An empirical explanation of the Chubb illusion

The brightness of any luminant stimulus varies, often quite markedly, as a function of the context in which it is presented. An especially intriguing example of this phenomenon is the illusion described by Chubb and colleagues (1989) in which the apparent contrast of a patterned target is reduced when it is embedded in a pattern of the same spatial frequency but of higher luminance contrast. Illusory percepts of brightness, like this one, are usually considered epiphenomena of inhibitory interactions between neurons tuned to the same attributes of the stimulus, in this case between neurons in the primary visual cortex similarly tuned to spatial contrast frequency. Here we tested a different possibility, namely that the Chubb illusion is generated according to the experience of the visual system with background textures seen through an imperfectly transmitting medium. In agreement with this suggestion, making the stimulus more consistent with a contribution to the target of imperfect transmittance increased the effect for naïve subjects, whereas making the stimulus less consistent with this possibility decreased the effect. Because the luminance contrasts and spatial frequencies of the stimuli were unchanged in these experiments, these results are difficult to explain in terms of the receptive field properties of neurons early in the visual processing stream. Rather, the results suggest that the Chubb illusion, like other illusions of brightness (and color), are generated empirically according to what the sources of the same or similar stimuli have typically turned out to be in the experience of both the species and the individual.

Duke Scholars

Author Dale Purves Duke Institute for Brain Sciences