All the way back in 1801, long before scientists knew about cones and neurons, English physician Thomas Young theorized that the human eye has three types of color receptors: blue, green and red. Young's trichromatic color theory was proven correct in the 1960s, when cones (named for their shape) were discovered to have special sensitivity to blue, green and red light [source: Nassau].
The opponent theory of color perception has been around since the 1870s, when German physiologist Ewald Hering first postulated that our vision was ruled by opponent colors: red versus green and blue versus yellow. Hering's opponent theory is supported by the fact that there are no colors that could be described as reddish-green or yellowish-blue, but every other color in the visible spectrum can be created by combining red or green reflected light with yellow or blue [source: Billock and Tsou].
Both trichromatic color theory and opponent theory were treated as immutable truths of color perception for more than a century. Taken together, the two theories argue that it's impossible for the human eye or mind to perceive certain colors described as red-green or blue-yellow.
Thankfully, there are always a few rogue scientists who like to push the realms of possibility. In the early 1980s, visual scientists Hewitt Crane and Thomas Piantanida designed an experiment with the goal of tricking the brain into seeing impossible colors.
In Crane and Piantanida's experiment, subjects were instructed to stare at an image of two adjacent strips of red and green. The subjects' heads were stabilized with a chin rest and their eye movements were tracked by a camera. With every small twitch of a subject's eyes, the red and green image was automatically adjusted so that the subject's gaze remained fixed on the opposing colors [source: Billock and Tsou].
The results, published in the journal Science in 1983, were mind-blowing. If people stared at adjacent opposing colors long enough, the border between them would dissolve and a new "forbidden" color would emerge. The resulting color was so new that subjects had great difficulty even describing it [source: Wolchover].
By stabilizing the image to track eye movements, Crane and Piantanida theorized that different areas of the eye were being continuously bathed in different wavelengths of light, causing some opponent neurons to get excited and others to be inhibited at the same time.
Oddly, Crane and Piantanida's experiment was dismissed as a parlor trick, and several other vision scientists failed to achieve the same dramatic results. It wasn't until the 21st century that impossible colors were given a second life.