Rethinking the Process of Vision

Gerald C. Huth, Ph.D. (physics) — Ojai, CA | Tucson, AZ
E-mail: gerald.huth @ gmail.com

PLEASE SEE “RUNNING COMMENTARY” ON THE LEFT OF THIS PAGE FOR CONTINUALLY UPDATED THOUGHTS

“Anyone who conducts an argument by appealing to authority is not using his intelligence; he is just using his memory.”
Leonardo da Vinci

“…we have learned that the eye must have a fantastic mechanism for finding a balance point within a band of wavelengths”
Edwin Land

“One theory maintains that the nerve ends on the human retina (rods and cones) are tuned to receive any of the 3 primary colors (red, yellow or blue) which constitute all colors”
Josef Albers, 1963, “Interaction of Color”

“We conclude…that there is no definitive evidence for three molecular substances mediating colour vision” “…one possibility is that the (one) identical molecular substance is arranged differently in each receptor type”
Snyder and Hall (”Unification of Electromagnetic Effects in Human Retinal Receptors with Three Pigment Colour Vision”, Nature, Vol. 223, 1969

“If aberrations do not significantly affect visual perception, then why has nature allowed such large amounts of some aberrations while minimizing others (spherical aberration)? The speculative answer to this question proposed here is that aberrations of the eye (whatever the eye tolerates of these) rather than interfering with vision, may play a positive role in visual perception.”
Michel Millidot, “Effect of Aberrations of the Eye on Visual Perception”, Visual Psychophysics and Physiology, Academic Press, 1978

In this view of the interaction of light with the retina of the eye, what was  previously considered defect - longitudinal chromatic aberration - is shown to be not an aberration at all but, in fact, forms the basis for the fundamental imaging mechanism of vision. Light wavelengths refracted by the lens of the eye are detected by three narrowly tuned circular areas surrounding the  central fovea on the retina. These areas are defined by receptor appositions (i.e. the cone-to-cone, cone-to-rod, and rod-to-rod separations) and not by the cone and rod receptors themselves. This pattern is derived directly from the historically well characterized distribution of  retinal receptors  (Osterberg, G, Acta Ophthalmol. Suppl.6, 1-103,. 1935). The conclusion is that retinal response is the materialization of the physical laws of light refraction - nothing more!  Further, in contrast to past thinking, this geometric construction of retinal light interaction defines the exact short wavelength 400 nm limit (rod-rod) and and the long wavelength 700 nm limit of the visual band detected by the eye.  Moreover, an exact center of the visual band.is geometrically defined at a retinal eccentricity of 7-8 degrees with this point serving are a fixed wavelength reference point (~550 nm) from which all other wavelengths can be geometrically calculated in the vision process. This fixed reference also undoubtedly serves as the basis for the color constancy of vision. The three wavelengths detected by the three regions represent what have been historically termed  the “primary” wavelengths, i.e.,  “red”, “green” and “blue”. The trichromacy of vision is intact. These are not yet , however,  “colors”. The three apposition-defined retinal areas consist of a variation in density of receptor sites across each band.  This variation corresponds to regions of peak absorption or brightness (or, the term used by Edwin Land…”lightness”). As Land deduced from measurements made external to the eye the hues of color (using that term for the first time ) are derived from a ratio of lightnesses on either side of the (nanogeometrically) fixed 500 nm reference. The all-cone fovea of the retina is seen to be solely sensitive to the exact long wavelength end of the visible band (Wald described the fovea as being “blue blind”)) that, via a Fourier transform, encodes the overall brightness of the visual scene. Higher spatial frequencies detected at eccentricities beyond the fovea to ~20 degrees detect the “primal” or outline sketch of the perceived image as proposed by David Marr. The peripheral retina beyond ~ 20 degrees dominated by rod-rod appositions functions as an integrated, large area “light meter” that controls pupillary constriction and thus light entrance into the eye. These rod-rod appositions define the exact short wavelength limit as described above act to exert this pupillary control function. The proposal that rods act in this manner in the peripheral retina is at variance with the historical misunderstanding that “rods are the low light level receptors of the eye”.

It becomes clear in this geometrically-deduced plan of light interaction with the retina that the retina is a diffractometric ,or Fourier transforming , surface. The retina is located at the “back focal or Fourier plane of the optics of the eye. To satisfy the Fourier equation, each individual receptor apposition light detector center must be capable of detecting light phase as well as light intensity.

The retina should be seen as an array of generic, energy-accepting “nanowires” (again in physics terms, “one dimensional quantum-confined electron sites”). Optical wavelength absorption is determined by a classical “light antenna dimensionality” that exists between these sites. It is then the inner segments of cone and rod receptors that are actually the biologically engineered, dimension-controlling structures that define three different nanowire spacings and ultimately the trichromicity of vision.

The retina of the eye is not the analogue of a piece of photographic color film as vision texts have for so long portrayed or, at least, implied. Single cone receptors do not detect “color” and I can see no fundamental physics basis for the contention that single rods possess the “low light level” sensing function attributed to them. Rather it is shown that the light detecting elements of the retina are composed of any two adjacent receptors and the wavelength-defining space between them. This is the only model that is in accord with, and relates, the distribution of receptors on the retinal surface and the light refractive properties of the structure of the eye. The retina is shown to be composed of three discrete light sensitive regions defined by cone-cone, cone-rod, and rod-rod appositions each detecting a specific narrow wavelength band that combine to define the exact visual bandwidth. Three areas of the retina are seen to concentrically surround the fovea corresponding to the long and short wavelength limits, and, exactly as Edwin Land intuited must be there, the exact center of the band. These limits are narrowly tuned and precise - if the short wavelength limit of visual response is proposed as being at 400 nanometers it is exactly at this wavelength (within the limits of antenna bandwidth considerations or quantum uncertainty). Geometrical definition of the exact center of the band I believe explains the phenomenon of color constancy in vision providing a geometrically determined “reference wavelength” against which other wavelengths are compared. The retina is actually sensitive to, and defines three primary wavelengths from whence intermediate colors are synthesized. The visual image itself is formed from optical (or Fourier) transforms derived from these regions that combine to present this information and the sensation that we term color to the “color centers” of the brain. This model is consistent with Edwin Land’s color vision experiments and deductions.

It is basic to this explanation that the retina must be in the living state to effect the spatial order that must be inherent in this light interaction mechanism.