The perception of the degree of transparency of a surface depends on the chromatic differences between stimulus areas. This experiment was designed to test the hypothesis that the degree of transparency of a transparent surface M seen on another surface N and on a background L depends on the comparison between lightness contrasts. M, N, and L were concentric and their contours did not intersect. The luminances of M, N, and L were m, n, and l, respectively. There were three stimuli for each luminance order: (i) m < n < l, (ii) n < m < l, and (iii) n < l < m. Eighteen observers participated in the experiment. They rated the lightness contrasts between M and N (CMN), L and M (CLM), and L and N (CLN), and the degrees of transparency of M, T, using integers from 0 (no contrast or opacity) to 100 (black vs white or perfect transparency). The results confirm that the ratio between CMN and the sum of CMN and CLM, which expresses an integration of these perceived contrasts, gives valid predictions of T for stimuli with orders (i) and (ii). For stimuli with order (iii), the rated degrees of transparency differ from the predicted ones, and this is probably due to the observers' tendency to base their ratings of T more on CMN rather than on CLM.
A ratio model of perceived transparency for stimuli with no contour intersections
TOMMASI, Marco
1998-01-01
Abstract
The perception of the degree of transparency of a surface depends on the chromatic differences between stimulus areas. This experiment was designed to test the hypothesis that the degree of transparency of a transparent surface M seen on another surface N and on a background L depends on the comparison between lightness contrasts. M, N, and L were concentric and their contours did not intersect. The luminances of M, N, and L were m, n, and l, respectively. There were three stimuli for each luminance order: (i) m < n < l, (ii) n < m < l, and (iii) n < l < m. Eighteen observers participated in the experiment. They rated the lightness contrasts between M and N (CMN), L and M (CLM), and L and N (CLN), and the degrees of transparency of M, T, using integers from 0 (no contrast or opacity) to 100 (black vs white or perfect transparency). The results confirm that the ratio between CMN and the sum of CMN and CLM, which expresses an integration of these perceived contrasts, gives valid predictions of T for stimuli with orders (i) and (ii). For stimuli with order (iii), the rated degrees of transparency differ from the predicted ones, and this is probably due to the observers' tendency to base their ratings of T more on CMN rather than on CLM.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
