Specifically infrared & ultraviolet, if not beyond. I guess this is more of a science-fiction kind of question than anything else.
When my uncle had cornea replacement surgery, he was able to see farther into the UV than I was, as measured by the end of the spectrum thrown by a prism. So, replacing the cornea with a more UV-transparent material would be the first step.
James T Fulton claims that the human retina is already sensitive to ultraviolet, with wavelengths between 400 nm and 300 nm; it’s the optics of the eye (the lens and cornea, etc) that block it.
He claims that there is a fourth colour-sensitivity in the retina, which, when the optics are unblocked, allows perception of new colours.
There have been experiments that show the residual sensitivity of the eye to infrared (longer than 700 nm, up to about 750 nm). They block out the vast bulk of the incoming non-infrared light that normally overwhelms the infrared. This is the same colour sensitivity as red, though, so it just looks like additional shades of red. No new colours.
As for colour perception further into the ultraviolet (shorter than 300 nm) and further into the infrared (longer than 750 nm), you might need different receptors.
My understanding is that receptors for electromagnetic radiation generally need to be comparable in length to the wavelength of the radiation they are receiving. Thus, to ‘see’ radio, you’d need a retina the size of a radio telescope.
There have been experiments done in connecting cameras to the human optic nerve, though; there was an article about it in Wired a while back.
Most other vertebrates besides mammals have four kinds of cones in the retina, as opposed to the three found in humans (as well as apes and Old World Monkeys) and the two in most mammals. The fourth cone type has pigments that enables them to see into the near-UV. So all you would need would be the genetics to produce that kind of cone cell. (Recognizing UV as some other color than the ones we perceive now, however, might also require some rewiring in the brain.)
The natural human lens filters out UV light. My eye surgeon tells me that, in the days when cataract patients simply had the original lens removed and wore very thick glasses afterward, the post-surgical patients could see UV light. Today’s procedures replace the original lens with a UV-filtering plastic lens. I have had that surgery, and I cannot see UV, any more than I could before the cataracts.
I can see pretty darn well, though. I wore glasses for half a century, and now I don’t need them.
As a point of interest, the eyes of the Mantis Shrimp are the most complex eyes known. Pretty amazing. Halfway down the page.
AskNott, could you have had a UV-transparent lens put in?
16 different colour receptors? I am now jealous.
Wouldn’t allowing the UV in potentially damage the retina?
Just think about what the poor little bastards have to spend on a monitor to get ‘realistic’ color.
I’ll stick with RGB; thankyou very much!
If enough people open up their UV sensitivity, we’ll have to start making RGBU monitors–U for ultraviolet. No idea how this will affect fellow Doper cmyk.
You can already get UV LEDs, so light sources won’t be a problem…
Very young children can see into the ultraviolet. They see it as blue, though without the odd echoing peak the red receptors have out there. The issue is transparency of the eye tissue to ultraviolet, which declines from birth. You know how paper and other things yellow with age? Well, it starts pretty quickly, in the ultraviolet.
Receptors don’t have to be the size of the wavelength to receive it. There is a sense in which the simplest antennae have to be a wavelength or a half or quarter wavelength to be efficient, unless they’re compensated, but your WWVB-listening wristwatch doesn’t have to be 3 miles long to get that signal.
Humans have sensors that work at 4 different peak wavelengths, but we’re only wired to use 3 of them to judge color. The 4th one is used by rod cells that aren’t ratioed to cone cells.
Human color sensors work by judging the rate at which they have to regenerate temporary dyes they produce. The dyes are what physically has a color dependance. Being exposed to light within the dye’s sensitivity range destroys the dye, and the retinal cell rebuilds it and reports the rebuilding process to the retinal nerves (which are an amazingly complicated netlike structure more properly thought of as part of the brain than as just nerve wiring). So, if we rearranged the right proteins to change the color absorption spectra of our dyes, we’d shift our color sensitivities.
If it’s more bands of color you want, ie more than 3 degrees of freedom in color sensation, you’d need cone cells for sensing those bands, but also need differentiation wiring in the retina. Right now we have wiring to ratio green versus red level, and wiring behind that to ratio its sum to blue. Think of this as modification to the hardware interconnection layer of the retinal extension of the brain.
Oh, a couple more thoughts.
Photons having around 2 or 3 electronvolts of energy are energetic enough to break some chemical bonds, and that’s what’s happening when they disrupt the dye we regenerate. It is very much within the realm of chemistry to tinker with the chemistry to change the photon energy the dye detects.
Ultraviolet photons have more energy and the further you go into the ultraviolet, the more kinds of bonds they can break. Eventually it starts getting hard not to break too many kinds of bonds, and the photons just kind of ruin all the structures you create, rather than changing them in tiny, constantly sensible ways. This is the UV damage issue. It would be much easier to design eyes so that they could see in the ultraviolet than it would be to make them durable while doing so.
Infrared photons have the opposite problem. There are fewer and fewer options available in your chemistry. That’s why you can’t get infrared photographic film to go much longer than a micrometer wavelength. To see further into the infrared, like far enough to see people glowing by their own thermal radiation, you’d need some different approach, like forming a thermal image on a temperature-sensitive kind of retina.
Also, water is more and more opaque to wavelengths longer than red. So, the eye’s watery nature starts making it black, at some point.
IIRC, those people able to see ultraviolet due to damage / removal of the lens, usually perceive it as white, because UV will trigger all the cones of the eye to an extent.
Can anyone correct me on this?
Also, wouldn’t it be great if we could just imagine primary colours at will? (or maybe not: you’d have no way of describing any of the colours you see to other people or of using the colour in the real world).
Oooh, and while I’m hijacking this thread, you know that condition where a person’s senses are intermingled (e.g. they may describe a food as tasting “blue”, or a sound as being “pointy”). Do people with this condition ever see colours outside of RGB?
Ah, you mean synesthesia.
http://www.bu.edu/neuropsychology/synvc.html
Here is a nifty link on birds seeing ultraviolet.
http://www.bio.bris.ac.uk/research/vision/4d.htm
IIRC they have a retina that can regenerate the damage done by UV.
Thanks, Napier. That makes things a lot clearer. I was only thinking of the simplest antennas, plus I had vague memories of an Arthur C Clarke story in which huge blimp-like beings in the atmosphere of Jupiter had radio vision with hundred-metre-diameter ‘eyes’.
:: spoiled by working with WiFi gigahertz signals these days, where even the simple antennas are only a couple of centimetres long ::
I wonder how much my own vision has yellowed and blurred…
I’m not aware that anyone makes UV-transparent implants. I discussed UV with my surgeon before the surgery. I told him I knew some birds and bugs use UV to identify flowers and track prey. Kestrels, I’m told, can spot rodent urine with their UV vision. All summer, I looked at lots of flowers to see if the colors looked different. I could not find any new color patterns, nor any vole pee.
That can happen even in people with normal vision. SAIL magazine talked of career sailors and fishermen who had to retire due to UV retina damage.
>those people able to see ultraviolet due to damage / removal of the lens, usually perceive it as white, because UV will trigger all the cones of the eye to an extent.
Can anyone correct me on this?
I have not heard this, and I don’t think it is correct, but I’m not certain. Can you tell me more?
In the world of color science, there are curves of sensitivity as a function of wavelength, separate curves for our three kinds of cone cells. It isn’t quite right to call them red, green, and blue, because all the curves are wide and they overlap considerably. But the curves are not equally sensitive to short wavelengths. I’m looking at them now, in Principles of Color TEchnology, by Billmeyer and Saltzman, and the “blue” curve is much more sensitive than the other two at short wavelengths.
Ultraviolet damage, mechanical damage to the lens, the clouding of the eye with age, and other things can make the eye glow more when exposed to ultraviolet light. Since you’d sense this glow, and it can be whitish, that is somewhat like sensing ultraviolet as white. But it isn’t the same as color vision. For example, you don’t see white coming from an ultraviolet lamp in this context, you see diffuse white everywhere if the ultraviolet light is hitting your eye anywhere.
Say, there is a way, I expect, to see colors most people have never seen. Since the tristimulus curves overlap, you can’t see green light with your green sensors without your red and blue ones being stimulated somewhat too.
But you can tire out cones.
So, I think if you create a bright light by combining very short and very long wavelengths, say 380 nm and 700 nm, which are hard to see on their own and are very far down on the green curve, and you stare at this light, you could fatigue your red and blue sensors. Then, you suddenly look at a pure green light, say 510 nm. I think you would see this as more intensely green than anything you had ever seen before, because you would have a higher green sensor to other sensors ratio than you’d ever had before.
Mind you, it’s a little dangerous using wavelengths that you can only barely perceive to deliberately push part of your retina into mild disfunction. I’m only speaking hypothetically here.
It is also worth noting that you can see very different wavelengths, if the source is bright enough. I have worked with half watt lasers of wavelength 785 nm, which are easily visible as red. Most people would consider 785 to be infrared, but if it’s bright enough, it isn’t. These lasers are also dangerous, and I have to have special training and special eye exams as part of getting to mess with them. Besides, they burn holes in your fingers if you’re not careful.
Well, I’m still planning on building a pair of these! 
How about iRGBu monitors (already sounds like an Apple product)?
Keep in mind this experiment in which mice (which have 2 types of cones) are given extended color vision of the type humans (which have 3 types of cones) have by injecting a human gene into a mouse’s chromosomes. Startlingly enough, the mice had no trouble taking advantage of their new color vision, and did not seemed particularly disoriented by what must have seemed an unusual development. Seems like something that could be done on humans.