Luminance ratios along shadow edges remain the same even when they cross reflectance borders. According to Gilchrist (1988, Perception and Psychophysics 43 415 – 424) this so-called ratio-invariance property is a crucial factor in the perception of shadows. However, Soranzo and Agostini (2004, Perception 33 1359 – 1368) suggested that in some conditions (named ‘impossible shadows’), a luminance pattern might still be perceived as a shadow even if the ratio-invariance property along its edge is violated. This can occur when an edge is collinear with another edge (contextual edge) which incorporates it, shares the same polarity, and generates a larger ratio. The hypothesis that impossible shadows are actually perceived as shadows is here tested by comparing the perceptual contrast of a luminance edge in the absence of a contextual edge (control condition) to that of both possible shadow edges (where the contextual and mediating edge share the same ratio) and impossible shadow edges (where the ratio of the contextual edge is larger rather than that at the mediating edge). We found that the perceived contrast of luminance edges shrinks in both possible and impossible shadow conditions rather than in the control condition. This evidence supports the hypothesis that a luminance pattern might be perceived as a shadow even when the ratio-invariance property is violated.
The perceptual contrast of impossible shadow edges.
GALMONTE, Alessandra;
2009-01-01
Abstract
Luminance ratios along shadow edges remain the same even when they cross reflectance borders. According to Gilchrist (1988, Perception and Psychophysics 43 415 – 424) this so-called ratio-invariance property is a crucial factor in the perception of shadows. However, Soranzo and Agostini (2004, Perception 33 1359 – 1368) suggested that in some conditions (named ‘impossible shadows’), a luminance pattern might still be perceived as a shadow even if the ratio-invariance property along its edge is violated. This can occur when an edge is collinear with another edge (contextual edge) which incorporates it, shares the same polarity, and generates a larger ratio. The hypothesis that impossible shadows are actually perceived as shadows is here tested by comparing the perceptual contrast of a luminance edge in the absence of a contextual edge (control condition) to that of both possible shadow edges (where the contextual and mediating edge share the same ratio) and impossible shadow edges (where the ratio of the contextual edge is larger rather than that at the mediating edge). We found that the perceived contrast of luminance edges shrinks in both possible and impossible shadow conditions rather than in the control condition. This evidence supports the hypothesis that a luminance pattern might be perceived as a shadow even when the ratio-invariance property is violated.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.