The Naked Scientists

The Naked Scientists Forum

Author Topic: How can we see violet if we only have blue retinal receptors?  (Read 5450 times)

MartinF

  • Guest
MartinF asked the Naked Scientists:
   
Hi

I have a question about colour vision.

As I understand it, we have three colour receptors in our eyes, Red, Green and Blue.

If we look at a rainbow, the colours go Red, Orange, Yellow, Green, Blue, Indigo, Violet.

As the frequency of the light increases, the different receptors respond in different amounts. That is, first the Red, then a mixture of Red and Green, then Green, then a mixture of Green and blue etc.

What I can't understand is how we see violet if the highest receptor we have is Blue?

I have not explained it very well, but I hope you get the idea

Thanks

Martin

What do you think?
« Last Edit: 12/05/2009 23:48:27 by chris »


 

Offline techmind

  • Hero Member
  • *****
  • Posts: 934
  • Un-obfuscated
    • View Profile
    • techmind.org
An excellent question!

The answer is that the "red" receptor in the eye, as well as sensing long wavelength light (ie red), also has a second, weaker, lobe of sensitivity in the blue/violet end of the spectrum. When the eye/brain gets the appropriate combination of stimulus from the "red" and blue receptors it registers violet.

From an applications perspective, this is why we can use mixtures of red and blue light (including on a computer/TV screen) to make violet colours for which the apparent spectral colour is a shorter wavelength than the blue lightsource itself.


See also my web page(s):
http://www.techmind.org/colour/
and
http://www.techmind.org/colour/spectra.html


An artists' colour wheel, or the CIE diagram (see webpage) via the "line of purples", completes the colours we see with the other red+blue combinations (purples and magentas). These colours do not correspond to any single-wavelength (monochromatic) source but can only be made by mixing red and blue (long and short wavelength) light.
« Last Edit: 12/05/2009 16:18:54 by techmind »
 

lyner

  • Guest
I am not sure what you are suggesting, technind, but you cannot produce a metemeric match outside the gamut of the three primaries of the display.
There seems to be some confusion between the analysis and synthesis situation. The analysis curves of the eye are not 'red, green and blue'. They are all very broad and include sensitivity to many wavelengths. The spectra of the primariy phosphors in displays are not broadband but optimised for colourimetry and brightness (and cost, no doubt).
 

Offline chris

  • Neilep Level Member
  • ******
  • Posts: 5335
  • Thanked: 65 times
  • The Naked Scientist
    • View Profile
    • The Naked Scientists
How can we see violet if we only have blue retinal receptors?
« Reply #3 on: 12/05/2009 23:50:26 »
I think the "blue" cones are sensitive to a range of wavelengths, centred on blue but including some shorter wavelengths in the violet, and hence whilst not maximally activated the cones can also register and respond to light of this colour.

Chris
 

Offline techmind

  • Hero Member
  • *****
  • Posts: 934
  • Un-obfuscated
    • View Profile
    • techmind.org
How can we see violet if we only have blue retinal receptors?
« Reply #4 on: 13/05/2009 21:54:27 »
I am not sure what you are suggesting, technind, but you cannot produce a metemeric match outside the gamut of the three primaries of the display.
Yes, very true. I'm guilty of some slightly imprecise thinking which I shall now clarify...

What you can do from a tri-primary display is create a colour which is a slightly desaturated version of a colour which is out-of-gamut ... and in particular you can mix a small amount of red-primary emitter with your blue primary emitter to make a colour which is equivalent (metameric) to a {monochomatic violet (of wavelength shorter than the dominant wavelength of your blue primary) diluted with a small amount of white}.


Quote
The analysis curves of the eye are not 'red, green and blue'. They are all very broad and include sensitivity to many wavelengths.
Totally agree. The sensitivity curves are broadband and for analytical purposes can be taken as being the CIE Xbar, Ybar, Zbar functions - although this isn't quite the complete story. The spectral response of all three receptor types substantially overlap, especially the medium and long (green and red).

Quote
The spectra of the primariy phosphors in displays are not broadband but optimised for colourimetry and brightness (and cost, no doubt).
Indeed. And optimised for efficiency and, in the case of CRT phosphors, lifetime and low-persistance.
 

lyner

  • Guest
How can we see violet if we only have blue retinal receptors?
« Reply #5 on: 13/05/2009 23:55:56 »
Quote
What you can do from a tri-primary display is create a colour which is a slightly desaturated version of a colour which is out-of-gamut
All the brightest coloured clothes in the crowd at Wimbledon look the same colour because they are all brought back 'inside the triangle' of the phosphors and sit right near the edge. I find it quite a disturbing distortion, actually - almost as bad as burned out highlights when a picture is overexposed. Strange that it doesn't look as bad in digital photographs.
Now that blue LEDs are not a problem, maybe they'll do something about the blue phosphor and move it towards violet. But it may not really matter, bearing in mind how low luminance the blue is.
 

The Naked Scientists Forum

How can we see violet if we only have blue retinal receptors?
« Reply #5 on: 13/05/2009 23:55:56 »

 

SMF 2.0.10 | SMF © 2015, Simple Machines
SMFAds for Free Forums
 
Login
Login with username, password and session length