To compute lightness, illumination is not estimated, it is held constant.

Area of Science:

• Visual perception
• Color science
• Psychophysics

Background:

• Surface lightness perception is complex, as light intensity reaching the eye (luminance) is confounded by illumination levels and surface reflectance.
• Previous theories, including Helmholtz's, proposed that the visual system unconsciously estimates light source direction and intensity to determine surface lightness.
• An alternative hypothesis suggests the visual system disentangles illumination and reflectance by comparing equally illuminated surfaces.

Purpose of the Study:

• To experimentally test whether the visual system estimates light source properties or compares equally illuminated surfaces to perceive lightness.
• To investigate the role of illumination cues (shadows, highlights) in lightness constancy.

Main Methods:

• A 3D scene was constructed where rotating a target surface would lead to predictable changes in perceived lightness based on two competing theories.
• Experiments manipulated illumination levels and surface reflectance, comparing outcomes against predictions from lighting estimation and equi-illumination comparison theories.
• A subsequent experiment removed visual cues for light source direction, such as cast and attached shadows, and glossy highlights.

Main Results:

• Experimental results strongly favored the equi-illumination comparison theory over the lighting estimation theory.
• The perceived lightness of the target surface changed as predicted by the equi-illumination comparison theory, not the lighting estimation theory.
• Eliminating cues for light source direction did not affect the observed results, further undermining the lighting estimation theory.

Conclusions:

• The visual system appears to achieve lightness constancy by comparing surfaces under equivalent illumination conditions, rather than by estimating light source characteristics.
• The findings challenge established theories and suggest a novel mechanism for understanding how we perceive surface shade under varying light.
• Lightness perception is robust to the absence of explicit illumination cues, supporting a comparative processing approach.