Semantics and Syntactic Representations

Jacob Galley gal2 at kimbark.uchicago.edu
Sat Dec 4 15:21:31 EST 1993


In article <2dgd63$lv9 at agate.berkeley.edu> lmk2 at garnet.berkeley.edu (Leslie Kay) writes:
>In article <2dftqe$d33 at truffula.fp.trw.com>,
>Harry Erwin <erwin at trwacs.fp.trw.com> wrote:
>.....
>>in the ionic currents associated with K0 systems in the brain. At the same
>>time, Walter J. Freeman has recently presented evidence that his KIII
>>model of the olfactory system generalizes to all the sensory modalities.
>
>Although we hope this is true, really he has shown that primary sensory
>cortices: visual, auditory, and somatosensory, show post stimulus regions
>where spatial patterns of activity measured at the surface of the cortex
>can be discriminated as to which stimulus they are produced in response to.
>In other words, the spatial patterns exhibited in the olfactory bulb also
>occur in the other primary cortices and in the prepyriform cortex (primary
>olfactory cortex).  They occur in response to a meaningful stimulus of
>the same modality as the cortex.

Two easy questions:

1) Can you give a reference for this work?  Has it been published yet?

2) I have been reading some of the more recent and less technical
papers coming out of Freeman's lab, but I have not yet found any
explanation for the Kn notation that is used to refer to various
models.  Can one of you refer me to a explanation of this notation or
explain it briefly here?

A harder question:

Leslie, I have been curious for a while about your and/or Walter
Freeman's attitude toward the "enactive" paradigm proposed by
Francisco Varela, Evan Thompson and Elanor Rosch in _The Embodied
Mind_ (1991) and elsewhere.  Your research is quite compatible with
their reasoning, yet I have only seen their work alluded to once in
sources comimg out of Freeman's lab.  That one reference was in a
short paper by Christine Skarda (1992) in a collection co-edited by
Varela himself, _Understanding Origins_, which was in fact a direct
reply to his recent book:

#Varela's critique of problem solving and representation, and his
#emphasis on what he terms "enaction", I take to be another way of
#getting at a view of perception that Freeman and I have developed on
#the basis of olfactory processing.

And what you wrote at the end of your previous post:

>actually, this is still hypothetical and my work in progress, but I would
>like to stress that this is probably not a "translation" and pattern
>matching process.  These sensory processing areas can be seen as a
>dynamical system where each part contributes to the whole system 
>entering a state which is the perception, thus dispensing with the need
>for pattern matching.

. . . is also compatible with the enactive paradigm.  (I may be
projecting into this paragraph quite a bit, but it seems that you
would probably find enactivism agreeable.)  But is there some
criticism of this paradigm that I am missing?  Or perhaps such grand
conclusions are simply not appropriate in a strictly neuro-
physiological context?  (What about "Representations: Who needs them?"
(1990), then, which could easily have touched on Varela's work?)

Here is a nutshell summary of enactivism, taken from "Ways of
coloring: Compartative color vision as a case study for cognitive
science" by Thompson, Adrian Palacios and Varela, 1992, BBS 15(1).

#The enactive view of perceptual content is . . . different from both
#the "externalist" view that perceptual content is provided by distal
#physical properties and the "internalist" view that perceptual content
#is provided by subjective qualities (qualia).  According to the
#enactive view, the contents of perceptual states are to be type-
#identified by way of the ecological properties perceived, and these
#ecological properties are to be type-identified by way of the states
#that perceive them.  One should not be put off by this circularity,
#for it is informative.  To specify perceptual content for a given
#animal we must investigate the relevant environmental properties, and
#to determine the relevant environmental properties we must investigate
#the sensory-motor patterns of activity that constitute the animal's
#perceptual states. . . .
#
#According to enactivism, color is neither a perceiver-independent
#property, as in objectivism, nor is it merely  a projection or
#property of the brain, as in subjectivism.  Rather, it is a property
#of the enacted perceptual environments experienced by animals in their
#visually guided interactions.  Unlike computational objectivism and
#neurophysiological subjectivism, this does not lead to an
#eliminativist position regarding color:  Color is not divested of its
#phenomenal or experiential structure in favor of spectral reflectance;
#nor is it divested of its extradermal locus in favor of neural states.
#Instead, color is a property of the extradermal world understood as an
#animal's environment, a world that is enacted by animal-environment
#codetermination.  Thus we arrive at the view announced at the
#beginning of this paper, according to which color is both ecological
#and experiential.

(I hope this query does not seem rude; I am merely curious about your
and your lab's attitude toward this theory.)

Thank you,
Jake.

-- 
* What's so interdisciplinary about studying lower levels of thought process?
				  <-- Jacob Galley * gal2 at midway.uchicago.edu



More information about the Neur-sci mailing list