Perception

[ameerat42]

Here's a 10-and-a-half-minute video I found in a link somewhere else.
It's here because it has implications for photography. The first bit was
(for me) surprising, and the rest may be of interest...

Now look here... (https://www.youtube.com/watch?v=_FlV6pgwlrk)

[ameerat42]

How are your eyes, G:crzy:ZZA?

[Gazza]

How are your eyes, G:crzy:ZZA?Havta re-think what we see/saw

[Steve Axford]

Eyes are very smart (maybe that should be - visual processing is vey smart). My left eye had a blood clot and part of the retina died. If I look at the page of a book with just my left eye, some of the letters or words are completely missing. But, if I look at a tree or a face or most things, I see them perfectly well. My brain just fills in the missing bits.

[John King]

Steve, sorry to hear that. However, we can all only see the blind spot in each eye by tricking the eye/brain system into revealing it ...

A lot of simple visual processing actually takes place in the bipolar and amacrine cells in the retina before the signals leave via the optic nerves for the optic chiasm and onwards from there to the visual cortex.

Also, each eye is in fact two half eyes ... The retina is in two halves, and the right and left visual field from each eye are combined with its other half from the other eye in the optic chiasm.

Retinal tissue is composed of neurons, i.e. brain cells, and the eye and brain have to be considered as a single functional unit in order to understand them.

Then there is the distinction between wavelength and colour. These are processed by two separate nuclei in the brain. Loss of the colour nucleus leads to complete loss of colour vision. However, loss of the wavelength nucleus causes vision to become bizarre.

We are complex little beasties!

BTW, the creature with the best colour vision that we know of so far is the Mantis Shrimp! Squid and Octopi are pretty darn good. Humans and other mammals come a fair way down the list ... ;).

@Am - interesting video, mate :).

[Steve Axford]

Do you mean wavelength and colour or brightness and colour? We pay very little attention to wavelength, even though we pretend to relate it to colour.

Thanks John. I can still see well enough for photography and it has taught me that what we see ain't necessarily there. It's just our approximation of reality, and that can be nothing like it on occasions.

[John King]

Wavelength and colour, Steve.

Brightness and the apparent dynamic range of our visual system is a whole other can of worms!

[Steve Axford]

How do our eyes see wavelength apart from through our rod cells that are the way we see colour? We only get a very rough idea of wavelength as, for example, a blue and yellow wavelength are indistinguishable from a green wavelength, and a red and a blue wavelength appear the same as a purple wavelength.

[John King]

The two nuclei interpret the wavelength and colour information. They are separate from the visual cortex, and only contain a few hundred neurons each. It appears from brain trauma studies that the interpretation of wavelength takes precedence over that of colour. Loss of the colour nucleus leads to normal perception, but in shades of grey. Loss of the colour nucleus leads to the loss of meaningful vision.

For something that's so ubiquitously useful, colour doesn't actually exist!
It is an almost universal property that is a mental construct from the way eyes detect wavelength, and how the brain processes and interprets this info ...

As an example, what colour is UV? Being one of the 10% or so of humanity who sees into this part of the spectrum, it appears sort of pinkish/mauve to me. An unfortunate consequence of seeing this way, I must wear full UV cut glasses, even at night (HID headlights are pretty painful ... ). If I'm driving a long distance, I wear pink night driving glasses.

Vision was one of my great interests when I did my degree in psychology (along with a few other subjects ... it's a long story, and a longer degree!). Specifically, how reception becomes perception. The study of visual illusions informs this understanding. It remains a great interest to this day, along with how a piece of wetware (a brain) becomes a conscious, self-aware persona, whether in a human or an ant ...

Some interesting stuff about the mantis shrimp is here:

http://theconversation.com/mantis-shrimp-have-the-worlds-best-eyes-but-why-17577

On reading that article more carefully, much of the information is wrong - e.g. they have six colour receptor chemicals, not 16 ...
They are still fascinating creatures. The Wikipedia article is more accurate (on yer, Jimmy and friends), here:

https://en.wikipedia.org/wiki/Mantis_shrimp#Suggested_advantages_of_visual_system

[ameerat42]

...An unfortunate consequence of seeing this way, I must ... wear pink night driving glasses...

Wha-a-a-t?! Forced to wear rose-coloured glasses:p

(Ta for the link, too.)

[John King]

LOL, and you're welcome :).

[Steve Axford]

I too have been fascinated by colour and what you say makes little sense to me. Do you have any references on wavelength and vision? My understanding is that we only have a very vague idea of wavelength. We call it colour, and yes it is a construct, but it serves a very useful purpose.
Mantid shrimps have lots of colour sensors, but are since we can't ask what they see, we can only guess.

[John King]

Steve, Richard Gregory's "Eye and Brain" used to be a great once-over-lightly, but I haven't looked to see if anything equivalent is currently available. Here:

http://press.princeton.edu/titles/6141.html

Interestingly, one of my lecturers in Physiological Psychology was a Dr Michael Wood - probably just a coincidence!

Knowing the colour receptor chemicals present in the retina of a creature allows analysis of what it is possible for it to see. That research has revealed some fascinating information about this fascinating creature.

[ameerat42]

Awl wright! I'm starting to come up with a formulation now...

Perception = 10% Reception + 90% Deception +/- Delusional Constant*

:confused013:nod::D

* The Delusional Constant is akin to Einstein's Cosmological Constant, the delusion arising from
evidence in an article I read today that it is changing.

[Steve Axford]

Steve, Richard Gregory's "Eye and Brain" used to be a great once-over-lightly, but I haven't looked to see if anything equivalent is currently available. Here:

http://press.princeton.edu/titles/6141.html

Interestingly, one of my lecturers in Physiological Psychology was a Dr Michael Wood - probably just a coincidence!

Knowing the colour receptor chemicals present in the retina of a creature allows analysis of what it is possible for it to see. That research has revealed some fascinating information about this fascinating creature.

Hi John, There has been a lot of hard science in analysing visual perception in recent years using such tools as PET scans to trace where signals are processed in our brains. But, wavelength is not something that is processed by our brain. I'm not sure how you got this idea.The book you reference has chapters on brightness, movement and colour, not wavelength.

[John King]

Haha.

Let's not diverge further into Einsteinian physics ... ;) :nod: ... or his cosmological constant; or the big bang theory; or the "standard model" ...

How to mix a Universe:

1) Take lots of Hydrogen (a Universe full ... )

2) place Bamix into mixture for no longer than 10 seconds (or 10^-47s, if you prefer ... )

3) Place in oven to bake for no more than 10 minutes

4) Ignore all evidence about plasmas, gravity not being a force but the topology of spacetime, 'tired' light, etc

5) Present Universe as whatever you want it to be

6) Explain away all inconsistencies by super string theory, multi-verses, inflation, or your choice of astrological happen-stances ...

7) Shout down non-believers (like engineers, observational astronomers, and their ilk ... )

Done ...

At least science is (generally) self-correcting, even if it does take hundreds of years sometimes!

- - - Updated - - -


Hi John, There has been a lot of hard science in analysing visual perception in recent years using such tools as PET scans to trace where signals are processed in our brains. But, wavelength is not something that is processed by our brain. I'm not sure how you got this idea.The book you reference has chapters on brightness, movement and colour, not wavelength.

Steve, a question.

What stimulates the rods and cones to change the chemical structure of the photoreceptor chemicals in them?
Is it colour?
Or is it the specific wavelength of light to which the particular photosensitive chemical responds to?
I no longer have my old copy of Gregory. Perhaps I should buy a more modern copy rather than relying on my (rapidly deteriorating) memory from nearly 40 years ago ... :nod:.

- - - Updated - - -

[Edit]

Steve, the chapter titles in Jimmy Watson's "Molecular Biology of the Gene" don't tend to give anything more than a vague idea of the content either ... :nod:.

[Steve Axford]

Wavelength certainly has a relationship with rods but they measure what we think of as colour, not really wavelength. 40 years is a long time in the biological sciences, and even more in neuro- psychology.

[John King]

Yes, 40 years is a long time. However, I have maintained my knowledge over that time ...

Colour is a function of our perception of wavelength, not the other way around.

[Steve Axford]

Whoops, I meant cone cells (the ones responsible for colour vision), not rod cells which are responsible for dim light vision. We obviously both need to keep reading. My apologies.

My point is that we process colour not wavelength. Our cone cells each contain one of 3 photoreceptor pigments that react to light. One reacts most strongly to light with a wavelength of 564–580 nm (red), one 534–545 nm (green), and one 420–440 nm (blue). They do not measure wavelength, but just react most strongly to a that wavelength, but also react to a lesser extent to other wavelengths. Our brain interprets those 3 signals to to represent colour. We do not detect anything like wavelength as it would make no sense. I'll give you an example. Imagine we see a light source that contains 2 wavelengths, one blue (say 450nm) and one yellow (say 570nm). What we see is green. Yet the wavelength is clearly not green. If we were shown light with a single wavelength of 510nm we would also see green. It is worth noting that almost all green birds are really yellow + blue birds.

The point being that our eyes do not see wavelength. If you ever use a spectrometer, then you will get wavelength information. You will see each component of the light as a separate wavelength and intensity. If we wanted to make a spectrometer, we would not choose the way our eyes are made as we could not get very much useful information on wavelength. On the other hand, when interpreting the real world, wavelength is at best confusing. Colour, even though it is a construct, is much better.

It occurred to me that maybe the confusion about wavelength comes from what is called colour constancy. This is where we try to keep colours constant in differing circumstances. For example we try to see the same objects as the same colour irrespective of the lighting conditions. This may be an apple which we know is red. If we see it under normal light we will see it as red. If we see it in predominantly green light we will also see it as red, yet when the light from the apple is measured it may be radically different from the red we saw under normal light. Our cone cells should also see it as another colour, but our visual system will adjust the colour depending on the surrounds. Perhaps this is where you got the wavelength confusion from.

Cameras are a very good analogy for our eyes. They have red, green and blue sensors, just like our cone cells. They also apply a white balance to the scene to adjust for colour constancy factors (not as well as we do, but really very good). Cameras, like eyes, do not measure wavelength. They react to colour.

[ameerat42]

Steve. How does what you say account for "colour blindness"?

Colour has to be a construct, IMO, to account for that.

Secondly, for the camera-eye analogy, the concentration of cones and rods (AFAIK) is in the
fovea.

[Steve Axford]

Colour blindness comes in various forms. The most common is where the red cones are missing and the person only sees blue and green. They cannot see the colours that contain red. Green or blue cones can also be missing giving rise to other forms of colour blindness. It has little to do with our construction of colour except that the signals from those cone cells are transmitted to our brain in order to construct colours. Missing signals lead to missing colours.
Complete lack of cone cells, ie fully colour blind, is very rare but does occur and is quite debilitating as just having rod cells makes us far to sensitive to light. Oliver Sachs wrote a book on this called "The Island of the Colour Blind".

Cones are concentrated in the fovea, rods predominate in the periphery of our eyes.

[ameerat42]

...or, The Evolution Of Walking - Why Kangaroos Don't Use Zebra X-ings:

(Please report missing :url: (http://www.michaelbach.de/ot/mot-feetLin/index.html))

Or just interesting :D

[Steve Axford]

Very very cool. I haven't seen that one.
The information contained in our visual field is enormous and our brains could not make sense of it if we tried to process it all. So we only attempt to process that which is of use to us and throw away the rest. It is an approximation of the reality that is important to us. Our brain uses many tricks (rules of thumb?) so as to avoid huge computation tasks that would waste time and energy. Of course approximations are sometimes wrong, hence those optical illusions.
Just think of the processing problems associated with crossing the road. We have to recognise where the road is, what the cars are, when the cars will arrive at our crossing point and when they won't. Then we have to estimate how long will it take us to cross the road and will a car arrive before we do that. With practice we can do that easily - and while talking and chewing gum at the same time. I am always amazed at how few people are killed crossing the road or driving cars.