RE: [huml] Re: a random citation

["Bullard, Claude L (Len)" <clbullar@ingr.com>]

I agree that primary and secondary emotions don't 
seem to apply to the notion of a timeline, but on 
the other hand, I doubt I've seen the full set of 
properties the users of those terms are expressing.

HumanML itself should be agnostic to the theorectical 
basis of the secondary vocabulary.  I'm not convinced 
the model the author expresses is correct, but it should 
be expressible.  I'd have to study it more to understand 
it and work out a way to do that.   The application of 
HumanML I envision based on the abstract types is that 
any number of different theories could be modeled and 
tested against observable qualities or applied to a 
visualization through an implementing engine such as 
one for an avatar.

len


From: Rex Brooks [mailto:rexb@starbourne.com]

I think it was more in the way of the odd occurence that she turned 
up the reference, but I may ask her later.

What is nagging at me right now is that I think the model these folks 
are building would be better served if they used the terms: 
InitialEmotion, and SubsequentEmotion in their format of course, 
rather than as the compound terms I have just coined. Primary and 
Secondary don't build a picture of the progression they are 
describing, especially not in the sense of a crescendo/dimuendo, 
which our intensity-range values can quantify, as implied by their 
description. Their usage also not how I envisioned/envision building 
a secondary HumanML vocabulary from the base complexType Emotion. I 
was thinking of primary emotions being the least complex, as opposed 
to the more complex as their usage describes.

Ciao,
Rex

At 3:19 PM -0500 9/4/03, Bullard, Claude L (Len) wrote:
>I pulled Constance from the cc list just in case she
>doesn't want to be in this loop.  If she does, add her
>to your reply.
>
>One point is that given a theory like this, a simple
>set of polar opposites for emotional state ranking is
>insufficient.  Either that, or we have to model the
>secondary emotion as the extreme polarization of the
>primary.  However, we use intensity as an offset
>value from neutral and I don't think that is adequate
>for modeling the emergence of the secondary.  We'd have
>to give this some thought as to the right way to represent
>this and what values affect it.  The author of that page
>suggests that one value is time (fear can become anger).
>So the event timeline is important here.
>
>A second point is that emotional management via using
>sign sets to evoke other emotions is probably a valid
>approach.  It fits into the theory that emotions have
>only other emotions as opposing forces and that manipulation
>of the sign set is one means for the intellect as the
>selector to manage emotions.  It is a trick question I
>ask people:  what is the opposite of emotions?  Adults
>usually reply, intellect.  Children usually reply,
>other emotions.  Children have a more correct model
>and that is also an intrigueing bit of information
>given one asks why do adults believe that they can
>oppose their emotions with their intellects, and is
>that a symptom of a dysfunctional personality.
>
>len


-- 
Rex Brooks
GeoAddress: 1361-A Addison, Berkeley, CA, 94702 USA, Earth
W3Address: http://www.starbourne.com
Email: rexb@starbourne.com
Tel: 510-849-2309
Fax: By Request