Why We See Colors — And What We Might Be Missing

The Physics: What Color Actually Is

Color begins not in the eye, but in the nature of light itself. Visible light is electromagnetic radiation with wavelengths between roughly 380 nm (violet) to 700 nm (red). But this "visible" range is a tiny sliver of the full electromagnetic spectrum, which extends from gamma rays (picometer wavelengths) to radio waves (kilometers long).

There is nothing physically special about visible light compared to, say, infrared or ultraviolet. The reason we call it "visible" is entirely biological — it's the range our eyes evolved to detect. The Sun's peak emission roughly coincides with this window, and water and atmosphere are relatively transparent to it, so natural selection homed in on these wavelengths.

A crucial point: a photon of 550 nm light has no inherent "greenness." It has a wavelength and an energy. Color is what happens when that photon meets a nervous system.

---

The Biology: How We Construct Color

Step 1: Photoreception

The human retina contains two types of photoreceptor cells:

• Rods (~120 million) — sensitive to dim light, no color discrimination
• Cones (~6 million) — responsible for color vision, requiring brighter light

We have three types of cones, each containing a different photopsin protein that is maximally sensitive to a different range of wavelengths:

These sensitivity curves broadly overlap. A photon of 500 nm doesn't exclusively activate one cone type — it stimulates all three to different degrees.

Step 2: Neural Encoding

The brain doesn't receive "wavelength data." It receives a ratio of activation across the three cone types. This is processed through opponent channels:

• Red vs. Green (L − M)
• Blue vs. Yellow (S − (L+M))
• Light vs. Dark (L + M, roughly luminance)

This opponent-process theory (Hering, later confirmed by Hurvich & Jameson) explains why we can perceive "yellowish-red" (orange) but never "reddish-green" — those sit on the same opponent axis.

Step 3: Perception

What we experience as "color" is the brain's constructed interpretation of these signals, shaped by context, memory, lighting expectations, and surrounding colors. This is why:

• A grey patch can look blue or yellow depending on surrounding colors
• The same dress can appear blue/black or white/gold
• Color constancy allows you to recognize a red apple under fluorescent light, sunlight, or candlelight, even though the wavelength distribution reaching your eye differs dramatically

Color is not detected. It is constructed.

---

The Limits of Trichromacy

Our three cone types define a three-dimensional color space. Every color you have ever seen or could ever see is a point in this space — a weighted combination of three signals. This means:

Metamerism

Two physically different light spectra can produce the exact same color experience if they happen to stimulate your three cone types in the same ratio. Your computer screen exploits this — it produces only red, green, and blue light, but you perceive millions of colors because the correct mixtures fool your trichromatic system. A spectrometer would instantly distinguish screen-yellow from sunlight-yellow; your eyes cannot.

The Missing Dimensions

A full spectral power distribution is essentially an infinite-dimensional signal (intensity at every wavelength). Our visual system collapses this into just three numbers. We are losing a staggering amount of information. Think of it this way: we're projecting an infinite-dimensional space onto a 3D subspace, and everything that gets projected to the same point is perceptually identical to us.

---

How Other Animals See Differently

Birds (Tetrachromacy)

Most birds have four cone types, including one sensitive to ultraviolet (~300–400 nm). Their color space is four-dimensional. This means:

• They can see UV patterns on flowers, feathers, and fruits that are invisible to us
• Two objects that appear identical to a human (metamers) may look completely different to a bird, because the fourth cone breaks the tie
• Their perceptual color space contains combinations that literally have no analogue in human experience

Mantis Shrimp

Mantis shrimp have 16 types of photoreceptors (12 for color), spanning from deep UV to far red. However, research by Thoen et al. (2014) revealed something surprising: despite this hardware, mantis shrimp are actually worse at discriminating between similar colors than humans. Their system appears to work more like a barcode scanner — binning wavelengths into categories rather than computing fine differences. More channels ≠ richer color experience, necessarily.

Pit Vipers

Pit vipers detect infrared radiation (~5,000–30,000 nm) via specialized pit organs, and this information is integrated into their visual processing in the optic tectum. They may literally "see" heat as a component of their visual scene — a dimension of perception we have no access to.

Butterflies

Some species of swallowtail butterfly have 15 classes of photoreceptors with demonstrated behavioral evidence that they use them for fine spectral discrimination — possibly achieving something closer to genuine higher-dimensional color vision than mantis shrimp.

---

Could There Be Colors We Can Never See?

This question has layers, and they become increasingly philosophical.

Level 1: Wavelengths we don't detect — Definitely yes

Ultraviolet, infrared, microwave, X-ray — these are all "light." We simply lack the receptors. Technologically, we can translate them into visible representations (false-color images of galaxies, thermal cameras), but we never experience them as native visual qualia. A UV-seeing bird doesn't see "dark violet, but more so." It sees something we have no word for, because we have no experience to attach a word to.

Level 2: Higher-dimensional color spaces — Definitely yes (for other species)

A tetrachromat's color space contains entire categories of distinction that are invisible to us. Imagine being unable to distinguish red from green (as in some forms of color blindness) — the tetrachromat's view of us is analogous. There are flowers that look uniformly yellow to us but have bold UV "bullseye" patterns to a bee. The colors in those patterns are not in our gamut. We don't just "miss" them — we lack the dimensional axis to even represent them internally.

There are, in fact, rare human tetrachromats — typically women who carry variant L-cone genes on both X chromosomes, giving them a fourth cone class. Research by Gabriele Jordan and others has identified individuals who demonstrably perform better on color discrimination tasks, suggesting they may access a richer color space. But even they remain limited to a narrow wavelength window.

Level 3: "Impossible" colors within our own system — Sort of

Through opponent-process tricks, some researchers have reported inducing experiences of "forbidden" colors like reddish-green or yellowish-blue using carefully controlled visual stimulation (Crane & Piantanida, 1983). Whether these represent genuinely novel qualia or artifacts of neural fatigue is debated, but they point to the idea that even within our existing hardware, the space of possible experience may not be fully explored in ordinary life.

Level 4: Colors that no biology could detect — This gets deep

Here we cross from science into philosophy.

---

The Philosophical Dimension

Is color in the world or in the mind?

This is the core of the qualia problem. Physicists describe photons by wavelength, frequency, and energy. None of these properties are "red." The redness of red — the subjective, felt quality — is something added by consciousness. This immediately raises the question:

If color is a construction of minds, then does it make sense to ask about colors that no mind constructs?

Functionalism would say: color is just a label for a functional role in information processing. Any system that discriminates electromagnetic frequencies is "seeing color" in the relevant sense. Under this view, there are as many potential "colors" as there are possible discriminable states in possible nervous systems — and the space is enormous, but it's tied to actual or possible perceivers.

Property dualism / qualia realism allows for a more radical possibility: perhaps there is a vast space of possible phenomenal color experiences — a "palette" of qualia — and biological evolution has only sampled a tiny region. Under this view, there could be colors that no actually evolved organism has ever seen, but that are nonetheless experienceable in principle by some possible mind.

Mary's Room (Frank Jackson, 1982)

The famous thought experiment: Mary is a brilliant color scientist who has lived her entire life in a black-and-white room. She knows every physical fact about color vision. When she finally walks outside and sees red for the first time — does she learn something new?

If yes, then there are facts about color (experiential facts) that aren't captured by physical description. And if that's true, the space of possible color experiences might not be fully characterized by physics — leaving open the question of what other experiences are "out there."

Could there be colors no conscious being could ever experience?

This depends on your metaphysics:

• If you're a physicalist: Colors are identical to brain states or functional states. A "color" that no possible brain state could instantiate is incoherent — it's like asking about a square circle. The concept simply doesn't extend beyond possible perceivers.

• If you're a qualia realist: There may be a vast phenomenal space, and our biology accesses only a tiny region. Other species access other regions. But perhaps some regions are inaccessible to any physically realizable system. Whether such "colors" exist is undecidable — by definition, nothing could ever verify them.

• If you're an eliminativist: Colors don't really exist even for us. There are only wavelengths and neural responses. The question dissolves.

---

Synthesis

---

Final Thought

What we call "color" is an extraordinary act of biological interpretation — a way evolution found to compress a complex physical signal into a useful, navigable experience. We live inside a three-dimensional shadow of an infinite-dimensional reality. Other animals live in different shadows — some wider, some narrower, some shifted into regions we can't access.

The most remarkable thing may be this: the universe is awash in electromagnetic radiation of every frequency, and none of it is colored. Color exists only in the small, warm space between a photon being absorbed and a mind making meaning from the event. We don't see the world as it is. We see a story our brains tell, in the only palette they know.

And we have no reason to believe that palette is anything but a tiny, arbitrary fragment of what experience could be.