Hi Robert,
Thanks for the link to the paper which I found fascinating and
is very much appreciated.
Quote:
Originally Posted by madbadgalaxyman
As we know, Humans generally have only three distinct colour photoreceptors (for red and green and blue light) in the retina of the eye, while most other mammals have only two types of cone cells (blue and green sensitive), but the mammalian colour response is very poor compared to that of this species of mantis shrimp, which encodes its visual response using 12 colours (using 12 distinct photodetectors), from the ultraviolet through to the infrared.
|
I suggest it is best not to refer to the three sets of cone photoreceptors in
the retina of the human eye as being for red, green and blue and instead
refer to them as long, medium and short wavelength receptors, lest
someone is left with the mistaken impression that they are for perceiving
colour in a way analogous to emulsion or digital photography or on
a colour television monitor.
We perceive colour owing to a remarkable piece of processing that
takes place largely in the lower back of the brain in the visual cortex.
Quote:
|
Originally Posted by Robert
Given the remarkable width of spectral response, and the fine sampling of the spectrum into 12 distinct colours by the eye of this Mantis Shrimp, it would seem that these crustaceans would make the ultimate visual astronomers!
|
Or perhaps not.
Quote:
But why does a Mantis Shrimp need to be such a good visual observer that it synthesizes its visual response by using 12 different colour detectors in its eye? Well, conceivably, it all boils down to needing to communicate with other Mantis Shrimps, perhaps for the purposes of mating or defensive hostility, or maybe for other forms of communication.(many Mantis Shrimps are very colourful creatures).
Hmmm....but what do Mantis Shrimps actually talk about, using this very sophisticated colour coding?
|
When I first saw the diagram of the wavelength responses you posted,
I had a personal hunch as to what purpose they might be for before reading the
article.
My hunch was, rather than being able to see "more" colours, they were
an evolutionary mechanism to allow the shrimp to perceive the colour
of any one object to remain the same under a range of lighting conditions.
It is what is called "colour constancy".
We have the remarkable ability to perceive the colour of an object to remain
essentially constant over a broad range of lighting conditions. For
example, we can view an orange or green leaves indoors in dim
light or under fluorescent light and yet when we view the objects outside
in bright sunlight, the orange and the green leaves still appear to be
orange and green. Likewise white still appears as white either in
sunlight or under incandescent or fluorescent light.
This is completely different to how colour emulsion film or digital photography
works. As we know, with colour film, a white piece of paper photographed
under fluorescent lights will often have a green cast, with tungsten lighting
it will be yellowish, whilst with a flash light it may be slightly blue.
Digital cameras have processing that tries to compensate for "white
balance".
So why do we, or the shrimps, need colour constancy?
For a start, without it, the world would be a confusing place.
As objects moved in or out of shadows or we went from indoors to
outdoors, the colours of everything would otherwise
appear to be continually changing.
The best reason for having evolved it would be for survival. We needed
to perceive the tiger as it moved in and out of the shadows as being the
same object. We see its colours as constant despite vastly changing
illumination conditions. We can track the threat. Or stalk our own prey
on the savanna or in the jungle.
The shrimp need to do the same. In the oceans, they have a large range of
predators and with varying depths of seas water at varying times of the
day or night, on both sunny and cloudy days, they need to be able to
perceive an object as having the same colour constancy.
Reading the paper, my hunch was right. If I am interpreting the authors
correctly, they suggested two contending theories. One, shrimp needed such
a multiplicity of receptors to be able to perceive and discriminate between
a larger range of colours, because they are in coral reefs or what-not.
Or two, they don't need that at all, just the ability to achieve colour
constancy.
My first hunch was based on the fact that compared to us, the shrimp
have very small brains. As I mentioned, the perception of colours and
colour constancy in humans is largely a function of what goes on in the
brain rather than what takes place in the eye.
The shrimp needed to evolve the trick of colour constancy in a different way.
Hence they appear to use the multiplicity of receptors in combination
with a poor-man's (poor prawns?) temporal eye-scanning technique to
perceive colours. What one might need to do lacking a big brain.
Humans perform something analogous to a fast Fourier transform (FFT)
in the back of our heads to see colour and achieve colour constancy.
The authors suggest the shrimp may use a cruder, serial technique similar to
what is known as a "push-broom" scanner, often deployed by engineers in
remote imaging satellites. Since the satellite is moving, it can scan and process
like the moving part of a photocopier.
The important part - the trick if you like - of the processing is comparing
between the relative levels of the receptors rather than mixing the absolute levels together.
This is completely different to how cameras work. In cameras, the red channel
is for red, the green for green and so forth. We don't perceive colour that way.
The authors point out that experiments with the shrimp appear to bear the
colour constancy theory out, namely that they can't discriminate between closely
positioned wavelengths and for that matter, perceive many more colours,
despite the richness of overlapping receptors.
Instead, it appears they may simply use them to achieve colour constancy
over a relatively small number of possible colours.
Because we all know how colour cameras work, it is a common misconception
that colour processing in humans is analogous. The misconception is probably
widespread with school teachers and photographers alike.
Thanks again for the post which I enjoyed.
Best Regards
Gary
