Re: Direct Perception vs. Representationalism
Peter Main
1. Brain computation is fast relative to your reaction time because of its high degree of parallelism.
2. The brain uses a number of different methods to compute depth. Many of them (e.g. occlusion, color change with distance, parallax) are not particularly sensitive to focus.
3. Color vision is poor in the visual periphery, due to paucity of cones there, so perhaps we simply don't notice color fringes arising from chromatic aberration.
Peter Main
At 15:57 01-05-05 -0500, NWR wrote:
>
> Experiment 1:
>
> I look out my window, covering one eye with my hand. I see the crab
> apple tree, in bloom. But it appears rather flat. Then I move my
> hand to uncover the eye. It seems that I am immediately aware of a
> more glorious tree, due to the depth perception now available. The
> change seems instantaneous, indicating that it takes only a fraction
> of a second. Given the relatively slow response timings of neurons,
> there isn't enough time for a complex computation to be done.
>
> Experiment 2:
>
> I again look out the window, with one eye covered. But this time I
> am wearing my reading glasses. They focus to a distance of around 16
> inches from my eyes. As a result, the crab apple looks somewhat
> blurry. I again uncover the eye, and the same depth perception is
> immediately available, although what I see is still just as blurry.
>
> If depth perception is based on static analysis of retinal images,
> the blurriness of those images should reduce the ability to perceive
> depth. Yet it didn't appear to have much of an effect. Why not?
>
> Thought experiment:
>
> A good camera has a compound lens. This is to reduce the various
> distortions caused by a simple lens. The eye, by contrast, uses a
> relatively simple lens. Thus we would expect a variety of
> distortions. One of these would be chromatic aberration. It shows
> up as colored fringes surrounding the objects that are focussed by
> the lens.
>
> Maybe there is somebody reading this list who knows whether there is
> chromatic aberration with the eye's lens. If so, I would be interested
> to hear about it.
>
> Assuming that there is chromatic aberration, then images formed on
> the retina should have colored fringes. If we are literally seeing
> the images on our retinas, then we should be seeing colored fringes.
> But we don't see any sign of such chromatic aberration.
>
> -NWR
>
>
Peter Main
First I would like to clear the decks of a very simple misconception. Surely no one thinks that vision is a matter of literally "seeing" representations in the head, for that would be the very *pons asinorum* of the matter - confusing a representation for what it represents. So I hope it was not necessary for Brook to remark that "all is pitch-black in the brain" - we are all well aware of that! Whatever it may be that we see, it certainly is not any part of the brain.
"Representation" is a word that I feel invites muddle and misconception. Let me explain. Representation is a three-part relation - the representation itself (the signifier), whatever is represented (the signified), and the user (or observer, if you prefer) of the representation. All three entities are necessary. But when "representation" is used in the context of the brain, it either implies some sort of homuncular observer (clearly erroneous), or one is left in mystery as to who or what the user of the representation might be.
Serious though that is (and a great deal more could be said about it), it is not the error I am concerned with here. Which is this: to speak of the end state of the computational process as a representation is immediately to raise the question "a representation of what", to which the answer is unavoidably, "why, the outside world, of course - what else could it be?" This, I think, is Brook's view: vision is perception of the external world, mediated by a computational process which culminates in a brain state he has called a representation. If this be granted it becomes a minor matter whether it is called direct or indirect realism.
By avoiding the word "representation", as I have tried to do, we can avoid those pitfalls. There can hardly be any doubt that visual phenomena supervene on brain state/s that are the causal result of cognitive (or computational) brain processes. And that those processes have been tuned by evolution to generate brain states leading (via conscious experience) to pro-survival behavior. But that does not mean, except in a loose way of speaking, that we see the external world. What it does mean is that our visual experience arises from brain activity, much as it does in dreams, in imagination, and in other cases where there is no external world involved.
If Andrew still feels that all this is mere assertion, without argument, then I can do no more.
Peter Main
At 15:07 01-05-05 -0400, Andrew Brook wrote:
> Same old error, same old utter absence of argument for it. "Computational
> process that must lie between the physical world and our visual experience" --
> no, the computational process lies between the world and our *representation*,
> the one that gives us our awareness of world. As I have said repeatedly, one
> is (typically) *not* aware of the end-state of the computational process (and
> certainly one does not see it: all is pitch-black in the brain) and one
> typically *is* aware of what kicked the process off, a causal (in the case of
> vision, light wave) input from the world. Now, if you don't like this reading,
> fine. Then *argue* against it, don't simply assume on the basis of an
> unsupported (mis)reading of signal processing that it *just has to be* wrong
> It does not just have to be wrong -- and might even be right, depending on how
> direct our awareness of world turns out to be. (My view: as direct as could
> be, by any concept of 'directness' that can be articulated.)
>
> Put another way, it is arguments that impress me, not repetitions of
> centuries-old misreadings of the implications of the fact that visual
> information is processed in the brain (as opposed to where?).
>
> Andrew
>
> --
>
> Andrew Brook, Professor of Philosophy
> Director, Institute of Cognitive Science
> Member, Canadian Psychoanalytic Society
> 2217 Dunton Tower, Carleton University
> Ottawa ON, Canada K1S 5B6
> Ph: 613 520-3597
> Fax: 613 520-3985
> Web: www.carleton.ca/~abrook
>
>
Alex Green
Best Wishes
Alex Green
Send instant messages to your online friends http://uk.messenger.yahoo.com
Sue Pockett
couple of seconds for the blood flow to an active bit of brain to
increase. Thus the time resolution of the method is really not
conducive to the sort of experiment you refer to.
Andrew Brook
> By avoiding the word "representation", as I have tried to do, we can avoid
those pitfalls. There can hardly be any doubt that visual phenomena
supervene
on brain state/s that are the causal result of cognitive (or
computational)processes. And that those processes have been tuned by
evolution to
generate brain states leading (via conscious experience) to pro-survival
behavior. But that does not mean, except in a loose way of speaking,
that we
see the external world.
Brook: It does not entail that we don't, either.
What it does mean is that our visual experience arises from brain activity,
much as it does in dreams, in imagination, and in other cases where
there is
no external world involved.
Of course there is are complex processes going on. I certainly grant
that some don't make us aware of the world around us. From this it does
not follow that none does.
> If Andrew still feels that all this is mere assertion, without argument,
then I can do no more.
That is too bad -- you have a good grip on what I am trying to say.
However, that X does not mean Y does not entail that X is incompatible
with Y or that Y is false.