Newsgroups: comp.ai.neural-nets
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!howland.reston.ans.net!vixen.cso.uiuc.edu!uchinews!cicero!fugue
From: fugue@cicero.spc.uchicago.edu (The Right Reverend <name withheld by request>)
Subject: Re: Properties of spreading activation nets?
X-Nntp-Posting-Host: cicero.spc.uchicago.edu
Message-ID: <D8n7Go.IE5@midway.uchicago.edu>
Originator: fugue@cicero.spc.uchicago.edu
Sender: news@midway.uchicago.edu (News Administrator)
Organization: /usr/local/lib/organization
References: <D8MI6G.DJH@midway.uchicago.edu> <3p83b2$elk@cantaloupe.srv.cs.cmu.edu>
Date: Mon, 15 May 1995 23:19:35 GMT
Lines: 47

In article <3p83b2$elk@cantaloupe.srv.cs.cmu.edu>,
Scott Fahlman <sef@CS.CMU.EDU> wrote:
>
>In article <D8MI6G.DJH@midway.uchicago.edu> fugue@cicero.spc.uchicago.edu (The Right Reverend <name withheld by request>) writes:
>
>     Could someone point me towards references that would give a detailed
>   analysis of the various properties of spreading activation networks?  I've
>   looked and looked, and haven't been too successful.  Ideally, I'd like
>   some explanation and/or proof of various behaviours of these nets under
>   various configurations/conditions.
>     Or, if someone has knowledge of these nets, a followup post or email
>   would be nice, too.
>
>The usual readers of this newsgroup might not be able to answer this
>one, so I will bend my usual policy of not answering people who hide
>behind "handles" or otherwise fail to identify themselves...

  Sorry.  I was not hiding.  "fugue" is, and has been for a long time,
my username.  "The R R etc..." is simply there as an ongoing annoyance
to the purveyors of a certain Draconian policy about aliases here at
our site.  It's not meant to hide my ID...I guess I'll have to start
using an....*shudder* .sig file.  For those who it irritated, I'm
Mark C. Langston
Dept. of Psychology
University of Chicago

>>distance.  The degree of formal analysis or proof available for such
>models depends on the specific model you have in mind.  A lot of these
>models come out of cognitive psychology, and some of the ones on the
>AI side pre-date the mania for formalism.
>

  Sorry.  Although I've worked with a few specific CogSci models
(e.g., Kintsch's CI model, van den Broek's work, etc.), I've had
a bit of trouble sussing out any formal explanation for some of their
behaviour.  I'd rather focus on my problem - text comprehension - as
my research concentration, than write a thesis on spreading activation
formalisms.

[...various references deleted...]
Thanks.  I'll look into these.

-Mark Langston




