Psychology › Section E · Sensation and perception › Topic E.3

Vision

Sight feels effortless, but it is a long chain of work: optics that focus light, receptors that transduce it, and brain pathways that rebuild colour, shape, motion and depth. This topic follows that chain from the cornea to the cortex.

The eye as an optical instrument

Light first passes through the cornea, the clear front surface of the eye, which does most of the focusing. It then enters through the pupil, an opening whose size is set by the muscles of the iris, the coloured ring around it. The pupil widens in dim light to let more in and narrows in bright light; it also widens with emotional arousal. Behind the pupil, the lens fine-tunes the focus by changing shape, flattening for distant objects and thickening for near ones. The focused image lands upside down on the retina, the light-sensitive layer lining the back of the eye.

Rods and cones

The retina holds two kinds of photoreceptor. Cones, about 6 million per eye, work best in bright light, give fine detail and are responsible for colour vision. They are packed most densely in the fovea, a small pit at the centre of the retina where focused images fall, which is why you look straight at something to read it. Rods, about 120 million per eye, are spread across the rest of the retina. They are far more sensitive to dim light but do not signal colour and give poorer detail. They also make the edges of the visual field good at catching movement.

From the retina to the cortex

Rods and cones pass their signals through bipolar cells to ganglion cells, whose axons bundle together as the optic nerve. Where that nerve leaves the eye there are no receptors, creating a blind spot. You do not notice it because the two eyes' blind spots fall on different parts of the scene and the brain fills in the gap. The two optic nerves meet at the optic chiasm, where fibres are re-sorted so that the left half of the visual field goes to the right hemisphere, and the right half to the left. After a relay in the thalamus, signals reach the primary visual cortex in the rearmost lobe, the occipital lobe.

Feature detectors and parallel pathways

Recording from single neurons in cats and monkeys, David Hubel and Torsten Wiesel found cells in the visual cortex that respond to edges and lines at particular orientations; they shared the 1981 Nobel Prize for this work. They also showed a critical period: when one eye of a young kitten was kept closed, cortex normally driven by that eye was taken over by the other, work that still prompts ethical debate. Beyond the primary cortex, processing splits. A 'what' pathway into the temporal lobe identifies objects, including areas that respond strongly to faces, while a 'where/how' pathway into the parietal lobe tracks location and guides action.

Worked example: walking into a dark cinema

You walk from a sunny street into a dark cinema and at first can see almost nothing. Explain what happens over the next half hour. 1. Outside, vision was dominated by cones, which need plenty of light. 2. In the dark the cones' sensitivity improves only a little, within roughly the first 10 minutes. 3. The rods keep gaining sensitivity for about 20 to 30 minutes, which is why seats and faces gradually appear. 4. The scene still looks nearly colourless, because the rods now doing the work do not signal colour. 5. Check: when a bright scene fills the screen, the cones respond again and colours return.

Colour vision: the trichromatic theory

Thomas Young proposed in the early nineteenth century, and Hermann von Helmholtz developed later, that colour vision rests on three kinds of receptor. They were right: the human retina has three cone types, most sensitive to short, medium and long wavelengths, with peaks near 420, 530 and 560 nm. They are often loosely called blue, green and red cones, but their sensitivities overlap widely. Any wavelength excites all three by different amounts, and the brain reads colour from the ratio of their responses. That is how a screen's red, green and blue pixels can mimic a huge range of colours.

Colour vision: the opponent-process theory

In the late nineteenth century Ewald Hering noticed things three receptors alone did not explain. No one sees a reddish green or a yellowish blue, and staring at one colour leaves an afterimage in its partner. He proposed that colour is coded in opposing pairs: red against green, blue against yellow, and black against white, so that a cell excited by one member of a pair is inhibited by the other. Both theories are correct at different levels: trichromatic coding happens in the cones, and opponent coding in later cells, from the retinal ganglion cells onwards, which compare the cones' outputs.

Worked example: a negative afterimage

Stare at a bright green square for 30 seconds, then look at a white wall. You see a reddish square. Explain it with the opponent-process theory. 1. While you stare, the green side of the red-green opponent cells is driven hard and adapts, becoming less responsive. 2. White light stimulates all colour channels about equally. 3. With the green side fatigued, the red side now outweighs it where the square was. 4. So the afterimage appears in the opponent colour, red. 5. Check: the same logic predicts that a yellow square leaves a bluish afterimage, which is what people report.

Colour vision deficiency

Some people inherit cones that are missing or shifted in sensitivity. The commonest form is red-green deficiency, carried on the X chromosome, so it is far more common in males: about 8% of males of European descent but about 0.4% of females, with lower rates in males of many other ancestries. Most people affected do see colour, but find some pairs hard to tell apart, so 'colour blindness' usually overstates it; seeing no colour at all is very rare. Eye specialists test for it with plates of coloured dots hiding numbers.

Depth perception

Depth perception is the ability to see the world in three dimensions and judge distance. Binocular cues need both eyes. In binocular disparity, each eye gets a slightly different view, and the brain uses the difference: hold up a finger and close each eye in turn, and the finger jumps. Nearer objects give larger disparity, the principle behind 3-D films. Convergence, the eyes turning inward for near objects, is a second binocular cue. Monocular cues work with one eye, which is how paintings look deep: linear perspective (parallel lines meeting in the distance), interposition (nearer objects block farther ones), relative size, height in the visual field, texture gradient and motion parallax. The neuroscientist Bruce Bridgeman, stereoblind all his life, reported first seeing stereoscopic depth in his late sixties, at a 3-D film in 2012.

Common mistakes to avoid

First, swapping rods and cones: cones give colour and detail; rods give dim-light and peripheral vision. Second, thinking the left eye feeds the right hemisphere: it is the left visual field, from both eyes. Third, calling the cones pure red, green and blue detectors: their sensitivities overlap. Fourth, treating the two colour theories as rivals: they describe different stages. Fifth, assuming colour vision deficiency means seeing in black and white. Sixth, thinking depth needs two eyes: monocular cues give plenty.

Key takeaways

The cornea and lens focus an inverted image on the retina. About 6 million cones per eye, concentrated in the fovea, give colour and detail; about 120 million rods give dim-light and peripheral vision, and dark adaptation takes about 20 to 30 minutes. The optic nerve's exit leaves a blind spot; at the optic chiasm each visual field crosses to the opposite hemisphere. Hubel and Wiesel found orientation-tuned feature detectors. Colour is trichromatic at the cones and opponent later on, which explains afterimages. Red-green deficiency is X-linked. Depth comes from binocular disparity and convergence plus many monocular cues.

Flashcards

Flashcard

Cornea

Clear front surface that does most of the eye's focusing

Light enters here first.

Test your understanding

Question 1 of 5

In a dimly lit room you can make out shapes but almost no colour. Why?

Question 2 of 5

A light flashes in your left visual field. Where does the signal first reach the cortex?

Question 3 of 5

After staring at a bright red circle, you look at a white page. What colour is the afterimage most likely to be?

Question 4 of 5

How do the trichromatic and opponent-process theories of colour vision relate to each other?

Question 5 of 5

Which depth cue depends on having two working eyes?

← Previous topicNext topic →