Method: Each of four to six observers with normal color vision was asked to adjust a black pole back and forth untilit appeared to be at the same distance with a colored surface whose vertical edge could be seen through an aperture of a black screen The difference between the matched distance of the pole for the colored surface and that for the neutral surface was adopted as the measure of advancement or recession.
Results: 1) Red was the most advancing color: Orange, Yellow and Red Purple also showed advancement. Green and Purple did neither advance nor recede. Blue, Greenish Blue and Blue Purple were receding colors.
2) The lightness of Gray, Blue or Red surface had little influence on this phenomenon.
3) Both of the advancement of Red and the recession of Blue were proportional to their saturation.
4) There was no difference between the results when colored surfaces were presented on the white background and those when they were placed in front of the black background.
5) In scotopic vision, hue of the surface had no influence on this phenomenon. This means that the color sensation rather than the wavelength of the stimulus light is the determining factor of the phenomenon, and consequently, any explanation on the basis of the chromatic aberration of human eyes is not appropriate.

View full abstract