PubMed · 8538736
The source did not provide an abstract. Follow the original record for more information.
Explore connections, maps & timelines
P Cavanagh. 1996-01-04. Vision. When colours move.. https://doi.org/10.1038/379026a0
Cite the original work for its findings. Save a collection to share your selection of sources.
KEEP EXPLORING
Explore the source record for details and available documents.
The perception of yellow has played a central role in distinguishing two main theories of colour vision. Hering proposed that yellow results from the activation of a distinct retinal-neural mechanism, whereas according to the Young-Helmholtz-Maxwell view, yellow results from the combined activation of red and green cone mechanisms. When red and green images are presented separately to corresponding retinal locations in the two eyes, the resulting sensation is yellow. As the pathways from the two eyes do not converge until the cortex, this suggests that yellow can indeed arise from the central combining of separate red and green channels. I now show that the reverse process can also occur; the visual system can decompose a 'yellow' stimulus into its constituent red and green components. A 'yellow' stimulus was created by optically superimposing a flashed red line onto a moving green bar. If the bar is visible only briefly, the flashed line appears yellow. If the trajectory of the green bar is exposed for sufficient time, however, the line is incorrectly perceived to trail the bar, and appears red. Motion processing occurs in the cortex rather than the retina in primates, and so the ability of motion cues to affect the perception of colour is consistent with the Young-Helmholtz-Maxwell notion of a 'central synthesis' of yellow.
A series of experiments were designed to examine how luminance and color cues influence the occurrence of neon color spreading for the Ehrenstein pattern plus the cross pattern. The proportion of "see" responses for color spreading was obtained for different combinations of the luminance and color of the pattern components. The following were obtained. (1) Increase in the luminance of the inducing pattern and/or the color difference between the cross and the inducing pattern raised the proportion of "see" responses for color spreading. This implies that luminance and color signals additively contribute to the generation of the color spreading, but in different ways. (2) The "iso-spreading contours" for the generation of color spreading were determined on the two-dimensional isoluminant plane composed of the L-M and S-(L+M) axes. The contours were approximately described by rotated quadratic or ellipse functions. The additive interaction within the chromatic systems [the L-M system and the S-(L+M) system] was less significant than that across the luminance and chromatic systems. (3) The pattern of the experimental results could not be explained straightforwardly by existing models.