The second of my interactive diagrams from yesterday showed how the three types of color receptors in the human eye respond to different wavelengths of light. These three kinds of cone-shaped cells detect, respectively, short, medium and long wavelengths, and so they are called S-cones, M-cones and L-cones.
One look at the diagram and it’s clear that light is never detected only by M-cones. What we think of as green light actually triggers both M-cones and L-cones. It just triggers the M-cones a little more strongly.
So of course I started to wonder what it would look like if only your M-cones were triggered. This would be a kind of “pure green” light that no human being has ever seen. Would it look like a completely new color?
The only way I can think of to do this would be to shine a very finely patterned light at your eye, with bright dots placed exactly where the M-cones are, but avoiding the L-cones.
To make this work you would need a high speed eye tracker that measures the slight movements of the eye, and continually moves the pattern of dots to match this movement. This would ensure that each dot illuminates only an M-cone, not an L-cone.
But how small do the dots need to be? Well, one degree of visual angle is about 0.3mm on your retina. And each M-cone near the center of the retina is about 1um across, which means you’d need a dot that subtended about 1/300 of a degree in visual angle.
The biggest problem I can foresee is that people can’t easily focus an image to such fine detail. To get around this problem, the first thing I would try is to use a laser to send the image into the eye through the center of the pupil, where the eye’s lens has little effect on depth of focus.
It would take a lot of engineering to get this all working. On the other hand, it would be cool to see a new color that nobody has ever seen before.
Hi Ken, people have thought about this colour before, sometimes called by the evocative name “psychadelic aquamarine”. But there is an easier way to see it – you can tire out your L-cones by staring at something red, and then quickly look at bright green. This is called the “eclipse of Mars” illusion.
Dr. Robert Hunt used to do a regular demonstration at the Color and Imaging Conference that produced just such a “super-green” as you describe. Dr. Hunt used two optical filters in his demonstration – when used with a projector they enabled projecting a square of extremely saturated green and magenta, respectively. The demonstration started with the magenta filter, and the audience were instructed to fixate their vision on a dot at the center of the projection screen. After a short time interval, Dr. Hunt substituted the green filter instead of the magenta one. The observer’s vision was strongly adapted to the magenta color – the pigments associated with the S- and L- cones were bleached and reacted very weakly. The M- cone was unaffected, and reacted strongly to the green light. Unfortunately, I personally did not attend any of these demonstrations, but I know people who have and they said the sensation produced was that of an unnaturally pure green.
Ah, that’s great! I had asked a few people and had tried to find an answer on-line, but had not been able to find anything useful. Blogs can be a wonderful thing!
I’ve created an on-line demo of the effect HERE. It fills your screen with solid magenta, and every time you click it flips between that and green.
Too bad the effect doesn’t last. Maybe for that we’d need to use my rather more expensive technique. 🙂
William – I wasn’t familiar with the Eclipse of Mars illusion, and looked it up after reading your comment. This illusion can produce color sensations beyond the monitor gamut, but on most displays probably not outside the spectral locus, due to the limited gamut of most monitors. That is why Dr. Hunt used extremely saturated filters rather than a digital projector for his demonstration. There’s probably on online source for such filters, and the demonstration doesn’t need much else (you could use a regular digital projector showing a white frame with a small black dot in the center to aid in fixation). Now I’m tempted to try it for myself…
Ken – see my comment to William. Unfortunately the limitations of computer monitors make the effect not quite work to the degree that you would like – you can get “a green more vivid than anything your monitor can normally produce” but that’s not quite as exciting :-). I wonder if a large-gamut display could produce colors saturated enough to cause the effect to break out of the normal perceptual color space. You could try a wide- gamut display – but optical filters are probably the best bet.
Yes, I agree. I just tried the circular “Eclipse of Mars” illusion, and it is much more compelling.
Clearly, as you say, it is something to be tried using proper saturated color filters, rather than a standard LCD monitor. But already it’s breathtaking!
It was described to me by Robert M. Boynton, the late color vision scientist, as a “super-saturated green” (or as I like to tell people, a green you’ve never seen.” A way to experience the super-saturated green, he said, was by red-adapting by completing depleting the color pigments in the red cones by exposing them to pure red light at a high level…. then it would take about seven minutes for them to recover…. I believe the red-sensitive pigment is called “erythrolabe”.
The reason why it is red, one depletes, is because color vision has an opponent color system, with a R-G channel, a Y-B channel, and a B-W channel. We’re talking about the Red-Green opponent channel.