Newsgroups: comp.ai,comp.ai.philosophy,sci.cognitive
Path: cantaloupe.srv.cs.cmu.edu!rochester!hood.cc.rochester.edu!news.acsu.buffalo.edu!dsinc!spool.mu.edu!uwm.edu!newsfeeds.sol.net!hammer.uoregon.edu!arclight.uoregon.edu!worldnet.att.net!ix.netcom.com!jqb
From: jqb@netcom.com (Jim Balter)
Subject: Re: Emotional computers
Message-ID: <jqbE450Cs.MEI@netcom.com>
Organization: Netcom
References: <32DBBF95.64C0@esumail.emporia.edu> <5bjiq1$1b0@usenet.srv.cis.pitt.edu> <jqbE43Hv8.GvJ@netcom.com> <5blv51$mkq@lastactionhero.rs.itd.umich.edu>
Date: Fri, 17 Jan 1997 05:24:27 GMT
Lines: 39
Sender: jqb@netcom.netcom.com
Xref: glinda.oz.cs.cmu.edu comp.ai:43545 comp.ai.philosophy:51148 sci.cognitive:14491

In article <5blv51$mkq@lastactionhero.rs.itd.umich.edu>,
Greg Stevens <gregs@umich.edu> wrote:
>Then why do you think simply implementing such an interrupt-signal
>would be sufficient to implement emotion?  It may be necessary,
>but since it does not differentiate 1 - 3 above, it must not
>be sufficient.

Ah, I think I see what's going on here.  You are interpreting my line, which
you neglected to quote:

	This is exactly the point: emotion == bother.

as referring to Simon's interrupt-signal.  But that leap, while sensible after
a fashion, was unwarranted, since I didn't mean that and didn't even notice
the association, a bit of a pun.  Rather, I was referring specifically to
Anders' question: "Why would Mr. Spock bother?" and suggesting that "emotion",
*whatever* it is, is precisely the answer to that question, it is the
mechanism that evolution gave us for "bothering", for formulating goals at
all, and thus is the answer to the question of why the goal reordering could
not be "completely dispassionate (free of emotion)".  No "Kantian rational
will" can serve to organize goals, because goal ordering is fundamentally
non-rational.  There is no such thing as a "dispassionate" goal reordering
because emotion *is* the basis for ordering goals.  Without emotion, we
wouldn't bother.  I was not suggesting an *explanation* or *model* of emotion,
certainly not one so simple as an "interrupt".  But to make a stab at it, my
concept is of a rather complex association between emotive states and
cognitive states, interrupts, other stimuli.  Our emotional responses become
objects of our cognition, and we act upon them.  So our emotions are somewhat
opaque primitives or tokens to our cognitive processes, but they are not in
fact primitive but are generated by various states of the organism, including
"interrupts".  So emotions are our cognitive representations of these organism
states (but not only; they are also accompanied by relevant physiological
changes, such as increased heart rate and such).  Once "I'm mad" is an object
of cognition, the cognitive faculties can execute the relatively "rational"
processes of selecting actions that are in accord with being mad.  Anyway,
that's how I speculatively see it.
-- 
<J Q B>

