Novel color via stimulation of individual photoreceptors at population scale
The olo experiment was very interesting, and it told me that today we even have the technology to stimulate a single cone cell one by one in time. I know that we can't accurately display the olo on screen right now, which also prevents any of these articles from actually containing a picture of the olo. I think it's very close to #00FFEE, and I'm making it the colour of my Hacker News's top bar.
Color is not a property of wavelength. There's nothing special about photons wiggling in the 380 to 750nm range.
In general it's not necessary to be this pendatic, but given the topic here, I think it's important to realize this. It takes a while because we are so good at projecting our internal experience outward.
Remember the blue / black dress?
did send a specific wavelength to your retina, but that wasn't violet.
It was, by definition
Color is not a property of wavelength.
Sure, it's a label
There's nothing special about photons wiggling in the 380 to 750nm range.
There is - they activate different receptors your brain relies on, hence leading to a distinct (from other wavelengths) sensation
What if we were sensitive to the 200 to 500nm range? What would be blue, violet and red then?
Our eyes and brain are the one constructing what we perceive as color. It doesn't exists outside of us.
Here's good article on the subject: https://anthonywaichulis.com/regarding-perception-photograph...
What if we were sensitive to the 200 to 500nm range?
And as such, they are both a construct of the brain, as any other colors, like... white.
What we label as "violet wavelength" is only a narrow projection of our experience outward. Case in point, we don't have such colorful (eh) names for other EM wavelength.
I say narrow because you could take this pure laser and change th surrounding and you will inevitably perceive it differently, even though the power and wavelength are the same.
Magenta and purples are constructs by the brain, as you mention.
If I shine some wavelength to your eyeball and you say "it looks blue", but then I change the surrounding and now it looks white, I don't think you would conclude that the original wavelength is blue.
We have a many examples like this, which prescribe that vision is not at all an accurate wavelength measurement device.
many people have never actually seen the colour "violet" which is a single wavelength of visible light
The violet seen in a rainbow (in nature, not a photo) is legit single wavelength violet. Same with the rainbows created from shining white light through a prism.
It's true that you don't really get to see it in isolation very often though. Maybe some flowers, birds, or butterflies? Or maybe the purple glow you get from UV lights?
Second, the term imaginary color already exists, and it refers to a specific thing[0], and the colors on the line of purple are not one of them. What you are describing is a non-spectral color. They exist in day to day life and in nature, they simply do not have an associated wavelength.
That's why red+blue=purple feels a bit like violet. It creates a similar double firing.
(And why red plus green gives an even more accurate yellow. The long and medium cones have a lot of overlap.)
I think the misconception comes from plots of XYZ color matching functions[3]. The X color matching function indeed has a local maximum in the short wavelengths, but X doesn't represent L cone stimulation; it's a mathematically derived curve used to define the XYZ color space, which is a linear transform of LMS color space selected for useful mathematical properties.
[1] https://en.wikipedia.org/wiki/LMS_color_space#/media/File:Co...
[3] https://en.wikipedia.org/wiki/CIE_1931_color_space#/media/Fi...
Magenta, formed by mixing red and blue, does not exist in nature. For that reason, "magic pink" (full-brightness magenta, #ff00ff) is often used as a transparency color when the image format does not support an alpha channel (e.g., sprite sheets, Winamp skins).
What 'does not exist in nature' is a single wavelength that produces the equivalent stimulation of your L, M and S cone cells as a mixture of red and blue light does.
But most of what we see in nature is not single wavelength light - it's broad spectrum white light reflecting off things with absorption spectra.
The reason stuff looks so weird under certain LED lights or pure sodium light is that the source light isn't broad spectrum - it's missing wavelengths already - so the way it interacts with absorption spectra is unintuitive. Something that looks blue under white light should still look blue under blue light - but a blue LED might just be emitting blue frequencies that the object absorbs, so it looks black instead.
The purple colour we see is formed from a mixture of red and blue, whether it's something in nature, screen displaying or printing.
Well if it’s on an RGB screen, or printed with CMYK inks then it’s not ‘real’ violet, but there must be plenty of natural and artificial pigments that are actually reflecting violet light and not blue + red light. I imagine any pure compound would be doing this. E.g cobalt phosphate (aka cobalt violet).
You could tell by illuminating a sample with different light sources. See metameric failure:
https://en.wikipedia.org/wiki/Metamerism_(color)#Metameric_f...
Some can be created at home without any special equipment. For example, you can't mix red and green and create a "redgreen," but if you cross your eyes and have one eye see red and the other see green, you might see a new color you haven't seen before.
I also see weird colors in displays with a high frame rate that cycle between colors quickly. And at one point, I had a laser shot in my eye, which destroyed part of my vision. Initially, in that spot, I saw a weird iridescent silver-greenish color I had never seen before. Although that was pretty cool, I wouldn't recommend repeating this involuntary experiment just to see that color.
Of course, colors are a hallucination our brain produces, so perhaps different brains deal differently with an unusual experience like Stygian blue.
It looks like black that glows blue
I could maybe buy that as a description, but...
I think it is fairly described as a new color—I would be quite unsettled if I encountered it in real life.
I don't feel this follows. There are a lot of things that would unsettle me if I saw them, like if someone gave off a visible aura. Heck, I even found a "black flame" a bit unsettling, and I saw a literal video of it on YouTube (look it up if you don't know what I'm referring to). I'd feel similarly if I saw a transparent human too. The feeling you get - or the fact that you haven't seen something visually similar before - doesn't really imply it's a new color, I think!
If we can't differentiate colors on saturation, hue or intensity
I don't think I follow. We can obviously distinguish all of these and do it on a daily basis... what do you mean we can't?
https://en.wikipedia.org/wiki/Blue%E2%80%93green_distinction...
Five subjects were recruited for this experiment ... Subjects 10001R, 10003L, and 20205R are coauthors on the paper and were blinded to the test conditions but were aware of the purposes of the study. The other two subjects were members of the participating lab at the University of Washington but were naive to the purposes of the study.
Is it normal for the authors to experiment on themselves and their colleagues like this? Or did they not like the idea of laser-stimulating the photoreceptors of random strangers?
I tooke a bodkin gh & put it betwixt my eye & the bone as neare to the Backside of my eye as I could: & pressing my eye with the end of it (soe as to make the curvature a, bcdef in my eye) there appeared severall white darke & coloured circles
https://www.newtonproject.ox.ac.uk/view/texts/normalized/NAT...
What I find fascinating is the neurological resilience that can be observed at cellular and behavioral levels to bounce back after an event like that.
Non-chemical interventions, like adaption wearing special glasses that flip vision(1), are quickly accounted for by a healthy brain.
1:https://www.npr.org/2012/12/14/167255705/a-view-from-the-fli...
The claim left one expert bemused. “It is not a new colour,” said John Barbur, a vision scientist at City St George’s, University of London. “It’s a more saturated green that can only be produced in a subject with normal red-green chromatic mechanism when the only input comes from M cones.” The work, he said, had “limited value”.
[1] https://www.theguardian.com/science/2025/apr/18/scientists-c...
Novel color via stimulation of individual photoreceptors at population scale
And they say, These results are proof-of-principle for programmable control over individual photoreceptors at population scale.
As for colour, language does not help very much with being able to see and understand them. What helps more is playing with photographic software and getting a feel for the relations within a system like HSL, or RGB.
It would be cooler still if this technique could be used for future VR technology, creating full immersion by targeting all photoreceptors individually. But unfortunately... the optics of the eye does not actually allow individual cones to be fully isolated, as the spot size would be below the diffraction limit. They discuss this in Fig. 2 and the first section of the results.
Even with a wide-open pupil and perfect adaptive optics, there would be 19% bleedover to nearby cells in high-density areas, while what they achieve in practice is 67% bleedover in a lower-density (off-center) area. This is enough to produce new effects in color perception, but not enough to draw crisp color images on the retina. :(
(All illuminated signage could be said to draw on one's retinas, after all. The major differences I see with this method beyond improved gamut are first that it rasters, and second that I think we have to worry what happens if it fails to raster...)
If I understand correctly, they first use one type of spectroscopy (AO-OCT) to image the eye and build a map classifying the type of cells, and then use AO-SLO to find the positions of cells in real time. I assume that AO-OCT can't image at a sufficient rate for the second part (or they would just use one type?) so they need to first build this classification map, and then use it to match the position of cells to their type (e g., by overlaying the positions of cells with the classifications and making them line up).
https://news.ycombinator.com/item?id=43736005 ("Scientists claim to have found colour no one has seen before (theguardian.com)" — 27 comments)