Abstract
How the luminance of a visual stimulus is translated into a perceptual value of brightness has been debated since the middle of the 19th C. Although it seems intuitively clear that the perception of brightness should scale directly with the intensity of the light that activates retinal receptors, this is not the case. Thus, doubling the luminance of a stimulus under the laboratory conditions in which such studies are typically done does not double its perceived brightness. The exponential relationship between luminance and brightness in these circumstances is referred to as the Weber-Fechner Law or, alternatively, as Stevens' Power Law. Although measurements of the relationship between luminance and brightness have become progressively more sophisticated, the reason for this relationship and its modification under more natural viewing conditions has never been explained. Operating with the assumption that the basis of these relationships might be based entirely on past experience, we analyzed the luminance of a visual stimulus as a function of its generative sources, i.e. the reflectance of and illumination giving rise to the visual stimulus in both restricted and more natural scenes. The results of the study show that the form of the luminance/brightness relationship changes predictably according to the relative contributions of reflectance and illumination that would previously have been experienced in the presence of the same or a similar stimulus. We conclude, therefore, that the relationship between luminance and brightness is determined empirically according the success or failure of visually-guided behavior.