Search

Using a color-naming procedure, two subjects described monochromatic lights, ranging from 450 to 630 nm, that were surrounded by perceptually unique-white fields of variable retinal illuminance. The test fields were 0.6 deg and 10, 100, or 1,000 Td. The surrounds, which were 4.5 deg in outer diameter, ranged from 0 to approximately 31,000 Td. From the resulting color-naming functions, equal-hue contours were derived, with surround intensity plotted against wavelength, for the spectral unique hues and the binary hues blue-green, green-yellow, and yellow-red. The wavelengths for unique blue, unique yellow, and blue-green were essentially invariant with changes in surround intensity. The spectral locus for unique green was also invariant at higher test-field intensities, but, at lower levels, it generally shifted toward shorter wavelengths as surround intensity increased. Nonmonotonic shifts were found for green-yellow and yellow-red. The contrast and the wavelength requirements for the color brown were nearly invariant with the changes of test-field intensity. Over the full range of surround intensities, subjects described the test fields as consisting of one or two hues plus white or black, depending on the surround level, but never (except for one subject at the lowest test-field intensity) white and black simultaneously and cospatially. This opponent aspect of black and white was compared with that associated with the chromatically opponent processes.

Three subjects viewed, foveally and monocularly, a monochromatic test field of 0.6-deg diameter that was surrounded by a white annulus of 0.6-deg inner diameter and 4.5-deg outer diameter. The wavelength of the central test field was varied in steps of 10 nm from 440 to 640 nm, and its luminance was set to 100 Td. Center and surround were flashed together for 2 sec every 4 sec. The subjects adjusted the luminance of the surround so that the central field was perceived as having equal amounts of whiteness and blackness. The luminance of the surround required for this balance point varied with the wavelength of the test field in a manner that closely resembled a heterochromatic brightness matching function obtained under similar conditions. Control experiments ruled out the possibility that the subjects were making brightness matches between center and surround fields. Additional evidence was provided suggesting that the spectral responsivity of the putative white-black channel is best represented by a photopic spectral sensitivity curve based on equal brightness.

Six subjects induced blackness within a circular broadband field by increasing the radiance of a surrounding monochromomatic annulus, which varied in wavelength. Between the central field and the annulus was a thin dark ring. Half of the subjects were instructed to increase the radiance of the annulus until the central field just turned black, and the other half were instructed to increase the radiance of the annulus until the contour between the central field and the dark ring disappeared. Spectral luminous efficiency functions measured by the methods of heterochromatic flicker photometry (HFP) and brightness matching (HBM) were determined for each subject and compared with the subject’s blackness-induction functions. The hypothesis that the contour-disappearance instruction would yield blackness-induction curves best fitted by flicker photometric functions and that the absolute-blackness instruction would yield blackness-induction curves best fitted by HBM functions was not confirmed. There was only one subject for whom the spectral efficiency of blackness was represented better by HFP than by HBM. There was one subject for whom blackness spectral efficiency was fitted better by HBM than by HFP. For the remaining four subjects, there was no difference in fits.

Photopic spectral sensitivity was determined in the mid- and far-peripheral retina by two methods. The first consisted of measuring increment thresholds on a background similar in spectral composition to CIE Source A. The resulting spectral-sensitivity functions displayed maxima at about 440 nm, in agreement with previous work. The second method consisted of measuring dark-adaptation curves following termination of the background. From these curves, spectral-sensitivity functions were derived for various times in the dark. The results showed that the 440 nm maximum quickly diminished. When photopic thresholds were estimated from the cone plateau of the dark-adaptation curves, the spectral-sensitivity functions peaked at about 550 nm and had much the same shape from the parafovea to the far periphery. We suggest that previous findings of maximum photopic sensitivity in the short-wave region of the spectrum resulted from chromatic adaptation induced by backgrounds (such as Source A) that were weighted towards middle and long waves.