Abstrset-The determination of the relative numbers of different cone types in the human retina is fundamental to our understanding of visual sensitivity and color vision; yet direct measurements which provide this basic information have not previously been made for ail cone types. Here we present a model which links the detection of a test light of small dimension to the number of cones contributing to detecfon of the light. We selectively isolated either the long-wavelength-sensitive (L) or the middle-wavelengthsensitive (M) cones, by choosing combinations of wavelengths of adapting backgrounds and tests to favor detection by the cone class of interest. Our model was applied to the detection functions measured for six color normal observers to obtain estimates of the relative numbers of L to M cones. Our estimates ranged between 1.46 and 2.36 for our observers with a mean value near two L cones for every M cone in human fovea centralis. ConesHuman fovea centralis Relative numbers of L to M cones ~RODU~ONThe determination of the relative numbers of different cone types in the retina is fundamental to our understanding of human visual sensitivity and color vision, and this information would be required for any quantitative models of human vision. Direct measurements which provide this basic information have not been previously made for all cone types. There continues to be a gratifying convergence of psychophysically derived evidence from humans ~Williams et al., 1981) and anatomically derived evidence from baboon (Marc and Sperling, 1977), macaque (deMonasterio et al., 1985), and human (Ahnelt et al., 1987) on the numerosity and distribution of the shortwavelength-sensitive (S) cones in the primate retina.In the cases of the long-wavelength-sensitive (L) and middle-wavelength-sensitive (M) cones, there are no previous direct psychophysical measurements from which the relative numbers of L and M cones can be derived, and estimates based on various indirect means vary widely. To our knowledge, DeVries (1946DeVries ( , 1948 was the first to suggest that the individual variability in luminosity functions could be related to individual variability in the relative numbers of different cone types. Rushton and Baker (1964) subsequently reported that retinal densitometric measurements yielding the density of M and L cone pigments could be correlated to the flicker photometric matches between red and green lights made by their observers. Rushton and Baker's estimates, based on densitometric measurements, of the relative numbers of L to M cones in normal trichromatic observers spanned a wide range of three times more L as compared to M cones to one third as many L as compared to h4 cones. Another approach has been based on estimates deriving from curve fits required to make various sets of psychophysical data consistent one to another. Examples of this kind of analysis include Walraven's (1974) and Smith and Pokorny's (1975) estimates based on fits of the cone primaries to the luminosity function; Vos and Walraven's (1971)...

The topographical retino-tectal projection of goldfish was electrophysiologically mapped a t various intervals after surgical removal of the nasal half of the retina and pigment epithelium. The remaining projection was initially restricted to the appropriate rostral half of the tectum, even if the nerve was crushed and allowed to regenerate. But later, after 137 days or more, it showed a progressive expansion onto the foreign caudal half of the tectum. The magnification factor, the number of micrometers of tectum per degree in the visual field, doubled in the rostro-caudal but not in the medio-lateral direction. Analysis of the sequence of the expansion showed that a few fibers originally projecting nearest the denervated area were the first to spread over it. Then, progressively more fibers moved caudally until a nearly uniform representation of the half retina was established on the tectum. Radioautography also demonstrated that retinal fiber terminals had invaded the caudal tectum. The retinae of these fish were also examined histologically. The density of ganglion cells had not increased, but they consistently showed the axonal reaction. This was not found to be associated with any initial surgical trauma, but rather with the movement of their fiber terminals within the tectum. Frozen sections, through half retinal and normal eyes, were cut and photographed for comparison of ocular geometry. Operated eyes were normal except for a slight but consistent loss of ocular volume. Analysis of the optical geometry showed that recording with fish in air produced two effects: Myopia (10" blur circle, or less) and enlargement of the visual field by 15% to 20%.The topographical retino-tectal projection of goldfish, which was once thought to be unmodifiable (Attardi and Sperry, '63; Jacobson and Gaze, '65; Sperry, '631, has more recently been found to be a very dynamic system (Gaze and Sharma, '70; Yoon, '71, '72a,b,c; Schmidt et al., '74). This shift in thinking began with the work of Gaze and Sharma ('70) who first demonstrated that, following removal of the caudal half tectum, the entire retinal projection could compress onto the remaining rostral half tectum.Although the results on half tectal compression have been verified by many others (Yoon, '72a, '71; Schmidt et al., '74; Meyer, '77), there is still disagreement about reorganizations following retinal ablations. Anatomical studies have shown little or no evidence for expansion of the half retinal projection. Attardi and J. COMP. NEUR., 177: Z~I -Z I~ Sperry ('63) waited only 30 days after optic nerve crush and partial retinal removal, and found that the regenerating fibers reestablished the original projection of the remaining retina but did not invade new areas of the tectum. More recently, Meyer ('75) using radioautography, found little evidence for expansion of the half retinal projection even after many months. Two short electrophysiological reports have also dealt with the projection of the half retina a t long postoperative intervals, but include ...

No abstract

Weintroduce and explore a color phenomenon which requires the prior perception of motion to produce a spread of color over a region defined by motion. Wecall this motion-induced spread of color dynamic color spreading. The perception of dynamic color spreading is yoked to the perception of apparent motion: As the ratings of perceived motion increase, the ratings of color spreading increase. The effect is most pronounced if the region defined by motion is near 1 0 of visual angle. As the luminance contrast between the region defined by motion and the surround changes, perceived saturation of color spreading changes while perceived hue remains roughly constant. Dynamic color spreading is sometimes, but not always, bounded by a subjective contour. We discuss these findings in terms of interactions between color and motion pathways.Neon color spreading (see, e.g., van Tuijl, 1975;Varin, 1971) shows that the colors we perceive do not always match predictions based on the spectral content of the stimulus. In instances of neon color spreading, color is seen over regions which in isolation would appear achromatic. Specific geometric, color, and brightness features are required in order to induce the neon color spreading into nearby areas. Consider the example in Figure 1: A full green disk can be seen, although only eight radial lines, part green and part red, are drawn. Another feature of the colored disk is that it appears luminous-hence the label "neon" color spreading.In this paper we introduce and explore a phenomenon of color spreading which differs from standard neon color spreading in that it requires the prior perception ofmotion to produce a spread of color over a region defined by motion. We call this motion-induced color spreading dynamic color spreading for short. Two frames from a typical display of dynamic color spreading are shown in Figure 2. Each frame consists of a white square containing 900 dots placed randomly (sampled from a uniform distribution) within the square area. The dots do not move from one frame to the next; only their colors are updated as follows: All dots are colored red except for those within a (virtual) disk, which are colored green. The center ofthe disk translates from one frame to the next, with the consequence that some red dots in one frame are green in the next, and vice versa. Again, the physical placement of dots remains unchanged from frame to frame; only the colors of a small number of dots (those at the leading and trailing edges of the virtual disk) change from one frame to the next. TheWe thank M. Albert, B. Bennett, M. Braunstein, 1. Yellott, and two anonymous reviewers for helpful comments. This research was supported by ONR Contract NOOOI4-88-K-0354 (D.D,H.), NSF Grant BNS8819874 (CM.C.), andNEI Grant IROIEY08200 (CM.C.). Correspondence should be addressed to C M. Cicerone, Department of Cognitive Sciences, University of California, Irvine, Irvine, CA 92717 (e-mail: cciceron@uci.edu), Mathematica (Version 2.03) program used for generating such frames is given in App...

Abstract. Exposure to constant light causes extensive rod photoreceptor damage but spares the photopic system in albino rats. The rod branch of tire dark-adaptation curve shows considerable elevation in threshold; the cone branch is hardly affected. Longer exposure and chromatic adaptation suggest that there are three cone mechanisms with peaks near wavelengths of 450, 520, and 560 nanometers.