On Edwin Land

by Gerald Huth on April 13, 2005

Someone has written re: the “two color system of vision that Land invented”. In my review of the Land color vision controversy, this was the “terrible heresy” of which he was accused….”how dare he refute the trichromicity of the vision findings of Young et al…etc.”. Land said repeatedly that he had proposed no such thing. It now becomes clear in this model of light interaction on the retina how this misreading of Land’s work arose. Land deduced, and proposed, that color was determined by “ratioing /two /lightnesses” on either side of a “mid-point”… the location of which within the eye he could not determine… but which now becomes clear (the only model of the eye that he had at hand was the “standard” one envisioning the retina as the analogue of photographic film). What Land could not know was that long wavelengths refracted by the eye to the central fovea additionally “decode” the outline sketch (the “primal” sketch that David Marr deduced must be present) of the perceived image. It is this sketch colored by Land’s /two lightnesses/ (or “brightnesses”), and via an optical transform mechanism (the Fourier equation), results in the final image transmitted to the brain.

It occurs to me that one should study the color vision work of Edwin Land. To this end one might obtain (from the Rowland Institute founded by Land and now subsumed into Harvard University) a compendium of Land’s papers describing his experiments and ideas about color vision. Such a collection was kindly sent to me by an associate of Dr. Land (Holly Perry) some years back that has proven invaluable in this work (I would wish that Rowland would make such a collection available if they do not now do so). Land’s writings are characterized by an uncommon directness and lucidity that make them very easy to understand (the exception is his Retinex hypothesis which was a rather complicated algorithmic attempt to simulate the processes that he discerned must be occurring within the eye). I believe that my model defines what Land did not have available to him, i.e., the mechanisms within the eye that geometrically determine the exact, mid-visual band point of light interaction on the retina (that Land deduced must be there), and, additionally, defines the method of “lightness” detection (via /density/ of “light antennas”) on either side of this point.


Leave a Comment

Previous post:

Next post: