From newshub.ccs.yorku.ca!torn!cs.utexas.edu!convex!darwin.sura.net!wupost!micro-heart-of-gold.mit.edu!news.media.mit.edu!nlc Wed Oct 14 14:59:02 EDT 1992
Article 7248 of comp.ai.philosophy:
Newsgroups: comp.ai.philosophy
Path: newshub.ccs.yorku.ca!torn!cs.utexas.edu!convex!darwin.sura.net!wupost!micro-heart-of-gold.mit.edu!news.media.mit.edu!nlc
>From: nlc@media.mit.edu (Nick Cassimatis)
Subject: Re: Brain and Mind (was: Logic and God)
Message-ID: <1992Oct13.074437.25305@news.media.mit.edu>
Sender: news@news.media.mit.edu (USENET News System)
Organization: MIT Media Laboratory
References: <1992Oct5.174528.20148@usl.edu> <1aqirgINN5u9@smaug.West.Sun.COM> <1992Oct12.165931.9154@news.media.mit.edu>
Date: Tue, 13 Oct 1992 07:44:37 GMT
Lines: 50

In article <1992Oct12.165931.9154@news.media.mit.edu> minsky@media.mit.edu (Marvin Minsky) writes:
>
>On pp310-313 of The Society of Mind is a theory of how some infantile
>"social groundings" might be built into brain hardware. But so far as I
>can tell, no one has ever read that far into the book.  Comments would
>be welcome.

The theory (if I read it right) is that there are built in recognizers
for "time trajectories."  These are grouped into gestrure recognizing
agencies which are then connected to "proto-specialists" that are
innate mini-programs or dispositions.  This is all either genetically
coded or resulting from "predestined learning."  This theory would
explain how certain "sentic forms" are universally abstracted across
cultures and sensory modalities.

This seems very plausible for many things, but would have radical
consequences for others.  Much of emotion is caused by things such as
the achievement of a goal, the fustration of a plan, etc.  Linguists
are finding striking regularities in "argument structure" -- the
ordering of the arguments of a verb in a sentence.  For example, it
seems that in certain syntactic contexts, the agent of the verb will
always precede the patient which will always precede the goal, etc.
Similar conditions obtain in many noun phrases whose heads are
nominalized verbs.  Roger Schank, in his book titled "Dynamic Memory",
gives good reasons to believe that something like "Thematic
Organizational Packets" are involved in memory.  There are cetain
brain injuries that make it very difficult for people to order the
objects of a verb appropriately.  (Does anyone know if people with a
severed corpus collosum (or other brain damage, for that matter) can
do things like pat their heads and rub their stomaches (or play
Chopin!) more easily than normal people?)

Thus, along with sentic forms, there seem to be "planning forms", and
"thematic forms."  The planning forms seem to be always there, while
the thematic forms seem to be partially in place as soon as verbs
begin to take objects in language acquisition.  Thus, if the proposed
mechanism for "social grounding" is extended to cover these other
phenomena, innate facilities for planning and argument structure would
be posited.

My hunch is that instead of being the result of a specific wiring, the
sentic and planning forms are results of more global properties of
brain function, which are innate or predestined to be learned.  The
would have something to do with how attention is switched during
certain processes and the "stack" needed to go along with such
switching.  The properties of argument structure could be the result
of using the planning in forming utterences.

-Nick



