Why do you need at-least three colors to make every other color?

[Avichal]

Since with the help of two different wavelength of light we can make every other wavelength of light why do we need three then - like red, green, blue or red, green, yellow? I guess red, blue or red, yellow will suffice.

 

[tiny-tim]

Hi Avichal!

Since with the help of two different wavelength of light we can make every other wavelength of light …

nope

you can't make a single wavelength out of any other wavelength(s)

… why do we need three then - like red, green, blue or red, green, yellow?

humans need three, because that's the way the human eye works

(i think some humans actually need four … so they sometimes see two things as different colours although most humans see them as the same colour)

each colour receptor responds to all wavelengths, but in unequal amounts

the response is a sort of bell curve, with a maximum in the red green or blue

see http://en.wikipedia.org/wiki/Color_vision for details ​

 

[D H]

Since with the help of two different wavelength of light we can make every other wavelength of light ...

nope

you can't make a single wavelength out of any other wavelength(s)

To elaborate, suppose you have a white wall, a red laser, a yellow laser, and an orange laser. Aim the red and yellow lasers at the same spot on the wall and aim the orange laser at a different spot on the wall. Next tune the intensities of the red and yellow lasers so that the colors of those two spots is indistinguishable to the human eye.

Just because those two spots are indistinguishable to the human eye does not mean that they have the same wavelength. That they appear to be the same color is just an optical illusion based on how the human eye works.

 

[dlgoff]

... Just because those two spots are indistinguishable to the human eye does not mean that they have the same wavelength. That they appear to be the same color is just an optical illusion based on how the human eye works.

And to add a little color to the thread

... there are three types of color-sensitive cones in the retina of the human eye, corresponding roughly to red, green, and blue sensitive detectors.

The Color-Sensitive Cones

 

[NemoReally]

... and not forgetting that many colours don't have a "wavelength" at all as they don't exist on the spectrum.

 

[D H]

... and not forgetting that many colours don't have a "wavelength" at all as they don't exist on the spectrum.

E.g., purple. Or white.

 

[1MileCrash]

E.g., purple. Or white.

Thought it was pink..

But yeah, nothing to do with actual colors or light, just part of our biology. Our vision is trichromatic.

 

[Avichal]

Oh ok...its not the light or color. Its how our eyes work! So there is a possibility that some animals need only two colors i.e. they are dichromatic(or whatever it is called)?

 

[snorkack]

Oh ok...its not the light or color. Its how our eyes work! So there is a possibility that some animals need only two colors i.e. they are dichromatic(or whatever it is called)?

Yes.

Almost all mammals other than monkeys are dichromatic, like colourblind. Not monochromatic, though.

 

[Avichal]

Yes.

Almost all mammals other than monkeys are dichromatic, like colourblind. Not monochromatic, though.

Why is dichromatic colourblind?

 

[NemoReally]

Why is dichromatic colourblind?

In a normally trichromatic (or better) species, it means that the animal lacks the ability to see the full normal colour "spectrum". A dichromatic human is regarded as colour blind (even though they can see some colours) but I'm not sure that a normally dichromatic species can be properly be regarded as colourblind. Using that definition, then you could probably make a case for humans being normally colourblind as they lack the colour range of a tetra- or penta-chromat. Indeed, we have no idea what colour really is and, consequently, what possible values it could take. I think a monochromat could be regarded as colourblind as there is no indication that they see "colour" at all.

 

[Alfi]

Why do we need four colours to make a map, if we can only see three?

 

[NemoReally]

Why do we need four colours to make a map, if we can only see three?

The 2 questions are not related. Stealing shamelessly from Wiki ..
The 4-colour map theorem states that no more than four colors are required to color the regions of the map so that no two adjacent regions have the same color, where two regions are called adjacent if they share a common boundary that is not a corner, and corners are the points shared by three or more regions. In fact, the word "colour" in this context is effectively just a label to indicate some unique marking (eg, letter, number or shading); for example, you could replace 'blue', 'green', 'red', 'yellow' with 'a', 'b', 'c', 'd', in which case you could call it the "4-letter problem"!

As for our vision, we can see significantly more than 3 colours - just open up the colour picking palette in Paint. As indicated in earlier threads, all things being equal, our eyes possesses 3 specialized sensors (the cones) that have differing sensitivities to light of different wavelengths (energy, frequency, take your pick). Our brain then (somehow) combines the outputs from adjacent cones to construct "colour". A (speculative) reason we see so many colours is to aid identification of fruits, plants and other objects that appear visually similar in grayscale but are quite different in their edibility or danger.

 

[D H]

Thought it was pink..

Pink is a non-spectral color, but a rather boring one. The same goes with light green, light blue, light anything. Pink is just light red, basically a white spectrum but with a bit of a peak in the red. (BTW, white is also non-spectral.)

Purple and magenta are much more interesting non-spectral colors. Suppose you shine a green laser and red laser at the same spot on a white wall. You'll see a yellow spot, a spectral color whose frequency is intermediate between the frequencies of the green and red lasers. A similar thing happens with a blue laser and a green laser. Now you'll see cyan, another spectral color.

Use a blue laser and a red laser and you'll get something very different. You don't see green, which is intermediate between blue and red in terms of frequency. You see magenta. Or purple. Or some other non-spectral color that in our heads is very, very different from green. Purple and violet are visually similar colors, yet spectrally they are very different.

Our brain then (somehow) combines the outputs from adjacent cones to construct "colour".

And what it does to make the linear range from red to violet wrap around to instead form the color circle is part of that "somehow". Purple is a pigment of our imagination.

 

[NemoReally]

... Purple is a pigment of our imagination.

Ouch! If you send me your address, I'll arrange delivery of a smart bomb ...

 

[Avichal]

In a normally trichromatic (or better) species, it means that the animal lacks the ability to see the full normal colour "spectrum". A dichromatic human is regarded as colour blind (even though they can see some colours) but I'm not sure that a normally dichromatic species can be properly be regarded as colourblind. Using that definition, then you could probably make a case for humans being normally colourblind as they lack the colour range of a tetra- or penta-chromat. Indeed, we have no idea what colour really is and, consequently, what possible values it could take. I think a monochromat could be regarded as colourblind as there is no indication that they see "colour" at all.

So can you say that a trichromatic species sees more colors than dichromatic and similarly higher i.e. tetra or penta see even more.

Now I feel as if I don't know what exactly color is?

 

[NemoReally]

So can you say that a trichromatic species sees more colors than dichromatic and similarly higher i.e. tetra or penta see even more.

In principle, simply on the basis that there is an extra 'dimension' for colour to extend into (by analogy, consider the rooms available in a single story dwelling to those in a high-rise tower of the same footprint and rooms per floor). However, the colour discrimination might be lower in a particular trichromat than in a dichromat (back to the analogy, d = 30 by 30 rooms and t = 5 x 5 x 5 rooms) as a result of cone sensitivity and / or brain processing. In addition, the colour dimensions might not contribute equally (eg, a 5 x 20 x 10 block or even (5 x 20) and (10) or some rooms might cover more than one floor).

... whether any of these happens in reality, I don't know.

Now I feel as if I don't know what exactly color is?

I don't think anybody knows what colour actually is. However, we can develop mathematical models that allow us to functionally describe and model it. You might the following webpage of interest ... http://www.archimedes-lab.org/color_optical_illusions.html#

 

[snorkack]

And what it does to make the linear range from red to violet wrap around to instead form the color circle is part of that "somehow". Purple is a pigment of our imagination.

We actually have an idea how. Another picture claiming to be the graph of human visual pigment sensitivity. But with vital features missing above:
http://www.huevaluechroma.com/032.php
Note how the red pigment has a rise towards secondary maximum, from 450 nm shortward. While the blue falls from maximum at 420 nm - which the other picture claims 445 nm.

True violet, at 390-400 nm, causes less excitation of the blue sensors than true blue in the range of 420-450 nm, because true violet is blueward of the blue maximum sensitivity. At the same time, the true violet causes more actual excitation of red sensors than true blue (because true violet is already into secondary maximum) and especially more excitation relative to excitation of blue sensors (because the blues are less excited, as stated above). Therefore, monochromatic short radiation looks much like mixture of blue and red.

And for comparison, a graph claimed applicable to birds (specifically finches):
http://en.wikipedia.org/wiki/File:BirdVisualPigmentSensitivity.svg

 

[Darwin123]

Since with the help of two different wavelength of light we can make every other wavelength of light why do we need three then - like red, green, blue or red, green, yellow? I guess red, blue or red, yellow will suffice.

First, you can't "make" any wavelength of light from two other wavelengths of light. The superposition principle says that light of any wavelength can exist superimposed on light of any other wavelength.

Second, the perceived color is not uniquely determined from the wavelength. It is incorrect to say that color is wavelength. The mixtures of light with different wavelengths are perceived as colors.

Suppose two beams of light are perceived as the same shade of "green". Spectrographs taken from these beams may not be identical.

Third, human beings have three types of color receptors. This is why they perceive of mixed light the way they do. Not all animals have three types of color receptors. A bird may not perceive of those two "green" beams as the same. A bird has a finer ability to distinguish colors because it has more types of color receptors. A bird may not even perceive of a color photograph the same as a human since the palate of the film was designed by humans.

There is no universal number of colors in vision. The uniqueness of a color signature is limited by the number of types of color receptor in the eye of the animal. The more receptors, harder it is to match the all the colors which the animal perceives. By harder, one needs a mixture of light with different wavelengths to match the color.

Three is not a universal constant in terms of color perception. It just happens to be the number of color receptors in the human eye. Humans are usually trichromic.

Many mammals with color vision have three color receptors to perceive of color
However, there are animals that perceive color using only two color receptors. These are called dichromatics. Here is a link and quote on dichromatics.

http://en.wikipedia.org/wiki/Dichromacy
“Dichromacy (di- meaning 'two' and chromo meaning 'color') is the state of having two types of functioning color receptors, called cone cells, in the eyes. Organisms with dichromacy are called dichromats. Dichromats can match any color they see with a mixture of no more than two pure spectral lights. By comparison, trichromats require three pure spectral lights to match all colors that they can perceive.

….
The exceptions to dichromatic vision in placental mammals are primates closely related to humans, which are usually trichromats, and sea mammals (both pinnipeds and cetaceans) which are cone monochromats.[11] New World Monkeys are a partial exception: in most species, males are dichromats, and about 60% of females are trichromats, but the owl monkeys are cone monochromats, and both sexes of howler monkeys are trichromats.
In recent investigations, it was found that some ground squirrels possesses dichromatic vision. It is beneficial to use situations in which less than the total visual system is functional when studying vision. This is present in systems in which cones are the sole visual receptors such as the dichromatic colour vision in squirrels.”Most birds are tetrachromatics. They have four color receptors. Here is a link and quote.

http://en.wikipedia.org/wiki/Tetrachromacy
“Tetrachromacy is the condition of possessing four independent channels for conveying color information, or possessing four different types of cone cells in the eye. Organisms with tetrachromacy are called tetrachromats.
In tetrachromatic organisms, the sensory color space is four-dimensional, meaning that to match the sensory effect of arbitrarily chosen spectra of light within their visible spectrum requires mixtures of at least four different primary colors.
Most birds are tetrachromats.[2] Tetrachromacy is also suspected among several species of fish, amphibians, reptiles, arachnids and insects.”

Humans that are not colorblind are trichromatics. Here is a link and quote.
http://en.wikipedia.org/wiki/Trichromacy
“Trichromacy or trichromaticism is the condition of possessing three independent channels for conveying color information, derived from the three different cone types.[1] Organisms with trichromacy are called trichromats.
The normal explanation of trichromacy is that the organism's retina contains three types of color receptors (called cone cells in vertebrates) with different absorption spectra. In actuality the number of such receptor types may be greater than three, since different types may be active at different light intensities. In vertebrates with three types of cone cells, at low light intensities the rod cells may contribute to color vision, giving a small region of tetrachromacy in the color space.”

There is no universal color palate! Different species of animals have different palates! The color palate varies even within some species.

 

[NemoReally]

This subject is one in which the terminology can cause a certain confusion. It is important to be clear that "colour sensors" do not actually detect "colour" any more than, say, a CCD camera is an "image" detector. These devices report certain patterns of incident light by responding to their spatially-distributed energy and intensity; the brain, as far as we can tell, creates the resulting "image" that "you" see.

 

[pumila]

Going back to the original question, mixing two monochromatic beams might be able to give you a wide range of colours. However, the brain does not sense purely colour - the brain reports saturation and tone. So pink is a low-saturation red. Brown is a high tone (dark) yellow, it is impossible to produce a brown colour sensation with an even illumination over the field of vision.

The three sensors of the reference human colour vision allow the detection and discrimination of a wide range of colours. They do not permit the discrimination of all possible colours as the mapping of a continuous spectrum into a tricolour sensor system results in a lost of information that causes many different spectral layouts to map into the same identical colour sensation. This is called metamerism.

Going onto some of the replies, colour blindness is the inability to discriminate colours adequately. There are two types of colour sensor, short wave cones that detect blue and long wave cones that detect red and green. The blue cones are in a part of the DNA that is very stable. However the red/green sensors are on a part that is a lot less stable. This leads to variability where up to six copies of one colour gene may exist in one person's dna. Only one of these is normally chosen by the physiology of the individual, and some can make an unusual sensor wavelength. Hence the gap in wavelength between the red and green sensors can reduce or even disappear, leading to variable degrees of red-green colour blindness. In very rare circumstance, where one parent has an abnormal sensor the individual - always a female - can have three active long wave sensors in the red-green region. These females are tetrachromats, and the first one discovered was able to discriminate the spectrum into 10 discrete colours instead of the reference 7.