Colors as you experience them are constructed entirely by your brain, but they are not hallucinations. They track real physical differences in the world: different wavelengths of light, different surface properties of objects. The honest answer is that both halves of the question are true at once. Electromagnetic radiation at various wavelengths is physically real, and the vivid experience of “red” or “blue” is something your nervous system generates in response to that radiation. What makes the topic genuinely interesting is everything that happens between the light entering your eye and the experience arriving in your awareness.
What Light Actually Carries
Sunlight and other white light sources emit electromagnetic radiation across a broad band of wavelengths, roughly 380 to 700 nanometers for the portion humans can detect. A strawberry absorbs most of those wavelengths and reflects a narrow band centered around 650 nanometers. A leaf reflects wavelengths near 550 nanometers. Those are measurable, objective physical facts, and they exist whether any eye is watching or not. In that strict physical sense, “color” is out there in the world, encoded in the spectral composition of the light bouncing off surfaces.
But here is where it gets tricky. The light itself does not carry redness or greenness. A photon at 650 nanometers has a wavelength and energy. It does not have a hue. The redness is something that happens later, after the photon hits your retina. This distinction might sound like a philosophical quibble, but it has real consequences: two physically different mixtures of wavelengths can look identical to you, and two identical wavelengths can look different depending on what surrounds them. The physical signal and the perceptual experience are related but far from identical.
Some colors you routinely see have no single wavelength at all. Magenta, for instance, is not on the visible spectrum. It is your brain’s interpretation of simultaneously receiving signals from both the long-wavelength (red) and short-wavelength (blue) ends of the spectrum with nothing in between. If someone asks “what wavelength is magenta?” the answer is that there isn’t one. It exists only as a neural construction.
How Your Brain Builds Color
The process starts with three types of cone photoreceptors in your retina, each sensitive to a different range of wavelengths. Your brain computes color by comparing the relative excitations across these three cone types, not by reading wavelength directly like a spectrometer.1PubMed. Spectral sensitivity of human cone photoreceptors The cones sensitive to longer wavelengths respond most strongly to light we call red and yellow. The medium-wavelength cones peak in the green range. The short-wavelength cones handle violet and blue. Every color you have ever seen is some ratio of activity across these three channels.
The cone signals are then processed through an opponent system, where neurons essentially pit signals against each other: red versus green, blue versus yellow, light versus dark. This recoding step is why you never see “reddish green” or “yellowish blue.” Those combinations are neurally impossible, because the channels that encode them are set up as opponents.
From the retina, signals travel to the visual cortex, where area V4 plays a particularly important role. V4 sits in the ventral visual pathway and is crucial for recognizing objects by their visual features, including color.2PubMed Central. Toward a unified theory of visual area V4 Damage to this area can leave a person unable to see color at all, even though their eyes and retinal cones are working fine. Conversely, electrically stimulating color-selective parts of the visual cortex can produce vivid color experiences with no light present at all. In one case, stimulating a specific electrode contact pair in a patient’s visual cortex produced a chromatic phosphene (a colored flash) in part of her visual field, while stimulating an adjacent pair produced only an achromatic one.3Clinical Neurophysiology. Color perception matches selectivity in human early visual cortex Earlier work on cortical stimulation in blind patients found that chromatic effects from electrode stimulation varied unpredictably from spot to spot on the same patient’s cortex.4PubMed Central. Phosphenes produced by electrical stimulation of human occipital cortex, and their application to the development of a prosthesis for the blind The takeaway is striking: color lives in the brain, and if you activate the right neurons, you see color whether light is involved or not.
Your Brain Corrects for Changing Light
One of the most impressive things your visual system does is color constancy, the ability to perceive an object as roughly the same color under wildly different lighting conditions. A white shirt looks white to you under bluish fluorescent office light, warm incandescent light, and reddish sunset light, even though the actual wavelengths reaching your eyes are very different in each case. Your brain automatically adjusts for the color of the illumination and recovers something closer to the stable reflectance properties of the surface.
Research into how this works has shown that a visual system with three cone types can recover stable color descriptions of surfaces when the illumination changes, provided it can view a few different surfaces at once. The changing illumination that creates the problem of color constancy also provides the information the visual system needs to solve it.5Journal of the Optical Society of America A. Color constancy: surface color from changing illumination In other words, your brain is not passively recording light. It is actively inferring what the surfaces around you are made of, using the illumination itself as a cue.
Color constancy is so robust that it usually goes unnoticed, which makes it easy to believe that colors are simply “out there.” But the mechanism reveals something important: what you perceive is not raw wavelength data. It is an inference, a best guess about the world computed by neural circuitry that evolved to be useful rather than literal. And sometimes the inference is wrong, which is why optical illusions involving color can be so disorienting. A gray square on a checkerboard can look white or dark depending on the shadow cues around it, because your brain insists on correcting for an illumination change that may or may not exist.
Why We Evolved to See Color at All
If color is a construction, it is a remarkably useful one. The leading explanation for why primates like us have three types of cones, rather than the two found in most other mammals, centers on food. Efficient detection and selection of ripe, reddish fruits against a background of green leaves has long been thought to be a major selective pressure favoring the evolution of trichromatic color vision in primates.6PubMed. Food search through the eyes of a monkey: a functional substitution approach for assessing the ecology of primate color vision Research on South American monkeys has confirmed that the spectral positioning of their cone pigments is well-matched to the task of spotting fruit against foliage.7PubMed Central. Fruits, foliage and the evolution of primate colour vision
This framing shifts the question. Color perception did not evolve to show us “the truth” about electromagnetic radiation. It evolved to help us find food, assess the ripeness of fruit, and detect subtle changes in foliage. Your experience of red and green is tuned to differences that mattered for survival in a forest canopy millions of years ago. If our ancestors had been deep-sea creatures instead, our visual system would be tuned to entirely different wavelengths, and the rainbow as you know it would not exist for us.
Not Everyone Sees the Same Palette
Even among humans, color experience varies more than most people realize. Roughly 8% of men and 0.5% of women have some form of color vision deficiency, typically caused by a genetic change in one of the cone pigments on the X chromosome. These individuals are not seeing the world “wrong” in any absolute sense. They are seeing it with a different set of filters, which compresses the distinctions in certain parts of the spectrum.
At the other end of the spectrum is the possibility of tetrachromacy, where a person has four distinct cone types instead of three. Because cone pigment genes sit on the X chromosome, women who carry one normal and one variant copy of the red or green pigment gene end up with four spectrally distinct photopigments in their retinas. This gives them the potential for enhanced color vision.8PubMed. The molecular basis of variation in human color vision The question is whether their brains actually use that extra channel. Testing women heterozygous for color deficiency genes, researchers found that many showed no evidence of functional tetrachromacy. However, eight carriers of anomalous trichromacy refused to accept color matches that trichromats found perfectly acceptable, suggesting an extra dimension of discrimination.9Vision Research. A study of women heterozygous for colour deficiencies
Whether or not full-blown tetrachromacy is common, the point is philosophically important: two people can look at the same surface under the same light and have subtly different experiences. There is no single “correct” human color perception. There is a range, shaped by genetics, and the variation is wide enough that some people literally see distinctions others cannot.
How Other Animals Experience Color
Step outside the human species and the landscape changes dramatically. Many birds have four cone types plus colored oil droplets in their retinas that further sharpen the distinctions between channels. This tetrachromatic system, which extends into the ultraviolet, enhances their ability to discriminate the colors of plumage and other biologically relevant signals.10PubMed. Tetrachromacy, oil droplets and bird plumage colours A bird looking at another bird’s feathers is seeing patterns and contrasts that are simply invisible to us. The feathers have not changed, but the perceptual world they inhabit is richer.
Mantis shrimp are the extreme case. They have 16 types of photoreceptor, which led to the popular claim that they see a dazzling spectrum far beyond anything humans can imagine. The reality is more surprising: behavioral testing revealed that mantis shrimp are actually poor at discriminating between similar wavelengths. Instead of using a fine-grained opponent-processing system like ours, they appear to use a recognition-based system. They can quickly categorize a color into a bin but cannot tell apart two colors within the same bin.11PubMed. A different form of color vision in mantis shrimp More receptors did not mean richer experience. The mantis shrimp traded sensitivity for speed, probably because rapid color identification matters more than subtle shade discrimination when you are a predator striking in milliseconds.
These comparisons undercut the idea that color perception is a window onto physical reality. Each species has a visual system tuned to its ecological needs, and each one builds a different perceptual world from the same underlying physics.
Colors That Exist Only in the Brain
Several phenomena demonstrate that your brain can generate color experiences with no corresponding wavelength of light present. Benham’s top is a classic example: a spinning disc painted with only black and white arcs fuses into concentric rings of color when rotated. Research has shown that these subjective colors emerge when at least two adjacent flickering regions stimulate the retina, with the temporal separation between light pulses determining which color is perceived.12Vision Research. On Fechner-Benham subjective colour Brain imaging during the illusion found that seeing these phantom colors was linked to enhanced feedback signaling from higher visual area V4 down to early visual areas V1 and V2, suggesting the brain’s color-processing machinery was actively generating color from achromatic input.13Cerebral Cortex. Neural Correlates and Effective Connectivity of Subjective Colors during the Benham’s Top Illusion: A Functional MRI Study
Synesthesia offers another window. People with grapheme-color synesthesia experience vivid, automatic color sensations when viewing letters or numbers. The letter A might always appear red, the number 5 always green, and the associations are stable across years. Brain imaging studies have found that synesthetes show larger responses in color-selective area V4 when viewing graphemes than non-synesthetes do, and the strength of the behavioral effect correlates with the strength of the brain response in early visual areas.14PubMed. Individual differences among grapheme-color synesthetes: brain-behavior correlations Magnetoencephalography work has suggested that activation in the grapheme-processing region and color area V4 occurs nearly simultaneously, within about 5 milliseconds of each other, which supports the idea of direct cross-activation between neighboring brain regions.15PubMed. Magnetoencephalography reveals early activation of V4 in grapheme-color synesthesia
For a synesthete, the colors are real experiences. They are not imagining red; they are seeing it, in the same brain regions that process color from actual light. The fact that the trigger is a shape rather than a wavelength only makes the case stronger that color, as an experience, is a product of brain activity rather than a property of the external world.
Does Language Change What You See?
Cultures divide the color spectrum differently. Russian has separate basic terms for light blue (“goluboy”) and dark blue (“siniy”), treating them as distinct colors the way English treats red and orange. The Dani people of Papua New Guinea historically used only two basic color terms. This observation has fueled a long debate about whether the language you speak shapes how you actually perceive color, a hypothesis associated with the linguists Edward Sapir and Benjamin Whorf.
The evidence points to a real but modest effect. Studies comparing color discrimination across language groups find that people are faster and more accurate at distinguishing two colors when those colors fall on opposite sides of a boundary in their native language.16PLOS ONE. The Sapir-Whorf Hypothesis and Probabilistic Inference: Evidence from the Domain of Color A blue-green pair that straddles the English “blue”/”green” boundary is easier for English speakers to tell apart than a pair that sits entirely within either category, even when the physical distance between the pairs is identical. The effect shows up primarily in reaction time rather than whether someone can see the difference at all, but it is consistent and replicable.
At the same time, there is strong evidence that color naming is not entirely arbitrary across cultures. Analyses of color-naming systems across many languages have found recurrent patterns and a relatively fixed order in which cultures begin to use new color terms: terms for light and dark come first, then red, then green and yellow, then blue, and so on.17PubMed Central. On the origin of the hierarchy of color names This suggests that the way we categorize color is partly constrained by the biology of our visual system and partly shaped by language and culture. Language does not create color perception from scratch, but it nudges where you draw the boundaries between categories, and those boundaries affect how quickly you process differences.
Memory, Expectation, and the Colors You Think You See
Your brain does not build color from scratch with every glance. It brings expectations to the table, drawn from a lifetime of experience with how objects typically look. A banana in a grayscale photograph can appear faintly yellow to many observers because the brain’s stored knowledge of banana-color bleeds into perception. This memory color effect is strongest for objects with a consistent, diagnostic color, like bananas, fire trucks, and grass. When those objects appear in an unexpected color, they pop out and grab attention more readily than neutral objects do, as though the brain’s mismatch detector is flagging a violation of its predictions.
This effect reveals something about the fundamental architecture of color perception. Your visual system is not simply a camera that faithfully records incoming light. It is a prediction machine that combines current sensory data with prior knowledge, and color is one of the channels where that combination happens. The yellow you see when you look at a banana is partly wavelength-driven and partly memory-driven. Where one ends and the other begins is not always clear, even to the brain doing the work.
The Philosopher’s Puzzle
Philosophers have a thought experiment that gets at the heart of why this question feels unresolved. Imagine a scientist named Mary who has spent her entire life in a black-and-white room. She has access to every physical fact about color: wavelengths, cone responses, neural pathways, everything described in this article and more. She knows, in complete physical detail, what happens when someone sees red. Then one day she walks out and sees a red rose for the first time. Does she learn something new?18Philosophia. Acquaintance and Mary’s Revelation: A Response to Alter
Most people’s intuition is yes, she learns what red looks like, and that knowledge is something no amount of physical description could have given her. If that intuition is correct, then the subjective experience of color is something over and above the physical facts. The experience of redness is real, but it is not captured by the wavelength, the reflectance spectrum, or the firing pattern of neurons, even though all of those things are necessary for it to happen. Physicalists push back: Mary might be surprised, but surprise is not the same as new information. The debate has been running for over four decades and shows no sign of settling.
What is interesting about Mary’s Room from the perspective of this article is that both sides agree on the science. The physics of light, the biology of cone cells, the neural circuitry in V4 are not in dispute. The disagreement is about whether those facts are the whole story, or whether the subjective experience of seeing red is an additional fact about the universe that physics cannot fully capture. That gap, between the measurable and the felt, is why the question “do colors exist?” does not have a clean yes-or-no answer. The wavelengths are real. The neural machinery is real. Whether the redness you experience when those systems do their work is itself a feature of the physical world or something your consciousness adds on top is still, genuinely, an open question.
Structural Color and the Tricks Surfaces Play
Not all color in nature comes from pigments absorbing and reflecting certain wavelengths. Structural coloration produces color through microscopic physical structures that are fine enough to interfere with visible light. A butterfly wing, a peacock feather, or an oil slick can produce vivid, shifting colors not because of any dye or pigment, but because of nanoscale layering or lattice patterns that selectively reflect certain wavelengths through interference. Tilt a peacock feather and the color changes, because the angle of viewing changes which wavelengths constructively interfere.
Structural color matters for this discussion because it highlights a case where the “color” of an object is not even a fixed property of its surface. The same feather reflects different wavelengths to different observers at different angles, at the same moment, under the same light. A blue morpho butterfly is intensely blue from above and dull brown from below. Which is the “real” color? Neither, of course. The structure is real. The light is real. The blue is an event that happens at a particular geometry between surface, light source, and eye. That framing applies broadly: all color is an event, a meeting point of physics and perception, rather than a static label stamped on objects.