h2g2 The Hitchhiker's Guide to the Galaxy: Earth Edition

From something I was wondering a couple of days ago, and was reminded by the QI question...

If the rods and cones in the human eye detect different frequencies of light, is there any part of the visible light spectrum humans can't see?

EG. Will there be a frequency of light which can (for example) be lased through a smoke filled medium without being seen?

"is there any part of the visible light spectrum humans can't see?"

I think the appropriate rejoinder is..."duh!"

B

The sensible answer is that visible light is just part of the electromagnetic spectrum, between infra-red and ultra-violet. There are plenty of animals that can sense these two wavelengths, and there is no reason why humans should not evolve to do the same.

B

I think the answer is probably 'no', but clearly we are more sensitive to different wavelengths. The three types of cone (red, green, blue) all have peak responses at specific wavelengths, so I guess you could find a colour that gives the minimal overall response if you wanted to hide something, for example.

Sorry, but this is quite informative:
http://en.wikipedia.org/wiki/Color_vision

OK, for the smartarse in the crowd, let me rephrase that.

Humans, as a whole, can see in the range of frequencies between about 300 and 700 nm. Given the design of the human eye, is there any particular flaw in the general design (allowing for the fact that there is no specific evidence as to a 'designer', as opposed to being an evolutionery self-improving pattern ) which will allow for being unable to see any specific frequency or frequencies within that range, rather than the human brain making up the colour(s) detected from a range of other frequencies which can be detected by the design used.

See why I asked the question I did before?

As to the IR and UV ends of the spectrum, there are undoubtedly those already who can see further than most into those frequencies. As to everyone developing the ability, I imagine it's not that likely. Technology can give the ability to percieve those areas of the electromagnetic spectrum to all, so any genetic 'survivor' trait is unlikely to surface through random selective breeding.

Ahem, Post 4. Is that not the answer you were looking for? The human eye has all sorts of flaws, blind spots etc, but as you can see from the link there are no gaps in the response across the visible spectrum.

"Technology can give the ability to percieve those areas of the electromagnetic spectrum to all, so any genetic 'survivor' trait is unlikely to surface through random selective breeding. "

Not sure that the mere availability of technology is enough to inhibit evolution - IR and UV imagers don't play a big part in my life, anyway.

There are of course some humans who have a congenital or acquired visual defect. Most of them have some response across the whole visual spectrum, just not the same as other people. If you have a cone dystrophy, you're going to lose response to some wavelengths of light outside the range of response by rods.

Sorry Dave,
The link is to wiki, which my work computer doesn't like much, so I hadn't clicked on it before my earlier response. Have now done so, and the articke shows that there isn't a specific blind spot. Thanks for the link, it's one of those wiki ones that actually works on this PC.

Re the "Not sure that the mere availability of technology is enough to inhibit evolution - IR and UV imagers don't play a big part in my life, anyway." bit...

No, having technology wouldn't inhibit the random development of abilities in the direction of IR and UV sensitivity, but it would remove the evolutionary advantage of doing so, which, in turn, reduces the liklihood of the ability developing.

Teas,
True, but they are a group of specific individuals (or, more accurately, individuals with specific conditions) which have reduced capacity / ability compared to the wider human population.

As an additional question I thought of last night,

Where has evolution taken the eye beyond human? Specifically, where has a "human style" eye evolved in another animal, and progressed further, rather than one of the other variants of eye that have evolved?

As most of the higher vertebrates have similar eye construction, bird's eyes are "human-style" but with a much greater visual acuity and an expanded color field (many birds can see well into UV).

My understanding is that, although there is no 'gap' in the visible spectrum, our eyes are more sensitive to colours that the cones are best at detecting - the primary colours red, green and blue.

The logical corollary of that is that our eyes are less sensitive to some other frequencies. So purple just doesn't jump out at us as much as scarlet.

Gif

Blimey! I've made quote of the day!

The first I'm aware of, as it happens.

Gif, the colours that are cones are best at detecting aren't actually RGB. You can see that they're actually something like blue, slightly-bluey-green and yellowy-green from looking at these two pictures from wiki:

http://en.wikipedia.org/wiki/Image:Cones_SMJ2_E.svg
http://en.wikipedia.org/wiki/Image:Spectrum441pxWithnm.png

IIRC, RGB are the colours where our 'discriminability' is relatively low (so that we tend to perceive reds and oranges as more similar than yellows and greens because discriminability is lower in the red-orange portion than in the yellow-green portion, even though the yellow-green portion of the spectrum is actually smaller than the red-orange part). Again, IIRC the discriminability of colours is related to the gradient of the graph of the response of the cones to different wavelengths, so that where the gradients are high, discriminability is better.

The eye has chromatic aberration, so that different wavelengths of light are focussed at slightly different positions. That gives an effect of red hues being perceived as closer compared with blue ones. I think that's why red against blue is rather uncomfortable to view.

I'd noticed that!

Gif

There is a great deal more to colour vision, and vision in general, than what's going on photochemically on the retina and electrically in the optic nerve.

I first became aware of this many years ago when I was boffing a psychology student whose favourite subject was the psychology of vision. I didn't even know there WAS a psychology of vision (I guess I was thinking optical illusions were a manifestation of faults in the mechanics of the sense, rather than the interpretation of the signal), and a whole world of new wonders opened up.

An excellent example of just how weird our colour vision is can be seen here. Enjoy: http://www.youtube.com/watch?v=mf5otGNbkuc

One effect I found interesting was the possibility of 'seeing' colours in essentially monochromatic input.

Years ago, I used to have a large area of a wall in a room covered in colour prints of photos I'd taken, generally 12x8s, which I'd got very used to seeing.

Playing around with some red LEDs one night, lighting up the wall with no other light sources apart from the LED, which has a close-to-monochromatic output, the prints appeared essentially as expected - like monochrome prints lit by pure red light.

However, after getting used to the lighting, I found I could 'see' the colours I knew should be there - the redness didn't disappear, but there were also definitely colours from across the spectrum which appeared to be visible in the photographs.
The colours were some combination of 'obvious' ones (blue sky, green grass, etc, but also had some that I'd have to have memorised, such as arbitrary colours for some people's clothing.

I'd guess what was happening was that I was actually getting pretty limited input from the colour vision system - the red light was presumably exciting low and medium wavelength receptors in some ratio that was effectively the same for all points in the scene, and many of the colour circuits didn't really have anything interesting to work on.
Maybe, given my knowledge of the 'correct' colours, there wasn't sufficiently powerful contradictory information coming in to override it, and so the knowledge won out, at least to a large enough degree for the colours to appear.

Hi Potholer,

Interestingly, I was watching a documentary on how they have recovered some of the 'missing episodes' of Doctor Who (yes, , I know, but stick with me here!). With some of the 3rd Doctor's episodes, they were made on colour videotape but transferred to B&W film for sale overseas. Now, only the B&W versions of some episodes exist.

They've previously had some success combining the high-quality B&W footage with low-quality colour home video recordings, but of course that only works were someone taped (and kept) the episode - not common in the early 70s!

So for at least one episode they have now restored the colour by extracting the colour information from the B&W film alone! They fine-tuned it by reference to existing epsiodes from the same story, but apparently the basic colour information can be deduced from the fine structure of the black dots on the film, or somesuch.

Don't know if that's linked to the effect you got - the guy did say he discovered it after seeing colour traces in a B&W title credit sequence...

Gif