How vision works

HANK GREEN: Take a good long look at this. We're gonna mess with your brain. This is the first stage of an optical illusion. Many illusions use patterns of light or perspective to exploit the disconnect that exists between sensation and perception—between what your eyes see, and what your brain understands. But not all illusions work that way. Some produce ghost effects, or afterimages, that take advantage of glitches in the physiology of human vision. Like this flag. I'm not trying to make a political statement here, and I'm not going to ask you to swear allegiance to the Republic of Hank or anything. I mean, If I was gonna start my own country, my flag would be way cooler than that, not that I've thought about that a lot. And now, look at this white screen. If you looked at this flag for at least 30 seconds without moving your eyes, you'll see something, even though the screen is blank—an afterimage of the flag. But instead of being turquoise and black and yellow, it's red, white and blue.

OK, so that's pretty cool, but I'm not just here to entertain you. This kind of illusion is actually a great way to explain your very complex sense of vision. And I do mean complex. Nearly 70 per cent of all the sensory receptors in your whole body are in the eyes. Not only that, but in order for you to see, perceive, and recognise something—whether it's a flag or a handsome guy in glasses and a sport coat sitting behind a desk—nearly half of your entire cerebral cortex has to get involved.

Vision is considered the dominant sense of humans, and while we can get along without it, and it can be tricked, what you are about to learn is not an illusion.

When we talked about your sense of hearing, we began with the mechanics of sound. So before we get to how your eyeballs work, it makes sense to talk about what they're actually seeing—light bouncing off of stuff.

Light is electromagnetic radiation travelling in waves. Remember how the pitch and loudness of a sound is determined by the frequency and amplitude of its wave? Well it's kind of similar with light, except that the frequency of a light wave determines its hue, while the amplitude relates to its brightness. 

We register short waves at high frequencies as bluish colours, while long, low frequencies look reddish to us. Meanwhile, that red might appear dull and muted if the wave is moving at a lower amplitude, but super bright if the wave has greater amplitude and thus higher intensity. 

But the visible light we're able to see is only a tiny chunk of the full electromagnetic spectrum, which ranges from short gamma and X rays all the way to long radio waves.

Just as the ear's mechanoreceptors or the tongue's chemoreceptors convert sounds and chemicals into action potentials, so too do your eyes' photoreceptors convert light energy into nerve impulses that the brain can understand. 

To figure out how all this works, let's start with understanding some eye anatomy.

Some of the first things you'll notice around your average pair of eyes are all the outer accessories, like the eyebrows that help keep the sweat away if you forgot your headband at raquetball, and the super-sensitive eyelashes that trigger reflexive blinking, like if you're on a sandy beach in a wind storm.

These features, along with the eyelids and tear-producing lacrimal apparatus are there to help protect your fragile eyeballs.

The eyeball itself is irregularly spherical, with an adult diameter of about 2.5 centimeters. It's essentially hollow, full of fluids that help to keep its shape, and you can really only see about the anterior sixth of the whole ball. The rest of it is tucked into a pocket of protective fat, tethered down by six straplike extrinsic eye muscles, and jammed into the bony orbit of your skull.

While all this gear generally does a fantastic job of keeping your eyeballs inside of your head, which is good, on very rare occasions, perhaps after head trauma, or even a really intense sneeze, those suckers can pop right out—a condition called globe luxation, which you really do not want to Google. I'll just sit here while you Google it.

Now, you don't need to pop out an eyeball in order to learn how it's structured. I'll save you the trouble and tell you that its wall is made up of three distinct layers—the fibrous, vascular, and inner layers. The outermost fibrous layer is made of connective tissue. Most of it is that white stuff called the sclera, while the most anterior part is the transparent cornea.

The cornea is like the window that lets light into the eye, and if you've ever experienced the excruciating pain of a scratched one, you know how terrible it can be to damage something so loaded with pain receptors.

Going down a little deeper, the wall's middle vascular layer contains the posterior choroid, a membrane that supplies all of the layers with blood. In the anterior there's also the ciliary body, a ring of muscle tissue that surrounds the lens, but the most famous part of this middle layer is the iris. The iris is that distinctive coloured part of the eye that is uniquely yours. It's made up of smooth muscle tissue, shaped like a flattened donut, and sandwiched between the cornea and the lens. Those circular sphincter muscles—yeah, that's right, you've got sphincters everywhere!—contract and expand, changing the size of the dark dot of your pupil.

The pupil itself is just the opening in the iris that allows light to travel into the eye. You can see how an iris protects the eye from taking too much light in if you shine a flashlight in your friend's eye in a dark room. Their pupils will go from dilated to pinpoints in a couple of seconds.

Light comes in through the cornea and pupil and hits the lens—the convex, transparent disc that focuses that light and projects it onto the retina, which makes up the inner layer at the back of the eyeball.

Your retinas are loaded with millions of photoreceptors which do the crucial work of converting light energy into the electrical signals that your brain will receive. These receptor cells come in two flavours—rods and cones—which I'll come back to in a minute. But the retina itself has two layers, the outer pigmented layer that helps absorb light so it doesn't scatter around the eyeball, and the inner neural layer. And this layer, as the name indicates, contains neurons—not only the photoreceptors but also bipolar neurons and ganglion neurons. These two kinds of nerve cells combine to produce a sort of pathway for light, or at least data about light.

Bipolar neurons have synapses at both ends, forming a kind of bridge—on one end it synapses with a photoreceptor, and at the other, it synapses with a ganglionic neuron, which goes on to form the optic nerve. 

So say you've just been hit with a blinding flashlight beam. That light hits your posterior retina and spreads from the photoreceptors to the bipolar cells just beneath them, to the innermost ganglion cells where they then generate action potentials.

The axons of all those ganglion cells weave together to create the thick, ropey optic nerve—your second cranial nerve—which leaves the back of your eyeball and carries those impulses up to the thalamus and then on to the brain's visual cortex. 

So that's the basic anatomy and event sequencing of human vision, but what I really want to talk about are those two types of photoreceptors—your rods and your cones.

Cones sit near the retina's centre, and detect fine detail and colour. They can be divided into red, green and blue sensitive types, based on how they respond to different types of light. But they're not very sensitive, and they really only hit their activation thresholds in bright conditions. Rods, on the other hand, are more numerous, and more light-sensitive. But they can't pick up real colour. Instead, they only register a greyscale of black and white. They hang out around the edges of your retinas and rule your peripheral vision.

Since these receptors function so differently, you might not be surprised to learn that your rods and your cones are also wired to your retinas in different ways.

As many as 100 different rods may connect to a single ganglion cell—but because they all send their information to the ganglion at once, the brain can't tell which individual rods have been activated, and that's why they're not very good at providing detailed images. All they can really do is give you information about objects' general shape, or whether it's light or dark.

Each cone, by contrast, gets its own personal ganglion cell to hook up with, which allows for very detailed colour vision, at least if conditions are bright enough.

And all this brings us back to that weird flag. Why does staring at this flag and then looking at an empty white space make us see a phantom flag of different colours? Well, it begins with the fact that our photoreceptors can make us see afterimages.

Some stimuli, like really brilliant colours or really bright lights, are so strong that our photoreceptors will continue firing action potentials even after we close our eyes or look away. The other part of the illusion has to do with another bug in our visual programming... and it's just that our cones can get tired.

If you stare long enough at a brightly coloured image, your cones will receive the same stimulus for too long, and basically stop responding. In the case of the flag, you looked at an image with bright turquoise stripes. Because your retinas contain green, red and blue-sensitive cones, the blue and green ones got tired after a while, leaving only the red ones left to fire. Then, you looked at a white screen. That white light included all of the colours and wavelengths of visible light. So your eyes were still receiving red, green, and blue light—but only the red cones were able to respond. As a result, when the afterimage began to appear, those stripes looked red.

The same thing happened to your rods, except since they only register black and white, the afterimage was like looking at a negative of a photograph—dark replaced with light. That's how those black stars and stripes turned white.

So yes, human vision is fallible, but those mistakes that it makes can help us understand that wonderfully complex system. And that wonderfully complex system probably helped you learn about the anatomy and physiology of vision today, starting with the structure of the eye and its three layers: the fibrous, vascular, and inner layers. We spent most of our time exploring the inner layer, which consists of the retina and its three kinds of neurons: photoreceptors, bipolar cells and ganglion neurons. And after learning how to tell our rods from our cones, we then dissected how the weird flag illusion works.