From what I have read the human eye is a rather poor optical instrument and relies on the brain to do a lot of ‘post-processing’ of images in order for them to be interpreted correctly, perhaps to the point where the brain infers certain things based on previous knowledge of the world that the eye may not have picked up.
So, first off all, to what extent is the above true? Secondly, are there any images of what the eye sees available, e.g. somehow linking a camera to the optic nerve to capture a live stream.
Is what we consider our vision a live feed from the eyes or is it the ‘photoshopped’ version that our brain has cleaned up for us?
The brain isn’t all there is, our system of vision is a combination of the eye and the brain.
For example, right now I can see my laptop, but i’m not seeing the laptop as my eye is seeing it, i’m seeing it after the brain has processed the image. So yes, the brain is kind of like a filter between the raw data captured by the eye and how we perceive the world. So, is there any way we can capture the raw data from an eye and what would that look like?
One deficiency of the eye that we’re not consciously aware of is that it’s only high-resolution in a small area near the center of the field of view: Most of the eye isn’t sharp enough to do things like reading normal-sized text, for instance. When you read, you’re quickly scanning your eyes across the page, so that small high-detail spot passes over all of the text. This is one reason why photographic memory speed-reading, which some folks claim to be able to do, is clear bunk: Even if your brain could store information that quickly (maybe it can; it’s hard to tell), your eyes just aren’t good enough to pick it all up at once.
No. “You” are your brain in this context. There’s no little homunculus in there watching what the brain produces.
But I think the question you’re really trying to get at is: If you used the human eye as the lens system and the human retina as the CCD array of a digital camera, what would the picture look like? Yes?
It’s an interesting question, and a quick Bing search doesn’t find any. Certainly it would be very blurry except right in the center.
Yes I know what you are saying TimeWinder, I know that I ‘am’ my brain, it’s just quite hard to get the idea across in words. If I described what I see as being what my brain ‘sees’ it would have sounded even stranger! For the purposes of the question it’s easier to think of the vision system being something like:
Eye -(1)-> Brain -(2)-> Vision/Final Image
What I am looking for is an idea of what the image looks like if it were captured at point (1) between the eye and the brain.
What the eye sees, and what the brain proceses are two different things.
Take colour for example. what The brain thinks it sees isn’t always the same. If you look at a red car, Parts of it may be highlighted, and you see bright orange or even white, other parts are in shadow, and you see dark brown, but the brain processes the image and sees it as red.
The eye, as people pointed out, is not a camera. It is a sense organ and was designed to be efficient and fast.
This means your eye cannot work without the brain and other inputs. For example, look to your left. Now, turn your head and look to your right. Now, take a camera, do the same thing, and watch the film. When you turned your head, your brain was able to understand that your head was moving, and stabilized what you saw. When you film the same thing in the camera, your brain didn’t detect the movement, and didn’t stabilize the image. Instead, you saw it quickly shift.
Your red receptor in your eye is actually pretty close to greenish-yellow in sensitivity. But, that’s enough to let you see red (which most mammals cannot do).
The center of vision is the most sensitive and the color sensing codes are more concentrated there. Fewer receptors are placed in your periphery vision. However, if something does capture your attention off to the side, you’ll move your eyes in that direction, or turn your head. Your brain will piece together a complete picture.
So, as a pure camera, our eyes have a lot to be desired. However, the entire body is used to help track what your eyes see, and your brain processes it into something that few cameras can capture.
Have you ever seen something you thought was amazing, took a picture, and this amazing thing is almost impossible to see in the picture? You stand in front of Notre dame and are in awe of its size and the image of this old building with the gargoyles on the sides. You take a picture, but the picture doesn’t show an imposing building, and the gargoyles barely show.
That’s because your brain builds a map of your environment, you see the entire church although it is impossible for you to get the entire church in your field of vision. Instead, as you move your head, your brain builds the image. You see the fine details while taking in the whole structure. Our vision isn’t a map of exactly what the photons tell us. Instead, the brain maps it, adds emotional context, highlights the details, and gives us a better image that what a mere photo could show.
I think that’s a bit oversimplified. In point of fact, information from the eyes go through multiple levels of processing before reaching the parts of the brain associated with conscious thought. Patterns and contrast are enhanced, color is added as a separate stream, etc. So it’s quite valid to say that images are processed by the brain before reaching “us”.
You say “deficiency”, I say “optimization feature”. The lower fidelity on the rod cells on the outer radius of the retina permit faster signal interpretation, i.e. permitting the brain to recognize patterns in low light and peripheral regions of vision, whereas the more detailed and coherent images formed by images that impinge directly upon the center of the retina require more processing. Recalling that the eye was evolved not for reading or other modern tasks, but rather to provide protective senses at distance and allow for multi-mode feedback for manipulation, the human eye can be regarded as a marvel of optimization for what it is evolved to do. Its greatest deficiency is in the mechanical structure; that is, as an adjustable pinhole camera rather than the translating camera eyes of the more sophisticated cephalopods. (<Homer Smpson>Stupid octopuses! They’re better as us than everything!</Homer Simpson>)
As for the division between the eye and the brain, there is no clear delineation in the vertebrate eye, at least insofar as the as the retina (the receptor) goes, as it is actually an outgrowth of brain tissue containing direct synaptic connections to brain cells. While human vision is still only poorly understood, processing occurs in several cortices and there are indications that some “preprocessing” of image data occurs even in the retina, roughly analogous to packet-switching in digital network theory, in order to avoid overloading the areas of the brain that process the information into cognitive images. What you recognize cognitively is a highly processed and interpreted image, but then, what the retina-brain system actually sees is a bunch of nerve impulses formed by the impingment of photons on different retinal cells that register different spectra and types of optical sensation (brightness, directional movement, et cetera). At no time does any part of the system see what can be extracted as an “image” in the way that information encoded on a helical scan videotape or digitally in a MP4 file is an image composed of discrete but common elements; the information is far more complex, comes in at different rates, and is composited together in a way unlike simple video images.
It’s not a simple answer, but that’s just the way it is, as with everything else involved in cognition.
No. That is worse than oversimplified, it is quite wrong, and what Alessan says (so far as it goes) is right. There are no “parts of the brain” where consciousness happens. That idea is just a hangover of the old picture (due to Descartes) where the brain just served to relay visual information through to an immaterial (and so scientifically inexplicable) conscious soul sitting at the center.
Of course, mittu’s original question, arises from the same misconception, as shown by the fact that s/he diagrams the visual system thus:
That is a nice illustration of just how how you should NOT think about the visual system if you are actually to have any hope of understanding it (despite the fact that it still gets diagrammed that way in many introductory textbooks:(). It is wrong to think of the eye transmitting an image (as opposed to information) to the brain, and it is also wrong to think of it transmitting information to a terminus in the brain somewhere, and that being the end of the process of seeing.
A much less misleading, (but still highly simplified) diagram would look like this:
[World] –-> [Retinal image in eye] –-> [Brain visual processing system] –-> [Brain action control systems] –-> [Muscles]
With then, crucially, an arrow looping back from [Muscles] to [Retinal image in eye] (which I can’t draw in ASCII art through the Dope’s formatting system).
This is crucial because your eyes are constantly moving about changing that retinal image so as to extract more visual information from the scene (Stranger has pointed out some of the reasons why it needs to do that). Of course, our muscles can also change the actual configuration of the external world itself, so an arrow should probably also loop back to there (and there are actually various other feedback loops within the larger one diagrammed here).
The only real image in the whole system is the retinal image, and that is never stable. The optic nerve does not function to transmit an image to the brain, like a the cable from a TV camera, but information. This information does not include everything you would need to reconstruct the momentary retinal image down the line (but that does not matter, because it is not what the visual system is trying to do) but it does include other things, such as locations of edges (in the world, not the retinal image) and information about relative movement of objects. This sort of information is derived partly from computations performed by neurons within the retina, and partly from registration of the way the retinal image changes as the eye moves.
Conscious visual experience is a property of the whole system, not of something inside the brain.
You will notice (or rather, you didn’t notice) that I said For the purposes of the question it’s easier to think of the vision system being something like. I don’t think that is how the visual system works at all, but for the purposes of clarifying my question it is easier to explain it that way.
The eye acts as a lens which (amongst other things) focuses an inverted image onto the retina, what if we took a disembodied (or otherwise) eye and replaced the retina with a CCD, what image would be produced?
Well, most people do think it works that way, you know. You would not be a fool for thinking so, just wrong. Maybe you did know better all along, but, if so, it is odd that you should ask your question at all, and a pity you should contribute to the general ignorance by appearing to endorse this misconception.
That is a very different question from what you originally appeared to asking, and, frankly, a pretty uninteresting one (which is why I am still pretty sure that it was not what you originally intended). Indeed, you have already gone most of the way towards answering it yourself. As you rightly say, and as is well known, the eye acts as a lens which focuses an inverted image onto the retina. If you replaced the retina with a CCD similar to one you might find in a digital camera (presumably that is what you mean, now), you would get much the same sort of output that you get from the CCD of a digital camera (except that it would not be so well, or so evenly focused). What else would you expect? Digital camera CCDs (unlike retinas) are designed such that their output can easily be reconstructed into an image that a person can look at.
Are you seriously telling me that was what you were asking about? I think you already knew that. What I am telling you, however, is that answering it is worse than useless as a step towards understanding how vision works, both because the retina’s output is not analogous to that from a CCD, and (more importantly) because the retina’s function within the visual system as a whole is not analogous (in more than a trivial way) to the function of a CCD within a digital camera. Unless one is very explicit about this, the question and its answer will lead one away from rather than towards a proper understanding of vision.
A less technological version of this “experiment” was done some 300 years ago, by scraping away the opaque layers at the back of a dissected-out cow’s eye, and replacing them with a piece of translucent parchment. It was then possible to see the inverted optical image through the parchment. This experiment was described first by Christoph Scheiner, and later, in a more widely read version, by Descartes. It has been misleading people about the true mechanisms of vision almost ever since.
Incidentally, if an eye with its CCD retina were otherwise left in its natural position in a living human’s face, hooked up to all the musculature of a normal eyeball (is that what you were thinking? I doubt it, but never mind), then, if the output from the CCD were fed to a screen, the image on the screen would be jittering about all over the place, because your eyes are constantly on the move, and would probably be very hard to interpret, much like trying to make sense of a movie made with a hand-held camera by someone who is dead drunk (or maybe, even better, seriously zonked on speed). This again, illustrates the fact that the function of the eye is not to get images (or information from which images can be reconstructed) into the brain.
If you are really seeking information about the defects of the eye as an optical instrument (most of which arise from the structure of the retina itself, and would thus be eliminated by replacing it with a CCD), you could try looking here (PDF).
Indeed. Why do you waste our time, sirrah? Presuming you are serious, you should be perfectly capable of doing your own research and not coming here to bother us at all. Do you think your prattling questions interest us, or that we do not have better ways to spend our time?
What the hell is your issue? I spent a lot of time and effort answering mittu’s original question is considerable detail. It so happens that in order to do so meaningfully it was necessary to explain how the question was misconceived (in a very common, but very significant way). Then mittu comes back and says, in effect,
“hey you jerk, I never misconceive anything, how dare you suggest it? There were weasel words in my second attempt o formulate my question which get me off the hook of ever having to own up to any misconceptions. I never meant to ask the interesting and significant (if misconceived) question I actually asked. I actually meant to ask this other very dull question, to which I already really know the answer.”
If I let my irritation show a bit, when I gave another detailed answer to that, I do not apologize. mittu’s response was churlish. (Though not as downright nasty as you are being, at least I do do something to offend mittu’s amore propre in pointing out the original misconception. You have no such excuse.) And I certainly do not apologize for bothering to take the time to find and provide a link to an article with detailed information relevant to both versions of the question. What? Am i supposed to precis it for mittu’s benefit with all the necessary diagrams redrawn? (No doubt to be told, “Oh no, that article would imply that I have been laboring under a misconception about the workings of vision, so that clearly does not answer what I really meant to ask!”)
A lot of signal processing is already going on in the retina of the eye. First of all, due to a complex system of molecular interactions that constantly tune the sensitivity of the light receptor system in each photoreceptor cell., The individual light sensor cells do not so much respond to the static level of illumination, but rather to changes in the light intensity. Then, several layers of neurons in the retina compare the signal from adjacent photosensor cells, leading to some neurons only firing if motion in a certain direction is detected, or a certain level of contrast. Therefore,the signal that reaches the brain from the eye is very different from the pixel-by-pixel representation of a digital camera. A lot of contrast enhancement, color normalization,motion detection and so on has happened before the signal even reaches the brain. Then, in a specialized structure of the brain, the signals received from the two eyes are compared to extract 3D- information before the signals are passed on to the visual cortex for further processing.