The colour magenta does not exist anywhere in the spectrum of visible light, and your brain manufactures it on the spot whenever red and blue cones fire together, inventing a hue to fill a gap that physics never bothered to provide.

The colour magenta does not exist anywhere in the spectrum of visible light, and your brain manufactures it on the spot whenever red and blue cones fire together, inventing a hue to fill a gap that physics never bothered to provide. Featured Image

Stare at a rainbow and try to find magenta. It will not be there. The visible spectrum runs from roughly 700 nanometre red at one end to about 380 nanometre violet at the other, with orange, yellow, green, blue and indigo sliding between them in a continuous band first split out by Isaac Newton with a prism in 1666.

Magenta — the hot pink of bougainvillea petals, the M in CMYK printer ink, the top stripe of the bisexual pride flag — has no wavelength. It is not a colour of light. It is a colour the brain invents, in real time, every time the red-sensitive and blue-sensitive cones at the back of the eye fire at the same time without anything in the middle telling green to fire too.

The human visual system is, in other words, making up a hue to paper over a gap that physics did not supply. And magenta is only the most famous example of this kind of editorial work. The brain is constantly fabricating visual information that was never there.

What the cones actually see

The retina contains three types of colour-sensitive cone cells, usually labelled S, M and L for the short, medium and long wavelengths they respond to most strongly. The S cones peak around blue, the M cones around green, the L cones around yellow-green and red. Every colour a person sees is the brain’s interpretation of how strongly each of those three cone types is firing at a given moment. Yellow light, for example, fires the M and L cones in roughly equal measure, and the brain reads that combination as yellow.

Red and blue sit at opposite ends of the visible spectrum. A single photon cannot be both. When the S cones (blue) and the L cones (red) fire strongly at the same time, with the M cones (green) quiet in the middle, the brain faces a small logical problem. There is no single wavelength that produces that signal. So it does not try to find one. It invents a new colour — a non-spectral colour — and places it on the opposite side of green on the colour wheel. That colour is magenta.

Magenta is, in a literal sense, what the absence of green looks like.

A pigment named after a battle

The word itself is younger than most people assume. The dye fuchsine was synthesised in 1859 and named after the fuchsia flower. Later that year, after the Franco-Sardinian victory over Austria at the Battle of Magenta in northern Italy, it was renamed magenta. As the Ashmolean Museum recounts in its history of the colour, the rebranding rode the wave of new aniline dyes pouring out of coal-tar chemistry. The colour exists in pigment because that chemistry made it possible. It exists in perception because the brain refuses to leave the colour wheel open at the bottom.

The hole in the middle of every eye

Magenta is one kind of invention. The blind spot is another, and it is more startling because it is geometric rather than chromatic. Every human eye has a region at the back of the retina where the optic nerve exits toward the brain, carrying about a million axons in a tight bundle. There are no photoreceptors in that patch. Light that falls on it is not seen. As Popular Mechanics describes it, the optic disc is the structural reason the gap exists in the first place — the wiring has to leave the eye somewhere, and where it leaves, there is no film to expose.

The blind spot sits about 15 degrees off the centre of vision in each eye, out toward the temple rather than the nose. The optic disc itself lies on the nasal side of the retina, but the eye’s lens flips the incoming image, so the gap shows up in the temporal half of the visual field. It is not small. As fine-scale measurements of the blind spot put it, the patch spans roughly five to seven degrees of vision — wide enough to swallow the full moon several times over. And yet nobody walks around seeing a black hole floating in the middle of the world.

The simplest way to prove the blind spot is there involves a piece of paper, a pen, and one closed eye. Draw a small cross on the left side of the page and a small dot about six inches to the right of it. Close the left eye. Stare at the cross with the right eye and slowly move the page toward the face. At a certain distance, somewhere around eight to ten inches, the dot will vanish. The paper does not go blank where the dot was. The white of the page simply continues, as if the dot had never been drawn.

That is the brain editing. It is not failing to notice the missing information. It is actively filling the missing information in, with whatever is around the edges of the gap. White paper around the hole becomes white paper inside the hole. A striped pattern around the hole becomes a striped pattern inside the hole. Laboratory work on this filling-in at the blind spot finds it happens almost instantly and entirely below awareness. The fill is so smooth that, until the demonstration works, it is genuinely hard to believe the spot exists at all.

Predictive processing, or seeing what should be there

The current best explanation for this kind of seamless invention is called predictive processing. The basic idea is that the brain is not a passive receiver of sensory data. It is a prediction engine. It guesses, constantly, what the world is most likely to look like next, and uses the actual signals from the eyes mainly to correct its guesses when they go wrong.

The blind spot is a place where there is no signal at all. So there is nothing to correct the guess. The brain looks at the texture, colour and pattern of the surrounding retina, predicts the most boring continuation of that pattern, and renders the prediction into conscious vision. As the philosopher and cognitive scientist Andy Clark describes the predictive brain, perception is less a recording of the world than the brain’s best guess about it, with sensory input brought in mainly to refine the guess when it misses.

Magenta works the same way. The cone signals do not match any wavelength, so the brain offers its most useful inference: a placeholder colour, distinct from anything spectral, that lets red-plus-blue feel as different from green as green feels from white.

The blind spot has become a small but serious tool in consciousness research. A team led by the neuroscientist Lars Muckli at the University of Glasgow has laid out a study protocol in PLOS One that uses the filled-in region to ask a sharper question: what exactly is the difference between sensory information that reaches awareness and sensory information that does not? Because the brain treats the fabricated patch of vision exactly the same as the real one — same confidence, same texture, same conscious presence — it offers a controlled way to compare invented experience against received experience inside the same eye, as reporting on the protocol lays out.

If invented vision and real vision feel identical from the inside, then whatever consciousness is doing in the blind spot, it is presumably doing everywhere else too. The fill is not a glitch. It is a window onto the normal operation.

Other things the brain quietly invents

Once a person starts looking, the list of fabrications gets long. Every time the eyes flick from one fixation point to another, several times per second, all day long, the motion blur disappears. The world looks stable even though the image on the retina is being violently yanked around. The brain throws away the motion blur and stitches the still frames together.

Peripheral vision is mostly colourblind and almost as low-resolution as a thumbnail image, yet a wide colourful world appears to extend smoothly out to the edges. The brain is interpolating. Large changes in a scene can go unnoticed if they happen during a blink or a cut, because the prediction did not flag them as needing attention.

Even the way colours look stays roughly constant under wildly different lighting — sunset orange, fluorescent green, candlelight yellow — because the brain silently corrects for the light source. A white shirt looks white at noon and at midnight under a sodium lamp, though the wavelengths reaching the eye are completely different. The famous internet argument about whether “the dress” was blue and black or white and gold was, in part, a fight about which light source the brain had silently subtracted — the same kind of mismatch that makes colours on an uncalibrated screen look wrong until the display is corrected.

From an evolutionary standpoint, a small blind spot in the middle of a hunter-gatherer’s field of view is a problem. A predator could be standing in it. The fill-in is not a vanity project. It is a survival feature, paired with the fact that the two eyes’ blind spots cover different parts of the world, so binocular vision plugs most of the gap with real data. The brain only has to invent when one eye is closed or when the same point falls into both blind spots, which almost never happens at normal viewing distances.

Magenta, similarly, is doing useful work. Without a non-spectral colour to close the loop between red and blue, the colour wheel would be a colour line, and discrimination between flowers, fruit, skin tones and signals would be coarser. The invention pays for itself.

The strange part

The strange part is not that the brain edits. It is that the edits are invisible from the inside. There is no flicker, no seam, no sense of a placeholder being slotted in. The experience of looking at a rainbow feels complete, even though magenta is missing from it and the brain has had to add the bridge between the ends in any picture that includes both. The experience of looking at a blank wall feels complete, even though a coin-sized patch of that wall is being painted in by inference rather than light.

The colours on a screen right now include several that the universe outside the screen has never produced as a single wavelength. The text being read is being assembled, letter by letter, from a retinal image that is upside down, low-resolution at the edges, interrupted by two blind spots and blanked out several times a second by eye movements. What arrives in consciousness is a clean page. Somewhere behind the eyes, a very old prediction engine is quietly filling in the parts that were never there, and has been doing it, without anyone noticing, for every waking second of every human life.

Subscribe to our newsletter!

Our latest tutorials delivered straight to your inbox

Make Tech Easier Editorial Team Avatar

Read next

When Bell Labs engineer Karl Jansky pointed a rotating antenna at the sky in 1932 looking for sources of transatlantic radio static, he kept picking up a faint hiss that peaked every 23 hours and 56 minutes, and he eventually realized he had become the first human to hear the center of the Milky Way.
On 28 May 2009, Google demoed a product called Wave on stage at I/O for 80 minutes and got a standing ovation from developers who had no idea what they had just watched, and 15 months later the company quietly shut it down because almost nobody could explain to a friend what it was actually for
When Clair Patterson set out in 1948 to measure the age of the Earth using lead in meteorites, his samples kept coming back contaminated, and the seven-year detour he took to find the source ended with him almost single-handedly forcing leaded gasoline out of American cars by 1986.
The IBM 305 RAMAC stayed in production until 1961, weighed more than a ton, stored five million characters on fifty spinning platters, and still drew customers because the alternative was a room full of punched cards
In 1977, Ann Druyan recorded an hour of her brainwaves and heartbeat two days after she and Carl Sagan agreed to marry, and NASA pressed the compressed minute onto Voyager’s Golden Record as a private love signal now more than 25 billion kilometres from Earth
Every Apollo guidance computer that flew to the Moon had its software literally woven by hand at a Raytheon factory outside Boston, where women threaded copper wire through tiny magnetic cores to encode each bit as either a one or a zero, a process the engineers nicknamed LOL memory for Little Old Lady.
When Soviet pilot Viktor Belenko landed a top-secret MiG-25 at a Japanese airport in September 1976, American engineers tearing it down expected to find titanium and microchips, and instead they found vacuum tubes, rivets popped by hand, and a stainless steel airframe so heavy it could only fly fast in a straight line.
In October 2002, a Russian scientist named Dimitri Malashenkov stood up at a space conference in Houston and quietly explained that the dog Laika, whom the Soviet Union had publicly mourned as a heroic week-long orbiter in 1957, had actually died of heat and panic within about five hours of launch.