It was just a colour out of space—a frightful messenger from unformed realms of infinity beyond all Nature as we know it; from realms whose mere existence stuns the brain and numbs us with the black extra-cosmic gulfs it throws open before our frenzied eyes.

Science-fiction author H. P. Lovecraft considered The Colour Out of Space his best story. In this 1927 classic tale of cosmic horror, a small Massachusetts farming community faces unspeakable evil from the outer reaches of the universe. The extraterrestrial villain is not a face-hugging or chest-bursting alien but something far more terrifying: a weird color.

Slowly but surely the otherworldly color mutates and destroys crops, insects, wild animals and livestock. It impregnates the land and the water. The unfortunate farmers who encounter the bizarre hue fall prey to insanity and untimely death.

And you thought vision research was for wimps.

This article features some of the most spectacular color phenomena this side of the galaxy. You won't see any extraterrestrials, but many strange illusions arise from taking colors out of place and putting them in an unusual context. Use caution: the peculiar shades and tints you are about to experience could blow your mind.

Here we have two moons out of space. One yellow and one blue. Or are they? Actually both moons are exactly the same color in this illusion by psychologist Akiyoshi Kitaoka of Ritsumeikan University in Japan; only the surrounding colors are different. If you don't believe it, cut out the two moons—you'll find them to be identical. The appearance of colors is all about their context.

Legend has it that Rome was founded by warring twin brothers, Romulus and Remus, born to a vestal virgin named Rhea Silvia and fathered by Mars, the god of war. Vestal virgins, as it turns out, are not supposed to conceive children, even if the father is a god. The family shame was too much for Rhea's father, who killed her and then condemned the twin baby boys to die of exposure. The wolf Lupa found the boys and adopted them. But hey, what about Lupa's biological pups, Rex and Fido, younger brothers to the feral Romans? These nonidentical twins (left) become identical when the background is removed (right). Had this pair been born before their mother discovered Romulus and Remus, surely Rome would have gone to the dogs.

Rubik's Cube is a three-dimensional puzzle in which the player rotates the tiled faces of a cube until each face shows the same color on all nine tiles. Sound easy? Only if the lighting conditions are stable. As this illusion by Beau Lotto and Dale Purves of Duke University shows, if the lighting changes, it can be hard to know which color is which. The masked version of the illusion (above, right) reveals that the blue squares on the left and the yellow squares on the right are actually all gray when viewed under white light. Color perception is not based strictly on the wavelengths of the light that strikes your retina; instead the brain assigns colors based on the lighting conditions and uses the wavelengths only as a guideline to determine which objects are redder or bluer than other objects in the same scene.

It looks like this Japanese manga girl has one blue eye and one gray eye. In fact, both eyes are exactly the same shade of gray. The girl's right eye only looks the same as the turquoise hair clip because of the reddish context. Part of the process of seeing color is that three different kinds of photoreceptors in the eye are tuned to three overlapping families of color: red, green and blue (which are activated by visible light of long, medium and short wavelengths). These signals are then instantaneously compared with signals from nearby regions in the same scene. As the signals are passed along to higher and higher processing centers in the brain, they continue to be compared with larger and larger swaths of the surrounding scene. This “opponent process,” as scientists call it, means that color and brightness are always relative.

This image by Kitaoka contains a number of blue-green circular structures. The red rings are purely a creation of your brain.

A process called color constancy makes an object look the same under different lighting conditions, even though the color of the light reflecting from the object is physically different. Color constancy is an incredibly important process that allows us to recognize objects, friends and family both in the firelight of the cave and in the bright sun of the savanna.

Because the rings here are drawn in shades of blue, the brain mistakenly assumes that the image is illuminated by blue light and that the physically gray rings inside the blue structures must therefore be reddish. The visual system subtracts the blue “ambient lighting” from the gray rings, and gray minus blue results in a pastel red color.

Here is another example of how the brain determines color depending on the context. In the bull's-eye structures in the left checkerboard, the center rings look either green or blue, but they are all the same color (turquoise). The center rings in the right checkerboard are all the same shade of yellow. Unlike the previous images, this type of color illusion is difficult to explain by an opponent process because the apparent color of the rings is more similar than dissimilar to the background.

All the hearts in this checkerboard are made out of the same cyan-colored dots, but they look green against the green background and blue against the blue background. The image, by Kitaoka, is based on the dungeon illusion discovered by vision scientist Paola Bressan of the University of Padua in Italy.

We have seen that the same colors can look different from each other, depending on context. This illusion shows that context can also make different colors look similar. Check out the red tiles on the top face of the left and right Rubik's Cubes. They look more or less like the same color. If we mask the rest of the tiles with white (above, right), you can see that the tiles on the left cube are actually orange and that the tiles on the right are purple.

We see four differently colored squares on a gray background, right? Wrong. The gray is actually a mixture of little blue and yellow pixels. Because the pixels are so small, they blend together and do not activate the opponent processes that would create contrast. This is how a color television creates different colors from just a few differently colored pixels (hold a magnifying glass to your TV and prove it to yourself). The turquoise and chartreuse squares are actually little green pixels mixing with either the blue background pixels (turquoise) or the yellow background pixels (chartreuse). Mixing red pixels with either the yellow or blue pixels in the background creates the orange and purple.

In 1979 Michael White of the Tasmanian College of Advanced Education described an illusion that changed everything in visual science. The gray bars on the left look brighter than the gray bars on the right. In fact, all the gray bars are physically identical. Before White discovered this effect, all brightness illusions were thought to result from opponent processes—that is, a gray object should look dark when surrounded by light and light when surrounded by dark. But in this illusion the lighter-looking gray bars are surrounded by white stimuli, and the darker-looking gray bars are surrounded by black. The brain mechanisms underlying White's effect remain unknown.

In Light of Sapphires, the blue dots appear to scintillate as you move your eyes around the image. But when you focus on one dot, the scintillation stops. The blue color appears more saturated for the dot in focus than for dots in the visual periphery. This effect is a colorful variant of the scintillating grid illusion discovered in 1994 by Elke Lingelbach of the Institute for Optometry Aalen in Germany and her colleagues Michael Schrauf, Bernd Lingelbach and Eugene Wist.

White's effect also affects the way colors look. The logo for the Best Illusion of the Year Contest is a combination of White's effect (the vase appears to be different colors behind the two curtains) and the famous face-vase illusion (in which the “vase” is a trophy for the winner). See more color combinations at

These spirals, created by Kitaoka, are particularly strong examples of White's effect as applied to color. The green and cream-colored spirals (bottom) are made from stripes that are physically yellow. In the other two examples above, the stripes are physically red and cyan, rather than purple, orange, blue and green.

The colors from the small crosses appear to spread onto the white expanse surrounding each intersection. The effect resembles the glare from a neon light. This illusion was reported in 1971 by Dario Varin of the University of Milan in Italy and a few years later by Harrie van Tuijl of the University of Nijmegen in the Netherlands. Its neural causes are currently unknown.

Here the neon color spreading produces a rectilinear grid of north-south and east-west streets on the map—but only in the periphery of your vision. It is absent from whichever intersection you happen to be staring at.

In this neon color spreading illusion, the yellow spreads in a direction that is perpendicular to the black bars.

In this illusion by Italian vision scientist Baingio Pinna, a thin, orange contour adjacent to a darker purple contour casts an orange tint over long distances—as though a watery paint was filling in the gaps between the orange lines [see “Illusory Color and the Brain,” by John S. Werner, Baingio Pinna and Lothar Spillmann; SCIENTIFIC AMERICAN, March 2007]. On the opposite side of the purple contour, the outlined areas look white.

In this image by Pinna, the inner square appears to have purple smog around the dots, and the outer square appears to be filled with blue. The illusion is caused by the watercolor effect.

The watercolor effect inspired the wave-line illusion by Japanese vision scientist Seiyu Sohmiya. In this version by Kitaoka, the white background behind the pattern is tinged by the color of the waves.

The red color behind the blue lines appears to be magenta, whereas the same red color behind the yellow lines appears to be orange. This “color assimilation” illusion shows that colors can blend with each other in some situations, rather than contrasting with each other.

During his blue period, Pablo Picasso painted everything—including shadows and gradations of sunlight—in shades of blue (left). How do we recognize the people if they are all the wrong color? Margaret S. Livingstone of Harvard Medical School has shown that although Picasso used blue, he was careful to maintain the luminance relations—contrasts in lighting within the scene [see “Art, Illusion and the Visual System,” by Margaret S. Livingstone; SCIENTIFIC AMERICAN, January 1988]. Those luminance relations, which we use to make sense of the image, are apparent in a grayscale version of the painting (right). This is why color-blind people see just fine in almost every way—sometimes they do not even know they have a deficit.

Here Livingstone and her Harvard colleague David H. Hubel took an Escher woodblock, Tower of Babel (left), and colored the white spaces light blue (center). You still see the tower, because the luminance relations remain intact. But when the black spaces are replaced by a green shade with the same luminance as the blue (previously white) spaces, the 3-D character of the image falls apart (right). Our visual system cannot perceive volume, form and distance with only color information available. Luminance information is also required.

A group of 20th-century European artists led by Henri Matisse and André Derain used such vivid, unusual colors in their paintings that one critic dubbed these works les Fauves (“wild beasts”). This style became known as Fauvism. Derain's 1905 portrait of Matisse (left) is characteristic of this style. Using a grayscale version (right) of a similar painting, Livingstone showed that the weird colors work because they have the correct luminance.

This painting by Picasso shows that coloring within the lines is unnecessary. Our brain assigns the colors to the correct shapes even though the shapes are depicted minimally with sparsely drawn lines.

Here is a great cognitive visual illusion that involves a conflict between the syntactic and symbolic processing systems in your brain. Look at the words one after the other without stopping or slowing, but instead of reading each word, just say its color out loud. It's hard, isn't it? You are experiencing the Stroop effect, named after psychologist John Ridley Stroop. Even if you try not to read the words, you cannot keep the content of the words from conflicting with their color.

Japanese ophthalmologist Shinobu Ishihara developed 38 color plates, including the two above, to test patients for color blindness. Each plate is a circle filled with dots of different sizes and colors. People with the most common types of color blindness find it difficult or impossible to see the numbers hidden within the patterns shown here.

Discovered by vision researcher Celeste McCollough, this illusion demonstrates that the interactions between color perception and form perception can be surprisingly long-lasting. The effect takes discipline, though, so suck it up before you try it, soldier! We can't make you do it, of course, but it won't work correctly unless you do, and we promise it will be worth the effort.

Look at the black dot at the center of the vertical magenta-striped grating for one full minute. (One minute will seem like forever, but trust us on this.) Then fixate on the dot in the horizontal green grating for one minute. Then shift back to the vertical magenta grating and then back to the green, for one minute apiece. Repeat another cycle. Okay, now you're ready. After six minutes of alternating between the two gratings, look back and forth between the uncolored patched gratings below. You will see that the horizontal patches are tinged magenta and that the vertical patches are tinged green.

This illusion shows that adaptation, the process by which neurons in the brain become less responsive to unchanging stimuli, can be simultaneously selective for both color and orientation of edges. That is, you have neurons that are attuned to both magenta and vertical orientations, and when you stared at the vertical magenta grating for minutes on end, that allowed the horizontal-detecting neurons that are sensitive to magenta to seem more responsive. So when you are presented with a horizontal white grating after the adaptation, it looks tinged with magenta. The same is true for the green adaptation, for the opposite orientations.

McCollough's illusion was the first to show that adaptation can last a long time. If you come back in an hour and look at the white gratings, you will still see an effect, albeit weaker.