease in compression) increases with: a) decrease of the illumination difference between the five rectangle display and the Mondrian; b) increase of the difference between display illumination and room illumination; c) decrease in lateral separation (20, 45, or 65 cm); d) decrease in depth separation (from 0 to 74 cm). We conclude that the visual system functionally combines articulation from separate fields of illumination, but with a loss of efficiency as the fields are separated either geometrically or photometrically.