Retinal Processing

17-2 How do the rods and cones process information, and what is the path information travels from the eye to the brain?

Imagine that you could follow behind a single light-energy particle after it entered your eye. First, you would thread your way through the retina’s sparse outer layer of cells. Then, reaching the back of the eye, you would encounter its buried receptor cells, the rods and cones (FIGURE 17.4). There, you would see the light energy trigger chemical changes. That chemical reaction would spark neural signals, activating nearby bipolar cells. The bipolar cells in turn would activate the neighboring ganglion cells, whose axons twine together like the strands of a rope to form the optic nerve. The optic nerve is an information highway to your brain, where your thalamus stands ready to distribute the information it receives. This nerve can send nearly 1 million messages at once through its nearly 1 million ganglion fibers. (The auditory nerve, which enables hearing, carries much less information through its mere 30,000 fibers.)

We pay a small price for this eye-to-brain highway. Where the optic nerve leaves the eye, there are no receptor cells—creating a blind spot (FIGURE 17.5 below). Without seeking your approval, your brain fills in the hole.

Rods and cones differ in where they’re found and what they do (TABLE 17.1). Cones cluster in and around the fovea, the retina’s area of central focus (see FIGURE 17.3). Many cones have their own hotline to the brain, which devotes a large area to input from the fovea. These direct connections preserve the cones’ precise information, making them better able to detect fine detail.

Rods don’t have dedicated hotlines. Rods share bipolar cells, which send combined messages. To experience this rod-cone difference in sensitivity to details, pick a word in this sentence and stare directly at it, focusing its image on the cones in your fovea. Notice that words a few inches off to the side appear blurred? Their image strikes the outer regions of your retina, where rods predominate. Thus, when driving or biking, rods help you detect a car in your peripheral vision well before you perceive its details.

Cones also enable you to perceive color. In dim light they become ineffectual, so you see no colors. Rods, which enable black-and-white vision, remain sensitive in dim light. Several rods will funnel their faint energy output onto a single bipolar cell. Thus, cones and rods each provide a special sensitivity—cones to detail and color, and rods to faint light and peripheral motion.

When you enter a darkened theater or turn off the light at night, your eyes adapt. Your pupils dilate to allow more light to reach your retina, but it typically takes 20 minutes or more before your eyes fully adapt. You can demonstrate dark adaptation by closing or covering one eye for 20 minutes. Then make the light in the room not quite bright enough to read this book with your open eye. Now open the dark-adapted eye and read (easily). This period of dark adaptation matches the average natural twilight transition between the Sun’s setting and darkness. How wonderfully made we are.

At the entry level, the retina’s neural layers don’t just pass along electrical impulses; they also help to encode and analyze sensory information. (The third neural layer in a frog’s eye, for example, contains the “bug detector” cells that fire only in response to moving fly-like stimuli.) In human eyes, any given retinal area relays its information to a corresponding location in the visual cortex, in the occipital lobe at the back of your brain (FIGURE 17.6).

The same sensitivity that enables retinal cells to fire messages can lead them to misfire, as you can demonstrate. Turn your eyes to the left, close them, and then gently rub the right side of your right eyelid with your fingertip. Note the patch of light to the left, moving as your finger moves.

Why do you see light? Why at the left? This happens because your retinal cells are so responsive that even pressure triggers them. But your brain interprets their firing as light. Moreover, it interprets the light as coming from the left—the normal direction of light that activates the right side of the retina.

17-3 How do we perceive color in the world around us?

One of vision’s most basic and intriguing mysteries is how we see the world in color. In everyday conversation, we talk as though objects possess color: “A tomato is red.” Recall the old question, “If a tree falls in the forest and no one hears it, does it make a sound?” We can ask the same of color: If no one sees the tomato, is it red?

The answer is No. First, the tomato is everything but red, because it rejects (reflects) the long wavelengths of red. Second, the tomato’s color is our mental construction. As Isaac Newton (1704) noted, “The [light] rays are not colored.” Like all aspects of vision, our perception of color resides not in the object but in the theater of our brains, as evidenced by our dreaming in color.

How, from the light energy striking the retina, does our brain construct our experience of color—and of such a multitude of colors? Our difference threshold for colors is so low that we can discriminate more than 1 million different color variations (Neitz et al., 2001). At least most of us can. For about 1 person in 50, vision is color deficient—and that person is usually male, because the defect is genetically sex linked.

Modern detective work on the mystery of color vision began in the nineteenth century, when Hermann von Helmholtz built on the insights of an English physicist, Thomas Young. Any color can be created by combinations of different amounts of light waves of three primary colors—red, green, and blue. Knowing that, Young and von Helmholtz formed a hypothesis: The Young-Helmholtz trichromatic (three-color) theory thus implies that the eye’s receptors do their color magic in teams of three. The eye must have three corresponding types of color receptors. Years later, researchers measured the response of various cones to different color stimuli and confirmed that the retina does have three types of color receptors, each especially sensitive to one of three colors. And those colors are, in fact, red, green, and blue. When we stimulate combinations of these cones, we see other colors. We see yellow when light stimulates both red-sensitive and green-sensitive cones.

Most people with color-deficient vision are not actually “colorblind.” They simply lack functioning red- or green-sensitive cones, or sometimes both. Their vision—perhaps unknown to them, because their lifelong vision seems normal—is monochromatic (one-color) or dichromatic (two-color) instead of trichromatic, making it impossible to distinguish the red and green in FIGURE 17.7. Dogs, too, lack receptors for the wavelengths of red, giving them only limited, dichromatic color vision (Neitz et al., 1989).

But why do people blind to red and green often still see yellow? And why does yellow appear to be a pure color and not a mixture of red and green, the way purple is of red and blue? As Ewald Hering soon noted, trichromatic theory leaves some parts of the color vision mystery unsolved.

Hering, a physiologist, had found a clue in afterimages. Stare at a green square for a while and then look at a white sheet of paper, and you will see red, green’s opponent color. Stare at a yellow square and its opponent color, blue, will appear on the white paper. (To experience this, try the flag demonstration in FIGURE 17.8.) Hering formed another hypothesis: There must be two additional color processes, one responsible for red-versus-green perception, and one for blue-versus-yellow.

Indeed, a century later, researchers also confirmed Hering’s opponent-process theory. Three sets of opponent retinal processes—red-green, yellow-blue, and white-black—enable color vision. Recall that the thalamus relays visual information from the retina to the visual cortex. In both the retina and the thalamus, some neurons are “turned on” by red but “turned off” by green. Others are turned on by green but off by red (DeValois & DeValois, 1975). Like red and green marbles sent down a narrow tube, “red” and “green” messages cannot both travel at once. Red and green are thus opponents, so we do not experience a reddish green. But red and blue travel in separate channels, so we can see a reddish-blue magenta.

How then do we explain afterimages, such as in the flag demonstration? By staring at green, we tire our green response. When we then stare at white (which contains all colors, including red), only the red part of the green-red pairing will fire normally.

The present solution to the mystery of color vision is therefore roughly this: Color processing occurs in two stages. The retina’s red, green, and blue cones respond in varying degrees to different color stimuli, as the Young-Helmholtz trichromatic theory suggested. Their responses are then processed by opponent-process cells, as Hering’s theory proposed.

17-4 Where are feature detectors located, and what do they do?

Scientists once likened the brain to a movie screen, on which the eye projected images. But then along came David Hubel and Torsten Wiesel (1979), who showed that our brain’s computing system deconstructs visual images and then reassembles them. Hubel and Wiesel received a Nobel Prize for their work on feature detectors, nerve cells in the brain that respond to a scene’s specific visual features—to particular edges, lines, angles, and movements.

Using microelectrodes, they had discovered that some neurons fired actively when cats were shown lines at one angle, while other neurons responded to lines at a different angle. They surmised that these specialized neurons in the occipital lobe’s visual cortex, now known as feature detectors, receive information from individual ganglion cells in the retina. Feature detectors pass this specific information to other cortical areas, where teams of cells (supercell clusters) respond to more complex patterns.

For biologically important objects and events, monkey brains (and surely ours as well) have a “vast visual encyclopedia” distributed as specialized cells (Perrett et al., 1988, 1992, 1994). These cells respond to one type of stimulus, such as a specific gaze, head angle, posture, or body movement. Other supercell clusters integrate this information and fire only when the cues collectively indicate the direction of someone’s attention and approach. This instant analysis, which aided our ancestors’ survival, also helps a soccer player anticipate where to strike the ball, and a driver anticipate a pedestrian’s next movement. One temporal lobe area by your right ear (FIGURE 17.9) enables you to perceive faces and, thanks to a specialized neural network, to recognize them from varied viewpoints (Connor, 2010). If stimulated in this area, you might spontaneously see faces. If this region were damaged, you might recognize other forms and objects, but not familiar faces.

Researchers can temporarily disrupt the brain’s face-processing areas with magnetic pulses. When this happens, people cannot recognize faces, but they can recognize houses, because the brain’s face perception occurs separately from its object perception (McKone et al., 2007; Pitcher et al., 2007). Thus, functional MRI (fMRI) scans have shown different brain areas activating when people viewed varied objects (Downing et al., 2001). Brain activity is so specific (FIGURE 17.10) that, with the help of brain scans, “we can tell if a person is looking at a shoe, a chair, or a face, based on the pattern of their brain activity,” noted one researcher (Haxby, 2001).

17-5 How does the brain use parallel processing to construct visual perceptions?

Our brain achieves these and other remarkable feats by parallel processing: doing many things at once. To analyze a visual scene, the brain divides it into subdimensions—motion, form, depth, color—and works on each aspect simultaneously (Livingstone & Hubel, 1988). We then construct our perceptions by integrating the separate but parallel work of these different visual teams (FIGURE 17.11).

To recognize a face, your brain integrates information projected by your retinas to several visual cortex areas, compares it with stored information, and enables you to recognize the face: Grandmother! Scientists have debated whether this stored information is contained in a single cell or distributed over a vast network of cells. Some supercells—grandmother cells—do appear to respond very selectively to 1 or 2 faces in 100 (Bowers, 2009). The whole face recognition process requires tremendous brain power—30 percent of the cortex (10 times the brain area devoted to hearing).

Destroy or disable a neural workstation for a visual subtask, and something peculiar results, as happened to “Mrs. M.” (Hoffman, 1998). After a stroke damaged areas near the rear of both sides of her brain, she found herself unable to perceive movement. People in a room seemed “suddenly here or there but I [did not see] them moving.” Pouring tea into a cup became a challenge because the fluid appeared frozen—she could not perceive it rising in the cup.

After stroke or surgery has damaged the brain’s visual cortex, others have experienced blindsight. Shown a series of sticks, they report seeing nothing. Yet when asked to guess whether the sticks are vertical or horizontal, their visual intuition typically offers the correct response. When told, “You got them all right,” they are astounded. There is, it seems, a second “mind”—a parallel processing system—operating unseen. These separate visual systems for perceiving and for acting illustrate once again the astonishing dual processing of our two-track mind.

* * *

Think about the wonders of visual processing. As you read this page, the letters are transmitted by reflected light rays onto your retina, which triggers a process that sends formless nerve impulses to several areas of your brain, which integrates the information and decodes meaning, thus completing the transfer of information across time and space from our minds to your mind (FIGURE 17.12). That all of this happens instantly, effortlessly, and continuously is indeed awesome. As Roger Sperry (1985) observed, the “insights of science give added, not lessened, reasons for awe, respect, and reverence.”