Temporal analog of the minimally distinct border

In this note, we present old data that are now seen in a new light. The experimental work, carried out a decade ago (Boynton, Kaiser and Shlaer, 1967), can now be viewed as a temporal analog of the method of minimally distinct border (MDB) which we developed somewhat later (Boynton and Kaiser, 1968; Kaiser, Boynton and Herzberg, 1971). More recent experiments with the MDB have shown that short wavelength sensitive (S) cones do not contribute to the formation of such borders (Tansley and Boynton, 1976; Tansley, 1976). For some conditions of our flicker experiment it can now be interpreted that the S cones do not contribute to the perception of flicker either. The principles of the method are illustrated in Fig. 1, where a full explanation is given in the caption. Whereas the homochromatic flicker curve must reach zero when the two components of the field are of equal luminance, the heterochromatic curve does not. Instead, the minimum of that curve defines a condition of equal luminance. The ordinate value at this minimum corresponds to a critical flicker frequency (cflJ that is obtainable for some specific mismatch r of luminances in the homochromatic case. Values of r determined in this way were used to define the equivalent achromatic contrast values shown in Fig. 3. We constructed an apparatus in which a 0.5” field, seen at 100 td, could be supplied by either of two optical channels. When the wavelength of the light variation of a bit less than 2% at peak, lasting for about 0.75 ms. The control condition (solid points) was suggested to us by R. M. Shlaer: it generates the same physical stimulus, in the ideal case, without