Search

The visual system often relies on prior assumptions when interpreting ambiguous visual inputs. A well-known example is the light-from-above prior, which aids the judgment of an object's three-dimensional (3-D) shape (i.e., convex or concave). Recent studies have revealed that the light-from-above prior also helps solve lightness ambiguity. This study aimed to examine whether 3-D shape perception and lightness perception share the same lighting prior. The study participants performed two tasks: one focusing on lightness perception and another focusing on 3-D shape perception. The dominant directions of the assumed lighting were calculated from participants’ performance in the two tasks. The results showed that the assumed lighting direction for 3-D shape perception were considerably biased toward the left, whereas the one for lightness perception was almost from directly above. The clear difference between these two directions supports the hypothesis that the visual system uses distinct lighting priors for 3-D shape perception and lightness perception. Experiments 1 and 2 involved Japanese speaking participants and European participants, respectively. The Japanese language can be read and written both horizontally (i.e., left to right) and vertically (i.e., up to down) with lines progressing from right to left. Nevertheless, the two experiments still produced the same result, which suggests that the present finding is universal regardless of reading/writing direction.

The amount of light reaching the eyes can be affected by changes in the level of illumination or by changes in the reflectance of external surfaces. It is well known that, at the retinal level, such a distinction loses its meaning, since there are only changes in luminous flux and spectral composition at this level. However, information about the two types of changes is recovered by the visual system at the perceptual level, allowing us e.g. to easily distinguish between self-luminous and reflecting surfaces. Surprisingly, the neural mechanisms underlying this distinction are largely unknown, even though the perceptual correlates are compelling. Using fMRI, we investigated in twelve healthy volunteers the neural basis for the perception of elements of the visual field which appear to be self-luminous (sources of light), as distinct from those elements which appear to reflect light. Stimuli were cross-like figures with luminance-gradients, as introduced first by Zavagno (Perception,1999, 28: 835-8) and Zavagno and Caputo (Perception, 2001, 30: 209-22). These stimuli can be modulated so that they produce a percept of luminosity or reflectance without changes in the luminance of the area which is seen as self-luminous or reflecting. Additional stimulus types were used to control for effects induced by the use of luminance gradients, and to distinguish possible luminosity-sensitive regions from brain regions that are sensitive to general luminance differences. Data were analysed with SPM'99. A group analysis revealed bilateral activation in the posterior part of the occipito-temporal sulcus that was specific for glowing stimuli. Individual analysis showed bilateral activation in 8 and unilateral activation in 4 of the 12 subjects. We conclude that a region in the human posterior occipito-temporal sulcus seems to be specifically activated by glowing objects. This region does not increase its activity with increasing luminance.