From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!utgpu!csd.unb.ca!morgan.ucs.mun.ca!nstn.ns.ca!news.cs.indiana.edu!att!linac!pacific.mps.ohio-state.edu!cis.ohio-state.edu!zaphod.mps.ohio-state.edu!mips!darwin.sura.net!europa.asd. Mon May 25 14:04:43 EDT 1992
Article 5577 of comp.ai.philosophy:
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!utgpu!csd.unb.ca!morgan.ucs.mun.ca!nstn.ns.ca!news.cs.indiana.edu!att!linac!pacific.mps.ohio-state.edu!cis.ohio-state.edu!zaphod.mps.ohio-state.edu!mips!darwin.sura.net!europa.asd.
contel.com!uunet!trwacs!erwin
>From: erwin@trwacs.fp.trw.com (Harry Erwin)
Newsgroups: comp.ai.philosophy
Subject: Brain and Perception
Keywords: neurophysiology, neurons, internal representations
Message-ID: <580@trwacs.fp.trw.com>
Date: 12 May 92 12:40:20 GMT
Organization: TRW Systems Division, Fairfax VA
Lines: 51

In preparation for a workshop next week, I have been reading Karl
Pribram's Brain and Perception very carefully. This book is difficult--it
resembles Sartre's Being and Nothingness in its density, but it also has a
great deal of interesting insight. I spoke with a research psychologist on
his reading of Pribram's research, and that gentleman indicated that he
felt that Pribram has a tendency to take 60 foot steps, but also that his
lab has always had a reputation for excellent research.

A key idea in Pribram's theory of the brain is that the internal
representation of objects is in the form of a Gabor ("wavelet") transform
of sensory data. He shows how that representation can result in an
environmental model that the observer can project into and explore with
alternative sensory modalities (touch, for instance, instead of sight or
hearing). 

An issue emerges here. If the representation of an object is an
transformation into a spacial/temporal frequency domain, how do we do
object-oriented programming? I have a picture of an n-dimensional
whiteboard, with objects interacting additively (and sometimes
non-linearly) and with the program probing the whiteboard with
Gabor-transformed objects and sensory fields to understand the progress of
events.

Pribram indicates the standard connectionist model of the neuron is far
too simple. He claims that there is significant processing in the
dendritic arbor, with non-stationary dynamics. The Gabor transformation
may be localized in the combined dendritic arbors of the sensory
processing cells...

My reading continues..., but there are two ideas that these concepts have
already stimulated. First, I have speculated that the secondary models of
significant others that we maintain to support complex social interactions
are at least potentially if not actually self-aware. What this would imply
is that individuals are distributed to some degree among multiple zygotes,
although primarily localized to one. The second speculation involves the
information structures of human communities. We know that information
processing is the primary function of these communities, with economic
production being secondary (this is a recent result coming out of the
archeological community). The question emerges--to what extent does the
information processing of the human community parallel Pribram's model of
information processing in the brain? In other words, do human communities
represent significant information in a Gabor-transformed format,
distributed among the members of the community, and allowing any member to
project himself or herself to any place or time in the community? If you
have studied the functioning of urban communities, this is a provocative
idea.

Cheers,
-- 
Harry Erwin
Internet: erwin@trwacs.fp.trw.com


