Purple exists only in our brains | Hacker News Reader

Purple exists only in our brains

snexplores.org

43 pointsgeox74 comments

I love the headline, but unfortunately both the headline and the explanation are wrong.

Better explanation:

The red, green and blue receptors in the eyes all respond to all wavelengths of light, they're just more sensitive at red, green and blue respectively.

When violet hits those receptors, it doesn't hit any of the receptors at their peak sensitivity point. All three receptors fire. However, the green receptor has much better sensitivity than the red receptor, so violet triggers the strongest response in blue and red. If the red receptor was more sensitive, we'd probably interpret violet as a shade of blueish-brown rather than of purple.

The headline:

Purple definitely exists. The eye & brain interpret purple and violet the same, but in the physical world they are different colors, and they both exist in the wild. Purple is a mixture of red & blue, violet is the highest visible frequency. A purple flower is a flower that absorbs green and reflects red & blue. A violet flower is a flower that absorbs red, green and blue and reflects violet.

Colors like purple are called "extraspectral" to distinguish them from the spectral colors like violet.

Same with pink.

P.S. Tetrachromats can distinguish between violet and purple, unlike us normal humans.

Not true. Human tetrachromats have an extra kind of receptor somewhere between the blue and red receptors' sensitivity. This doesn't help with colors like violet that are outside of that range.

Also, purple (a non-spectral color) is easily distinguished from violet (a spectral color) if you see them side-by-side.

There exists a shade of purple that is indistinguishable from violet because it triggers the cones of the eyes at the same level that violet does.

You can buy paint called "violet". This isn't the spectral violet, it's a shade of purple that looks very similar to spectral violet.

Tetrochromats can distinguish between that purple shade and real violet. But if you mixed the paint using 4 tints rather than 3 you could fool them too.

Source?

edit: You may be confusing tetrachromacy with people who don't have a lens and can therefore perceive ultraviolet light that's normally filtered out. These folks can see shades of violet where other people don't because the blue receptors are being stimulated by the ultraviolet light.

>All three receptors fire. However, the green receptor has much better sensitivity than the red receptor, so violet triggers the strongest response in blue and red.

I'm confused - it sounds like you're saying it's so off-peak for all of them that, even though it's more proximal to blue, it is effectively equally out-of-range for all three - except that the green receptor is more sensitive. But then that would mean it should look more green, rather than more red and blue, no?

More sensitive is a relative term rather than an absolute term -- the green cone selects for its color better than the red cones do.

Agreed. It's simply philosophically sloppy to suggest that purple, brown, and really any other colour don't exist without qualifying what one means by existing in this context, and why that definition of the word is used.

In this case, it seems the "what" is that for a colour to "exist", it has to be a distinct, isolated frequency of light, and (for some reason) it needs to appear in a logical place in the frequency space in relation to other colours. They don't answer the "why" question, and I think if they tried, they would find it difficult, because it's completely arbitrary and in conflict with reality.

Just because some colours do in fact correspond to frequencies of light, that doesn't mean all colours have to. There's a reason we have the terms primary and secondary colour. Some colours are emergent from mixing other colours together. Does that mean they can't appear in the natural world, being discriminated by sensory systems? No. Does purple cause things to happen? I'm sure you could find myriad examples of this in the natural world. Ironically, the only thing here that only exists in our brains, is the notion that purple doesn't exist.

>I'm sure you could find myriad examples of this in the natural world.

Can you think of any?

yaky (https://news.ycombinator.com/item?id=43584334) gave a great example of hummingbirds liking purple.

Even better, I can tell you one I experienced literally today.

I was at a barbeque, and the people hosting put out one bowl of diced red onion, which despite the name was purple, as it tends to be, and a bowl of crispy fried onion, which was brown. I reached for the fried onion, and avoided the red(purple) onion, because I always hated red onion.

Now, if you want to be annoying, you could make all sorts of red(heh) herring arguments about the cause of me avoiding the red onions being their flavour and aroma in the past having caused me not to like onions, and that's all true. But effects can have multiple causes, and it's also true that the distribution of pigments in the onion leading to an emergent property we dub "purple" did in fact convey information to me that caused me to decide not to eat them. I didn't smell them or taste them. Information was transmitted from onion to mind, via the electromagnetic field. Clearly the emergent property of purple colour has a causal effect here.

As for other examples in the natural world of non-human plants an animals, I don't know, but feel free to use a search engine. But since the colour purple is found in all sorts of life forms(including humans, and all sorts of life forms(including) are able to perceive and discriminate it from other colours, it's fair to assume that the colour purple is out there having all sorts of causal effects.

The purple you saw in those onions is literally a neurological glitch. Your brain inventing a color that doesn’t exist in the spectrum. When red (long) and blue (short) wavelengths hit without green (mid), your visual cortex makes up purple as a placeholder. Your avoidance wasn’t caused by the color itself (a mental construct), but by the brain using this imaginary hue as a proxy for past onion trauma.

This mirrors how we treat UI error messages. A "404" doesn’t cause missing data, it’s just the system’s way of flagging underlying issues. The real causal chain was anthocyanins → wavelength reflection → neural pattern-matching → memory recall. Purple was the middleware, not the root process.

Fun twist. Those fried onions’ brown does have causal ties to flavor. Maillard reaction products directly interact with taste receptors. The universe trolls us with color semantics, but chemistry always wins.

It might not exist in the spectrum, does that mean it doesn't exist? You're arguing and conflating two different things here. On the one hand, you're implicitly arguing that a colour can't exist unless it corresponds to a singular frequency of light, which I've already argued against. This is no more meaningful than arguing that tables and chairs are mental constructs because it's all quarks and electrons at the end of the day. Emergent properties exist and can have causal effects, most philosophers and scientists are in agreement about this.

The other is that a qualia or the mental experience of seeing purple is the same thing as perceiving purple as distinct from other colours in a physical object. I'm not talking about the qualia. In fact, I hate the concept of qualia, because whenever it's introduced into philosophical discussions, the discussion devolves into epicycles of meaningsless discussion of definitions and nomenclature and ends up going nowhere.

No, the purple was there. You say all that was there was some chemicals that only reflects certain wavelengths. I say this is what defines the emergent physical property we call the colour purple. You say electrons and quarks, I say tables and chairs. Both are accurate, and certainly not in conflict.

You might say, so how is this distinct from qualia? Well, for the qualia of seeing purple, there is no way even in principle to decide whether my qualia is the same as your qualia. But I can still look at a red onion and tell you it's purple, and you likely would agree unless you're colour blind. So this property of purple is, unlike a qualia, objective, not subjective.

Your critique reveals a crucial conflation between structural emergence and perceptual categorization, a distinction that clarifies why "purple" (as a color category) lacks the causal efficacy you ascribe to it. If you gift me some of your valuable reading time, let's dissect this.

1. Two Types of Emergence

- Structural emergence (tables/chairs): Arises from physical interactions between components. A table's causal power (holding objects) derives from its atomic structure creating macroscopic rigidity. These properties are observer-independent. A laser would detect the table's structural integrity even with no humans present.

- Perceptual categorization (color): Emerges from evolved neurobiology + cultural reinforcement. The "purple" label applied to red onions is a compression algorithm for "reflects 400-450nm + 600-700nm with minimal 500-600nm". This categorization has no causal power beyond its role as an information tag.

2. The "Objective" Color Fallacy

Your intersubjective agreement about purple stems from:

- Shared cone cell biology: 94% of humans have L/M/S photopsins with peak sensitivities at ~560nm (red), ~530nm (green), ~420nm (blue)

- Cultural conditioning: Modern color lexicons standardized via Pantone systems and CIE charts

Yet this consensus doesn't make purple an emergent physical property.

Consider this.

The Himba tribe uses "zoozu" for dark colors (blue/purple/black) and doesn't distinguish purple as a category

Industrial paint manufacturers recognize 12,000+ color terms, far beyond basic spectral labels

Your "purple" onion would register as #6A1B9A in HEX, 17.3° hue in CIELAB, arbitrary numerical tags, not causal agents

3. Causal Efficacy Lies Elsewhere

The chain you described:

Photons → Retinal Activation → Neural Coding → Avoidance

Contains zero causal nodes requiring "purple" as an explanatory variable. Replace "purple" with "wavelength combo X" and the physics/neurology remains identical. Contrast with a table's causal power. Replace "table" with "carbon lattice configuration Y" and you lose the explanatory utility.

4. The Qualia Dodge

You're right to reject qualia-centric debates, but the alternative isn't reifying color categories. Instead, recognize that:

a) The onion's surface selectively reflects wavelengths

b) Your visual system detects this pattern

c) Your brain applies a culturally-learned label

d) The label activates memory associations

The causal oomph lives in the biochemical aversion pathways, not the color label. Change the label (call it "ploobalooba") while keeping wavelength data and aversion remains. Change the wavelengths while keeping the label, and behavior shifts.

5. The Real Emergent Culprit

What does have causal power here is pattern recognition heuristics. Your brain evolved to:

- Create color categories as survival shortcuts ("red" = blood/danger)

- Link these to outcomes via associative learning

These heuristics are genuine emergent properties with causal effects, but they're neural algorithms, not spectral properties. The purple label is their UI, not their codebase.

TL;DR

You're mistaking the map (color categories) for the territory (wavelength interactions). Tables derive causal power from structural emergence, "purple" derives consensus from neuro-cultural emergence. One explains why plates don't fall through surfaces, the other why we argue about onions at barbecues.

>I don't know, but feel free to use a search engine.

I did. I didn't find any. I don't think there are. You made the assertion, so, I thought you might know some.

Violet in the rainbow is not a purple hue, it is a deep blue. Illustrations often color it more purplish but that is inaccurate

You usually can't see the really purple-y violet in a rainbow because it's quite dark -- it's easily absorbed by the atmosphere and our color cones aren't very sensitive to that wavelength. But it's there if you amplify it.

Exactly this, purple ≠ violet. They don't even look the same.

You won't see violet on a computer screen because it's a higher frequency than what blue LEDs produce. You won't see it on the output of consumer-grade printers for similar reasons regarding the color of the ink.

The easiest way to see actual true violet is to pass sunlight through a prism onto a white surface.

Purple on the other hand is a mixture of red and blue frequencies that stimulate both kinds of receptors in your eyes. It looks like a reddish blue that can't be produced by any one frequency of light.

True violet looks like a deep, deep blue without any red tint.

I think you're misinterpreting color and wavelength. No color exists "in the wild". Colors only exist in the consciousness of beings that can capture photons. Wavelengths do exist in the wild I guess, or at least it's a human description of how electromagnetic waves move in the wild.

Color is a mixture of wavelengths of visible light. Mixtures of wavelengths of visible light exist in the wild.

Color, as we know them, are a human interpretation of wavelengths and one we know is semi-subjective based on things like color blindness. A creature with less sensitive receptors could see one color were we'd see a much wider range. The reverse is also true that a creature with more sensitive receptors could see far more colors than we do.

No, color is a subjective experience a conscious being has. Different colors each correspond to different mixtures of wavelengths, but that doesn't mean they are the same thing. Color is the thing that you see with your consciousness. The difference is fundamental: wavelengths exist independently of the existence of conscious beings, while colors only exist if a conscious being can capture photons and has a certain brain system that can interpret those photon's wavelengths.

Colors don't even require photons. There's a phenomenon called phosphenes that produces colorful lights (green and purple for me). You can see them by rubbing your eyes sufficiently hard, or copy Newton and stick a blunt needle in your eye. They're caused by mechanical deformation in your optical cells rather than any photons being captured, but they produce the experience of color regardless.

You're confusing the qualia of colour("what it's like" to perceive colour) with the property of colour. Both can be referred to by the word "colour".

You're essentially just arguing that one definition of a noun is "more valid" than another, which is completely arbitrary.

And of course the qualia of colour is entirely subjective and only in the mind. That's literally the definition of qualia. That doesn't mean the property of colour is also subjective, just because it's referred to by the same noun. And besides, qualia is a philosophically controversial term in the first place.

If you were trying to cross a river and someone told you walk over that bridge over there, you wouldn't respond that it's impossible because a bridge is a device that links two computer networks together at the link layer, and couldn't possibly be used a cross a river, would you?

you're right, I'm just arguing that when we talk about colors we talk about the subjective perception of it, the thing that we see, which is what is usually referred to as color in any normal conversation. I didn't know there was a different definition of color which was a synonym of wavelength, in that case you're right. So as you say, the debate is arbitrary because it's about different definitions of color.

However, both measurements of wavelengths, the qualia or subjective one and the numerical wavelength or objective one, both exist only in our minds. There exists a measurement if and only if there exists someone that measures, even if it is an objective measure

Purple is a measurement, but wavelength is a measurement too. Ultimately everything is a measurement to us as that is the only way we can experience "the wild".

But wavelengths are a property of the waves that exist in the wild, while color is a property of our brains interpreting those wavelengths. There's a fundamental difference there: one could say that wavelengths exist independently of the existence of conscious beings (even if no one measures them), while colors only exist if a conscious being can capture photons and has a certain brain system that can interpret those photon's wavelengths.

Wavelength measures something in the wild. Whatever that something is, colour (and thus purple) offers a different way to measure the same thing. While the measuring unit differs, the fundamentals are the same.

interesting, so you're saying wavelengths are also a construct of the human mind (handed with a device to measure them). Maybe then the Buddha was right when he said "form is emptiness"? Which would mean there's no way that something really is independent of the way of looking at it, that things are empty, there are only ways of looking...

There's a common abuse of language that the ancestor comment uses, where we also use the names of colors for the wavelength of the pure light that we perceive them as; perhaps they could have been more specific about that. Hence because of this abuse of terminology we have words like infrared and ultraviolet to refer to spectra beyond human color perception, but are so called for their relation to the particular pure wavelengths that we perceive as red and violet.

I think you are misunderstanding the purpose of a headline. "Purple exists only in our brains" is a reasonable summary. Your last paragraph is not a headline.

It's a reasonable summary of a portion of the article. I assert that the portion it summarizes is wrong, making the headline also wrong.

More than purple not existing, brown really doesn't exist. Brown is just reflected orange-y light in the right context. You can play around with it, by looking at a picture of say, earthy soil, color picking that pixel and painting it on a white canvas in a different way. Or by trying to shine an RGB light with that color code. It will look orange in a direct light.

Also the purple hallucination is because the same cone that mostly detects red also has a small peak futher along the spectrum next to blue. Since it slightly reacts to blue/violet light, you get that purple tickle.

Technology connections made a video on it.

https://www.youtube.com/watch?v=wh4aWZRtTwU

Came here looking for this haha, Alec's videos are always a treat.

Thank you, I couldn't remember who made it, thought it was Knowing Better for some reason

According to this [1] that extra bump is a myth. But I'm not qualified to judge.

[1]: https://physics.stackexchange.com/questions/673991/red-cones...

I wonder if near-UV like 380nm starts exciting the red receptor at 760nm like a resonant frequency or harmonic. Would also explain why something that's deeply red enough doesn't start looking purple, because the lower-energy photons can't get the higher energy receptors to trigger, unlike the other way around for deep blue. Which would make the color purple a distinctly quantum phenomenon.

> brown really doesn't exist. Brown is just reflected orange-y light in the right context.

The Pirahã language only has two words for "color" – which more or less translate to "light" and "dark". Therefore red, green, blue, even orange don't really exist either. Except they do, because we've defined them to exist. Brown is no different.

Brown is different, because you can only see it in the right context, otherwise it becomes orange. You cannot create brown light sources.

https://www.youtube.com/watch?v=wh4aWZRtTwU&pp=0gcJCdgAo7VqN...

Not naming certain colours does not mean they cease to exist. They can be split, created, and we have the receptors to perceive them.

> Brown is different, because you can only see it in the right context, otherwise it becomes orange.

But that's how it is defined, so no, it is not different.

Well.. no color “exists”, they are just wave like particles.. not even “reality” as we see it exists.. it’s just in our minds.. I think that what they really mean is that there is no wavelength for the purple we see

Except for colors that are combinations of different, disconnected, wavelengths. This is why there are red and blue stars, but no green or magenta stars.

That's not really the reason. Green is a spectral color, and in fact the sun peaks around the green part of the spectrum. But to see a star as green, it would need to exclusively emit green light.

This is silly. Everything exists “only in our brains”. The ancient Greeks knew this - see Plato’s Allegory of the Cave.

You don't only exist in my brain.

That sounds like something one of Plato's cave dwellers would say.

Are you sure about that?

Are you just a text message? I don't think my brain generates all the text messages in the internet, it isn't that creative.

The article says this (and pretty much all the other comments here).

It doesn't however mention James Clerk Maxwell who took a break from inventing equations to torture EE students forever to figure this stuff out, including making the first color photograph.

(Yes I know Heaviside actually came up with the equations we use)

I understand it more intuitively from a chromaticity diagram, depicting color divorced from brightness: https://en.wikipedia.org/wiki/Purple#/media/File:Line_of_pur...

The curved rim of the horseshoe is the spectral locus; that of the purest, single wavelength spectral colors.

Purples are what you get when you travel between the ends of the horseshoe, where no spectral color lies. This is what is meant when one says purple exists only in our brains; it is non-spectral.

Furthermore, our vision is such that objectively different spectra can look the same; cf. metamers. This follows from the fact that the process of projecting an infinite dimensional spectra into three dimensions (because color is 3D) is lossy. But our vision optimizes the dimension reduction for spectral discrimination in parts of the electromagnetic spectrum that are beneficial for evolution.

The spectral locus is highlighted in this one (for clarity),

https://en.wikipedia.org/wiki/CIE_1931_color_space#Chromatic...

That outer *bolded* edge is monochromatic light, as perceived by humans. The number labels are the wavelength in nanometers.

I forgot to add that in addition to being non-spectral, they are the most saturated colors of their hue. All colors inside the horseshoe are non-spectral; they are mixtures of spectral colors.

https://en.wikipedia.org/wiki/Line_of_purples

Fun fact: such "purples" exist for other species. There was a study that found that hummingbirds are more attracted to a color that is a mix of green and ultraviolet (imperceptible to humans, but similar to our concept of "purple"), than just green, violet, or ultraviolet.

I mean, all colors exist only in our brains. "Color" is a perceptual experience that may start as spectral power and cone excitation, but is then mapped to color sensation through a fantastically complex visual system in the brain. That's why color-based optical illusions work. That's why "The Dress" was controversial.

There's nothing that special about purple. The article makes a big deal about purple "connecting" two opposite ends of the spectrum, but this is just an artifact of humans seeing rainbows and realizing that pink-purple would connect the two ends, even if they're not present in the rainbow. It's just a human framework, not some kind of physical truth.

Put another way, there are a ton of colors that are not in the rainbow, because it's a 1D simplification of our color perception, which is 3D (three cones). Would you say brown "exists only in our brains?"

This is equivalent to saying a C Major chord only exists in our brains since it isn't a note.

It seems reasonable to me to say that soundwaves exist in the world, but music only exists in our brains. There is something added in our perception of the soundwaves that turns them into music.

Something exits in the world, it seems. Sound waves and music are merely our interpretation of it. Maybe what actually exists in the world is music, and sound waves are what our brain invents when the music is too complex for it to grasp?

I guess my point is that sound waves are not terribly controversial. A simple device can measure sound waves present in the air. It would be much more difficult to build a device that told you whether music was playing. Reasonable people could disagree about whether certain sound waves constitute music or noise.

Well if we get on to hearing, our ears do a lot better than our eyes. From the source entering into hole on either side of our head we can split the sound into a myriad of frequencies and gather a lot of information from it.

In a similar situation from a single point of light out eyes would say "sort of blue-ish". Most visible frequency information is ignored.

Color "exists only in our brains".

"It is common to say that certain wave lengths of the electromagnetic spectrum are a given color, but in truth, it is more correct to say that those stimuli are perceived to be of a certain color when viewed under specific conditions. So without the human observer, there is really no color, and practical standards for color must take this factor of human perception into account."

"Charting Color from the Eye of the Beholder" 2005

https://www.americanscientist.org/article/charting-color-fro...

Our brain makes up every color. Color does not exist. Wavelengths exist, but color does not. Our brains turn the wavelengths we see into colors.

Everything is a wave.

Does 700 nm wavelength light exist? If so, so does red.

Pretty stupid to label something "not real" just because it's not a pure spectral colour.

I guess white isn't "real" either.

  The visible spectrum is the range of wavelengths of light our eyes can detect.

I mean, that sounds nice, but pink, brown, white, and black clearly aren't on that spectrum, so it's a ton more complicated than that, eh? Of those 4, only "white" is mentioned at all in the article.

The article is flat out wrong. The reason purple and violet look similar is not a trick of the brain, and has nothing to do with the color wheel "wrapping around"; it's a natural result of the frequency response curves of the three types of cones in our eyes. The two colors stimulate our cones in the same way, so of course they naturally look similar.

Most diagrams of our cone frequency responses are subtly wrong. Diagrams typically show three separate smooth, overlapping peaks, centered around red, green, and blue. What they leave out is that our L-cones (the "red" cones) also have a separate little sensitivity bump way off in the violet end of the spectrum. So when you see violet light, it's actually stimulating both the cones that are most sensitive to red light and the ones that are sensitive to blue light. This is pretty much the same stimulation pattern you get if you send both pure red and blue light into your eyes together, which is why purple and violet look so similar.

If you Google "cone sensitivity diagram" you'll mostly find the misleading versions of the diagrams, but you can see one that includes the extra bit of high-frequency L-cone sensitivity in this paper, for example: https://hal.science/hal-01565649/file/Vienot_ConeFundamental...

> There is something unique about the color purple: Our brain makes it up. So you might just call purple a pigment of our imagination.

You know, in the past, my curmudgeonly self would have rankled at a pun like this in a science article, but I don't think AI slop could come up with this, it's pretty good.

me:

for a science article come up with a pun about purple being imaginary

ChatGPT said: "Purple: Just a pigment of your imagination."

---

me: For a pop science article: finish this sentence with a pun

There is something unique about the color purple: Our brain makes it up. So you might just call purple a

2/2

ChatGPT said: pigment of your imagination.

---

me: For a pop science article about the color purple being imaginary, write an opening paragraph with some punning word play...

ChatGPT said: Purple is a pigment of our imagination—quite literally. Violet may bloom in nature, and red and blue dance on opposite ends of the rainbow, but purple itself is royally absent from the spectrum's party. Instead, it's a chromatic concoction our brains whip up to bridge a colorful gap, making purple less of a wavelength and more of a wave-lengthy illusion. So, the next time you spot lavender fields or don a plum-colored sweater, remember: you're seeing something that's vividly not there.

(all of these were separate clean contexts with chatgpt 4.5)

Sadly, these clean contexts also include vast swathes of the internet. I bet someone already came up with "pigment of your imagination".

I was responding to the parent argument that the pun couldn't have been written by chatgpt, when that is literally the _only_ pun that ChatGPT comes up with.

Come on guys, read yourselves.

The article is pretty good and to everyone not well versed in colors and wavelengths, it is very interesting and well phrased.

Attention economy -- dismiss apparent flaws asap.

Not "Purple exists only in our brains" instead " “All colors are made up by the brain. Full stop,” says this visual scientist"

Reply on news.ycombinator.com