Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Path: utzoo!mnetor!seismo!columbia!caip!nike!ucbcad!ucbvax!sdcsvax!sdcc6!sdcc3!ln63szf From: ln63szf@sdcc3.ucsd.EDU (Rick Frey) Newsgroups: net.ai,net.philosophy,net.cog-eng Subject: Re: Gibson's theory of perception Message-ID: <3490@sdcc3.ucsd.EDU> Date: Thu, 24-Jul-86 19:07:17 EDT Article-I.D.: sdcc3.3490 Posted: Thu Jul 24 19:07:17 1986 Date-Received: Fri, 25-Jul-86 08:01:21 EDT References: <1782@mtuxo.UUCP> <3483@sdcc3.ucsd.EDU> <14987@ucbvax.BERKELEY.EDU> Organization: U.C. San Diego, Academic Computer Center Lines: 145 Xref: mnetor net.ai:1066 net.philosophy:2200 net.cog-eng:242 >> Rick Frey > Michael Ellis > Major mistake? Simply isn't true? Does anyone understand the > relevance of this to Adam's original statement? As far as I can > tell, you have simply refused to acknowledge a difference between > your word "perceive" and Gibson's. Ok, maybe I overstated it a bit. But, when people start playing with words and simply assume that because they feel a certain way they can redefine a word, that's not correct. Perceive is a word with a definition and Gibson's idea of perception doesn't fit the known facts. What more do you want me to say? > >> The information contained in this > >> awareness is the original, and not a re- (or transformed), presentation > >> of the object to consciousness. > >What is awareness? What stage of processing are > >you referring to? Aside from the actual collection of excited and > >inhibited receptors at the retinal level, *all* perception is transfor- > >mation. Sorry, no homunculae. (sp??) > My, we're being doctinaire! No awareness, no homunculae? All > perception is transformation? Are you really sure?? > Are you telling us that there is nothing in the real world that > corresponds to "awareness"? How quaintly Cartesian! Yeah, I feel pretty comfortable saying that all perception is transformation. And be somewhat careful, I never said there was no such thing as awareness, I simply pointed out that it didn't have a clear definition. What do you see in the real world that corresponds to awareness? > Homunculism is by no means dead. Dennett in "Brainstorms" speaks > of mental processes in terms of progressively stupider homunculae. And Don Norman refers to Cognitive demons that make decisions for us, but the idea of the homunculus is that he's sitting in there looking at a television screen that identically corresponds to what we see. This idea of the homunculus is wrong. The little Homunculae you refer to are looking at excited and inhibitted neurons in specific locations. Not quite a tv screen. > As to awareness itself, are you sure you really wish to overlook > the concept? Is all we wish to know about perception simply how a > mess of brain cells work? Or is it not at the same time to understand > how it is that we become AWARE of our environment? I don't want to overlook it at all. But how do we define it? Obviously perception is more than the excitement of neurons, but as far as we can tell, even awareness seems somehow to be linked to a neural state. If you can define awareness, I'd love to know exactly what it is. > >In some cases, this is a valid claim of Gibson's, but his whole theory > >rests on it and it isn't 100% true. He'd love to have us believe that > >the external world is rich in information and that perception is an easy > >task of picking out of the multitude of cues available. That's simply > >not true. > Must the Gibsonians be 100% right lest their entire case come > crashing to the ground? Who says so? The anti-Gibsonians? Why? Nope, he doesn't need to be 100% right. But, since he isn't on a major premise, you have to wonder about his whole theory (that is based on that premise). > And if Gibson would "love to have us believe the external world is > rich in information", whatever do you find hard to believe about > that? Do you mean to imply that the world is poor in its > information content? As far a I can tell, there IS an enormous > wealth of unused information in the natural world. There is an enormous wealth of information, true. But, there is not an enormous wealth of useful information. There's a big difference. > I did not know Gibson's "theory" was either true or false; rather, > I thought that it was more unified approach, a redefinition > of just what it is our perceptions are, But can't general redefinitions be wrong? Obviously he's got some quantity of truth about what he's saying, but in general, his redefinition is off-track. > the wider view of evolution, not of just species, but of > interdependent systems ranging from genes to ecologies. If so, > there doesn't seem very much to be "wrong" about it, at least none > of the criticisms you've offered. But it doesn't make sense in terms of evolution. It works well with identifying faces and familiar objects. That is a far cry from the task of locating predators and prey in a natural environment. > >Perception developed under situations > >where the available information was far from over-abundant. Without > >this over-abundance of information, Gibson's ideas start falling apart. > Totally absurd. Life evolved in the presence of light, which > contained vast amounts of environmental information that was > almost completely unused until creatures evolved that were > complex enough to exploit vision. There's a lot of evolution from light sensitive cells to the eye of a frog, for example. Keep thinking about the task of picking out a cameleon on a tree branch. Do you really have an overabundance of information? Cameleon like characteristics aren't that rare in nature. > Or are you denying that light carries information? Of course not. But it doesn't carry with it how to make sense out of itself. Lots of information, not all of it useful. > >Perception developed in a piecemeal fashion such > >that there wasn't an overabundance of information at various stages in > >the development. So where did this 'perceptual laziness' come into > >play? You try finding a small tree frog in a huge tree. The whole idea > >of an overabundance of perceptual information might apply to recognizing > >people's faces, but it has no basis in the reality of where and how > >perception developed. > How different is recognizing people's faces (for beings who need > to do so, like other people) from recognizing the kinds of places > where frogs hang out (for beings who need to eat frogs)? People's faces aren't typically presented on a background of identical color. People's faces usually don't match the texture of whatever is behind them. People's faces present clear occlusions of other objects such that depth information is pretty clear. Keep thinking about a bumpy green tree frog on a green leaf. Where does the leaf end? A bumpy leaf? Another leaf behind? How do you know? You don't have all day to carefully search each leaf and making a mistake can be fatal. Tree snakes are green, too. > Now my perceptions did not evolve so that I could locate frogs > more readily than many other kinds of things, but if humans had > evolved into tree climbers with a diet consisting largely of > frogs, I have little doubt that we'd have no trouble locating > that frog, even blindfolded. Well tell that to a frog that'll starve to death in a cage full of dead flies because he can't *see* them unless they're moving. An overabundance of information, but little of it actually being useful. Rick Frey (...ihnp4!sdcsvax!sdcc3!ln63szf) (ln63szf%sdcc3@sdcsvax.UCSD.EDU or .ARPA)