[Aleader-dev] Re: Roseman96

[William L. Jarrold]

Hi Joshua,

A long overdue reply...

First of all I am not that familar with Roseman.  I've read
one of his papers ways back.  He is more empirically based
than OCC.  His goal is less computationally minded than OCC
thus his model might be less rule-ifiable.

Because of my lack of familarity with the theory these comments
won't be worth much.

On Wed, 6 Aug 2003, Joshua N Pritikin wrote:

> On Tue, Aug 05, 2003 at 03:51:04PM +0530, Joshua N Pritikin wrote:
> > I will offer some comments on Roseman soon though, in any case.
>
> Another big email.  :-)
>
> Here are my reactions while reading Roseman96.  If you prefer to read
> my conclusion first then skip to the bottom.  I've tried to organize
> my thoughts into three categories:
>
> 1. How Roseman's appraisal maps to Aleader's appraisal.
> 2. What I agree with or don't understand.
> 3. How Aleader's appraisal maps to Roseman's appraisal.
>
> +++ + +++
>
> Here we go:
>
> 1. How Roseman's appraisal maps to Aleader's appraisal.
>

If you really want to do a good mapping you should say exactly
how concepts being compared are the same or are different.

> + The appraisal of probability corresponds pretty well with Aleader's
> concept of tension.
>
> + The appraisal of "an event's control or influence potential by the
> self" corresponds to Aleader's appraisal of intensity.  I am still
> refining the exact method of appraising intensity.  Perhaps intensity
> is the most subtle component among Aleader's menu of appraisals.
> Maybe KR is needed to help me get rigorous.
>
> + I am confused by the terms "positive emotion" and "negative
> emotion".  It seems tautological that "improving things" is positive
> and "made worse" is negative.  On the other hand, I acknowledge that
> some way is needed to differentiate positive & negative emotions.  I
> guess this roughly corresponds to Aleader's appraisal of individual
> intention.
>
> + The appraisals of causation by self and causation by other are
> represented implicitly in Aleader when the intention of two
> individuals are combined to form the situational intention.
>
> 2. What I agree with or don't understand.
>
> Agree with:
>
> + I agree that the appraisal of "whether one can cope with an event"
> is not relevant for differentiating emotion.
>
> + The appraisal of legitimacy and problem source are not represented
> in Aleader because the emotions which they differentiate are
> considered composite / sequential emotion patterns.  Superficially,
> I do not see a problem with modelling these emotion patterns,
> but I have not attempted it.

Your use of the term "emotion patterns" confuses me.  Is an "emotion
pattern" just a fancy name for "emotion"?

>
> + I think that surprise could be modelled in an Aleader as a sequence:
> tension != relaxed then phase = after. (Don't worry about it if you
> don't understand my notation.  Surprise is just one emotion.  We can
> come back to it later.)
>
> Don't understand:
>
> + The idea of "motivational state" just seems confusing.  Maybe I
> don't understand what it is suppose to mean.  I guess I agree that
> motivational state needs some revision, as noted on page 261.

Hm\m.   I wonder what I can get this article.  Do you have a URL handy or
a copy of the article you can email?

>
> + I'm not sure whether I understand the appraisal "causation by
> circumstances". Somehow it seems related to probability, but maybe
> not.
>
> 3. How Aleader's appraisal maps to Roseman's appraisal.
>
> Aleader's appraisal has five main components: initiator, intention,
> phase, tension, and intensity.
>
> + Both tension and intensity are well represented in Roseman's model,
> as noted above.
>
> + Intention is somewhat represented, but Aleader's method of appraising
> intention is more complex than Roseman's.
>

Occam's razor favors simplicity.  Maybe you meant to say that Aleader
allows for more richness or organicity or something like that.  Whatever.
Not a big deal.

> + Aleader appraises phase. I didn't find phase in Roseman's model.  If
> you recall my eariler email, a precise explanation of "phase" can be
> found at the end of KM's situation manual.

Phase refers to whether the emotion refers to a past, present or future
circumstance, right?

>
> + I did not find anything about initiator in Roseman's model.  Perhaps
> this is due to his experimental methodology.  The subjects are asked
> write about an event in which they were a participant.  Therefore, the
> point of view will usually (always?) be the subject's point of view.
> To contrast, in Aleader there is no preset preference among the two
> participants' point of view.

I'm not getting you.  Probably bc I am not familiar with Aleader theory.
What are participants?  What is initiator?  I seem to remember that there
is no paper that describes your theory.

>
> W Jarrold wrote:
> > i believe there is a hole in the literature.  there is a more general
> > class of concepts that should be called "affective states".  emotions
> > are a focused subclass of these.  ambandoned is a classic affective
> > state which is not an emotion.
>
> Now I understand what you are talking about.  Yes, I agree.  Many of
> Aleader's "emotions" are not what people typically expect as an
> emotion.  While Roseman seems to stick with a more traditional
> definition of emotion, Aleader follows the affective state idea.
> However, instead of inventing a term "affective state", I re-defined
> emotion to mean what you call "affective state".  I still remain
> undecided whether it is better to introduce a new term or to re-define
> "emotion".  I don't think emotion is very well defined (in general)
> and "affective state" is a mouth-full.  What do you think?

It depends on how the field has defined emotion.  Do what the field
does before arbitrarily redefining emotions.  OCC defines emotion
one way.  How do others define it?  Reading happens fast, so long
term names are okay.

>
> OK, I suppose I should write some sort of conclusion now.
>
> Despite confidence in my own introspection & creativity, I had quite a
> lot of anxiety approaching this type of article.  What if Aleader's
> appraisal model included a bunch of factors which turned out to have
> poor empirical performance?  That would be a problem.

Yep.  Research causes anxiety for the experimenter.  One antedote is
do design studies such that no matter what the outcome, the results
are interesting.  E.g. comparing Aleader to another model or models
is one way to guarantee interestingnes.

> Now my worry is
> mostly finished.  It is not always easy to gauge the similarity of
> appraisal questions, but I find that my intuition is mostly supported
> by the numbers.  I look forward to reviewing OCC.

Gerald Clore (the second author of OCC) once emailed me an article
bc I had trouble retrieving it from the library.  Maybe he'd do
something similar for you if you had to wait too long.

Bill

>
> --
> .. Sensual .. Perceptual .. Cognitive .. Emotional .. Oh My!
>