Happily, what we see in various circumstances has been well documented over the years, providing plenty of grist for the mill of this or any proposed explanation of vision. The documentation falls into two general categories: all the visual illusions whose bases have been debated during the last century or more, and all the less well-known (some would say boring) psychophysical functions that vision scientists have amassed. The focus in what follows is on the more intriguing "gee whiz" category of so-called illusions, setting aside psychophysical functions for the time being. Work on these phenomena had, for the most part, been relegated to psychology; most neuroscientists (myself included until the late 1990s) regarded illusions as a by-product of the neuronal interactions that neurophysiologists were unraveling, and of no great interest in their own right. There were some exceptions to this parochial mindset. A few outstanding vision scientists—Horace Barlow at Cambridge is perhaps the preeminent example—combined work on visual physiology (Barlow described receptive fields in the frog retina about the same time Kuffler described them in the cat retina) with studies of perception. But this was not the tradition that I had been exposed to. A case in point is Hubel and Wiesel's magisterial retrospective published in 2006 called Brain and Visual Perception, a book that combines their key papers with commentaries on what they did and why they did it. Despite the title, the book has no index entry for perception, no definition of this term in the glossary or elsewhere, and little or no discussion in the papers or commentaries of the parallel work on perceptual issues that was going on in psychology and psychophysics during the 25 years of their collaboration.
Perception is defined as the conscious product of sensory processing, given to us as the qualities that we attribute back to objects and conditions in the world (the way things look, sound, feel, smell, and taste). This concept, however, is a very limited description of what is really occurring in the nervous system. Although we think in terms of vision, audition, bodily sensations, taste, and smell, we are oblivious to the output of many other sensory modalities; the information produced by sensory systems that monitor muscle length and tension that Hunt and Kuffler had studied, blood pressure, blood gas levels, and a host of other parameters critical to survival never enters our awareness.
Earlier in the twentieth century, the Gestalt school of psychology was the most popular approach to explaining perception. The most famous representatives of this way of thinking were a trio of German psychologists who emigrated to the United States in the 1920s and 1930s: Max Wertheimer, Wolfgang Kühler, and Kurt Kofka. The assumption of the Gestaltists was that many perceptual phenomena arise because we tend to see scenes in terms of organized forms or wholes. ( Gestalt means "form" or "shape" in German.) The organizational rules that presumably applied to a complete scene compared to its elements would be different and would therefore lead to different appearances. This way of thinking had the merit of taking the complete content of scenes into account, but the Gestalt laws that emerged seemed little more than another way of describing the perceptual effects. Although Gestalt psychologists documented many challenging phenomena (the Gelb room is a good example), they didn't link their "laws" to biology. As a result, enthusiasm for this approach even among psychologists had begun to fade by midcentury.
By the 1950s, explaining the discrepancies between lightness or brightness perceptions and luminance was done primarily on the basis of electrophysiology. Based on the evidence that had begun to emerge from such studies, one way of rationalizing the effects elicited by stimuli such as the standard simultaneous brightness contrast stimulus in Figure 8.1 was to consider them incidental consequences of visual processing—a price paid, so to speak, for achieving some larger goal of the visual system. For example, the central region of the retinal output cells' receptive fields was known to have a surround with the opposite functional polarity, an organization that presumably enhances the detection of the edges ( Figure 8.5). As a result, retinal neurons whose receptive fields lie over a light-dark boundary but with their central regions either within or without the dark region of an image would be expected to generate action potentials at a somewhat different rate. This physiological observation made it attractive to suppose that the patch on the dark background in Figure 8.1 looks lighter or brighter than the patch on the light background because of this difference in the retinal output to the rest of the visual brain. Kuffler, who had reported the organization of receptive fields of neurons in the cat retina in 1953, avoided this sort of assertion, probably because he felt that tying receptive field properties directly to perception was a dangerous game, which it is. An appreciation of Kuffler's conservative philosophy might be one of the reasons why his protégés Hubel and Wiesel paid little attention to visual perception.
However, the perceptions elicited by other stimulus patterns show that simultaneous brightness contrast effects such as those in Figure 8.1 are not an incidental consequence of retinal processing. In Figure 8.6, the four target patches on the left are surrounded by relatively more higher-luminance (lighter) territory than lower-luminance (darker) territory; nevertheless, they appear lighter than the three targets on the right, which are surrounded by relatively more lower-luminance (darker) territory. Although the average luminance values of the surrounding contexts of the patches in the stimulus are effectively opposite those in standard simultaneous brightness stimulus in Figure 8.1, the lightness/brightness differences elicited are about the same in both direction and magnitude. The effect of the scene in Figure 8.2 would also be difficult to explain as an incidental consequence of retinal neuron receptive field properties.
Figure 8.5 The receptive field properties of retinal ganglion cells and their output to higher visual processing centers. The upper panel shows the receptive fields of several retinal neurons (ganglion cells) overlying a light-dark boundary. The lower panel shows the different firing rates of ganglion cells as a function of their position with respect to the boundary. In principle, this difference might cause the lightness/brightness contrast effects in Figure 8.1 as an incidental consequence of neural processing at the input stages of vision. (After Purves and Lotto, 2003)
ctf Cl tidge
Was this article helpful?