Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!isi.edu!vaxa.isi.edu!smoliar
From: smoliar@vaxa.isi.edu (Stephen Smoliar)
Newsgroups: comp.ai.philosophy
Subject: Re: "Emotion" vs. "Understanding" (was: Re: emergent properties)
Summary: bringing time into the picture
Message-ID: <15276@venera.isi.edu>
Date: 12 Oct 90 14:02:23 GMT
References: <3549@media-lab.MEDIA.MIT.EDU> <45348@apple.Apple.COM> <3560@media-lab.MEDIA.MIT.EDU> <3129@idunno.Princeton.EDU> <15268@venera.isi.edu> <3679@media-lab.MEDIA.MIT.EDU>
Sender: news@isi.edu
Reply-To: smoliar@vaxa.isi.edu (Stephen Smoliar)
Organization: USC-Information Sciences Institute
Lines: 44

In article <3679@media-lab.MEDIA.MIT.EDU> minsky@media-lab.media.mit.edu
(Marvin Minsky) writes:
>  Who did that classic research
>on speech + nonspeech perception, in which random clicks were
>perceived as closer to phrase-boundaries than they really were?

The classic click experiments I know about were performed by Jerry Fodor and
Tom Bever, who wrote them up in the paper, "The psychological reality of
linguistic segments," published in the JOURNAL OF VERBAL LEARNING AND VERBAL
BEHAVIOR, Volume 4 (1965), pages 414-421.  I do not remember this work for its
results on distinguishing speech from non-speech.  Rather, I remember that they
used the phenomenon of click displacement for evidence of segmentation.  From
this, they attempted to produce real-time models of parsing which takes place
as the words are heard.  These models were subsequently implemented as
augmented transition networks by Ron Kaplan ("Augmented Transition Networks
as Psychological Models of Sentence Comprehension," ARTIFICIAL INTELLIGENCE,
Volume 3 (1972), pages 77-100).

>  The
>point is that each sub-agency of the mind can - and must - construct
>its own model/theory of what happened recently.  
>
This actually seems to be one of the more salient points which Husserl was
trying to make in his work on time-consciousness.  However, Husserl was
particularly taken with the "processing" of time in BOTH directions.  In
other words the mind not only constructs a model of the recent past but
also projects that model into the near future.  This is in the same vein
as Schank's argument in DYNAMIC MEMORY that understanding involves our ability
to form expectations and recognize which of them are satisfied and which are
thwarted.  Leonard Meyer has tried to deal with music the same way, although
he spends so much time looking at scores that he tends to lose track of the
issues of real-time processing.  Why hasn't anyone been able to define an
analogous set of click experiments for listening to music?

=========================================================================

USPS:	Stephen Smoliar
	USC Information Sciences Institute
	4676 Admiralty Way  Suite 1001
	Marina del Rey, California  90292-6695

Internet:  smoliar@vaxa.isi.edu

"It's only words . . . unless they're true."--David Mamet