The pattern of the light that falls on the retina is a conflation of real-world sources such as illumination and reflectance. Human observers often contend with the inherent ambiguity of the underlying sources by making assumptions about what real-world sources are most likely. Here we examine whether the visual system’s assumptions about illumination match the statistical regularities of the real world. We used a custom-built multidirectional photometer to capture lighting relevant to the shading of Lambertian surfaces in hundreds of real-world scenes. We quantify the diffuseness of these lighting measurements, and compare them to previous biases in human visual perception. We find that (1) natural lighting diffuseness falls over the same range as previous psychophysical estimates of the visual system’s assumptions on diffuseness, and (2) natural lighting almost always provides lighting direction cues that are strong enough to override the human visual system’s well known assumption that light tends to come from above. A consequence of these findings is that what seem to be errors in visual perception are often actually byproducts of the visual system knowing about and using reliable properties of real-world lighting when contending with ambiguous retinal images.
Lightness constancy is the remarkable ability of human observers to perceive surface reflectance accurately despite
variations in illumination and context. Two successful approaches to understanding lightness perception that have
developed along independent paths are anchoring theory and Bayesian theories. Anchoring theory is a set of rules that
predict lightness percepts under a wide range of conditions. Some of these rules are counterintuitive and difficult to
motivate, e.g., a rule that large surfaces tend to look lighter than small surfaces. Bayesian theories are formulated as
probabilistic assumptions about lights and objects, and they model percepts as rational inferences from sensory data.
Here I reconcile these two seemingly divergent approaches by showing that many rules of anchoring theory follow from
simple probabilistic assumptions about lighting and reflectance. I describe a simple Bayesian model that makes
maximum a posteriori interpretations of luminance images, and I show that this model predicts many of the phenomena
described by anchoring theory, including anchoring to white, scale normalization, and rules governing glow. Thus
anchoring theory can be formulated naturally in a Bayesian framework, and this approach shows that many seemingly
idiosyncratic properties of human lightness perception are actually rational consequences of simple assumptions about
lighting and reflectance.