Wednesday, November 24, 2010

Gregory on Downplaying the Mystery of Inverted Vision

We now think of the brain as representing, rather as the symbols of language represent characteristics of things, although the shapes and sounds of language are quite different from whatever is being represented. Language requires rules of grammar (syntax), and meanings of symbols (semantics). Both seem necessary for processes of vision; though its syntax and semantics are implicit, to be discovered by experiment.
     Some puzzles of vision disappear with a little thought. It is no special problem that the eyes' images are upside down and optically right-left reversed-for they are not seen, as pictures, by an inner eye. (Gregory, 1997, p. 5).
Noë, and others, often suggest that upside down images are pseudo-problems that representationalism invokes.  But, Gregory pretty quickly dispenses with this putative problems.  He clearly presents them as only prima facie puzzling features of vision--features that might confuse common sense or early scientific theorizing, but ones that do not give rise to deep perplexity for representationalists.


  1. If the retinal image isn't the basis for vision, for Gregory, then what is?

  2. Gregory thinks the eye is like a camera, but that what happens after a camera image is projected onto the retina, what happens beginning with the retina, is distinctly not camera-like.

    Reading through Gregory's Eye and Brain again is very nice, since he too objects to many of the same things that Gibson does. So, it dispels a lot of myths about what cognitivists are supposed to believe. For example, on page 1, Gregory rejects sense data.

  3. What about this quote from Gregory:

    "Perception is but indirectly related to objects, being inferred from fragmentary and often hardly relevant data signaled by the eyes, so requiring inferences from knowledge of the world to make sense of the sensory signals"

    Gibson would never say that the information signaled by the eyes is "hardly relevant" or "fragmented". Gibson would say that there is an redundancy of ordered information available at the eye, and the job of the brain is to discriminate from that field of information. These seem like opposite views to me. Gregory is a constructivist who thinks that the sensory stimulus is impoverished and in need of enrichment; Gibson is a direct realist who thinks that the sensory stimulus is rich in information and only needs to be differentiated. It's this difference in emphasis on either enrichment or differentiation that theoretically separates Gibson and Gregory.

  4. I know that Gibson and Gregory differ on many important scientific issues, such as direct versus indirect perception and visual images, but there are more similarities that one might expect from reading Gibson or Noe.

  5. I'm not sure those similarities can possibly be more than skin deep, given Gregory's clear stance on the indirectness of perception.

    Plus he thought Gibson was talking rubbish (according to a friend of mine who was at Bristol :)

  6. Well, Gibson does seem to have dismissed most everything that has ever been done in the study of perception, so that does seem to leave pretty little room for agreement with anyone.

    But, let's stick to cases. It's not all that interesting just to say Gibson and Gregory agreed or disagreed on a lot. Let's just look at the particulars.

  7. So take an example. How many times have you, Andrew, said that perception is not based on sensation? And, I've commented on how there is sensation and then there is sensation. Well, Gregory rejected sense datum theories of perception, just as did Gibson. So, then, we have to turn to the idea of sensations as, say, retinal stimulations. But, you have also maintained that retinal stimulation has to play some role in vision. So, this opens the door for some engagement. This Gibsonian dismissiveness is not very productive.