[Aleader-dev] Re: direction

[William L. Jarrold]

On Tue, 5 Aug 2003, Joshua N Pritikin wrote:

> [I am aware of your limited time.  I don't expect you to acknowledge
> everything I wrote in this email.  I am keeping track of any issues
> which I feel are unresolved so I can feed them back to you later.
>
> For example, over the last few days I have made an effort to better
> separate the science & philosophy portions of Aleader.  Eventually
> (not now!) I want to revisit whether the new organization hides
> the philosophy sufficiently well.]

Yes, I meant to say that your .sig looks much better.  Note that I have
not yet looked at the (updated?) website to which it points.

>
> On Mon, Aug 04, 2003 at 08:49:53PM -0500, William L. Jarrold wrote:
> > I am finally attaching my comments on the tutorial.  They  way
> > they are written is sort of halphharzed.  I first only
> > put new stuff by commenting it out with % and putting at
> > WLJ at the start of it.  Then I did a big comment blurb with
> > WLJ.  Might be best to respond to that via email discussion
> > rather than attachment.  Then I went back and changed a few
> > lines of the original text without adding any WLJ indicators figuring
> > you can use a diff like tool to find 'em easy.
>
> The best way is just to change stuff without putting in any %WLJ
> marker.  Diff (or emacs's emerge) works great.
>
> I see a lot of uncontroversial edits, such as:

...

>
> It is best if you simply commit these changes to CVS, but it's no big
> deal.  I can also do the commit.

Yes.  Sorry.  I will eventually bite the CVS bullet.  Maybe during the
next go around.  Might make most sense to work over some of the
fundamental philosophical, methological issues first.

>
> > My main belief is that
> > there may be some robustness to your categories but not much.  Perhaps
> > the following will make the point..."Arto Chirps a Reply" is slight
> > different.  Different than what?  I forget exactly, my written notes
> > (made a few days ago are unclear).  But it is probably different from
> > all or most of the other "Celebrate Presence" clips.
>
> Frankly I'm surprised to read this because I clearly stated in
> the tutorial:
>
>   The second row which says ``Artoo chirps a reply'' is dim gray
>   (this indicates that the example may be weak or incorrectly classified).

Okay.  Could be a brain fart on my part.  My main point -- that
there is still some heterogeneity within each category -- remains.

>
> In other words, you should have ignored the examples which are
> shown in dim gray.  (Oops, now that's a bug.  Why do I show
> these situations at all?  I'll fix that.  Sorry for the confusion.)
> OK, let's move on ...
>
> > Both "Master Yupas" are distinct from one another...One involves
> > a set of collegial regard and the other is more like, yay, papa is back.
> > The former emotions depicted are more adult in tone, the latter more
> > childlike in tone.
>
> Agreed.
>
> In any classification scheme, some variation is unavoidable.
> The challenge is to see how well we can minimize the variation.

Yes, you want to minimize variation within a category.  However the
more categories you have, the less manageable and or harder to verify your
theory becomes.  Yes, in the limit as time approaches infinity you
will want to maximize the number of your categories.  But in the short
term, I believe that the most fruitful theoretical work will be applied
to emotions that 3 y.o. children can distinguish between.

>
> In this particular case, I don't have any idea how to do better.
> Subjectively, I find both "Master Yupa"s similar enough.  I do not
> feel urgent motivation to further distinguish them.

You don't, but it is quite likely that other humans will see things
differently.

>
> > o I did not really get passed a thorough checkout of "celebrate
> > presence", i.e. Item #31.  However, I did a random leap and ended up
> > looking at "You think I'm afraid of you big fuck?" from Goodwill
> > Hunting.  This appeared to be classified as admiration (there was a box
> > checked next to "[+] admires [0].").
>
> Perhaps it seems strange to classify this situation as "admires"?
> "Admires" is actually a _general_ category.
>
> This situation also classifies to the _specific_ category
> "haughty / arrogant" (which is a sub-type of admires).

I don't see haughty as being a sub case of admires, unless admires
includes self-admiration?

>
> One more comment: I have not put much effort into translating
> Aleader's affective assessment in idiomatic, common-sense
> English.  This is one of the many reasons why I request people
> to stop reading my explanation and start watching film clips.

Sure.  Film clips are a nice way to illustrate an emotion concept by
way of several examples....Btw, it would help if your writings made the
following clear:

o Any given character in a given film clip may be depicting
several different emotions at once.

o A set of characters that are simultaneously present in a film clip
need not be depicting identical emotions.

o The overall emotional tone of depicted by a film clip is something that
is distinct from what each of the individual characters in a film is
depicting.

....Then again, perhaps you disagree with the above points.  If so, lets
talk.

>
> > okay, well, basically I think I get it.   ...
> > ... there is some similarity.  Sure, there is a happy feeling
> > felt by certain characters in all of those scenes.
> > ...
> > So far, the chief value of your system seems to me to be able to
> > provide a vivid and reliable depictor of emotions.
>
> Great!  Now I won't have to run around in circles trying to verbally
> explain what you have now seen & quickly absorbed first-hand.
>
> If we publish an article, I guess we should "strongly recommend"
> that readers try out the CD?

I guess.  But I feel like I knew this before I saw the film clips.

>
> For those readers who don't try the CD, how much of an attempt
> should we make to verbally explain what is on the CD?  Perhaps
> such a description should go in an appendix or something?

I'm not sure...Ask again later...However, an important first step to
answering this question of yours is getting the following out into the
open: My sense is that this "seeing is believing" deal is more important
if one is attempting to promulgate the use of the CD for personal growth
purposes...The research community wants cold hard data...Of course, the
research community subconsciously likes watching cool videos, seeing neat
demos etc but you have to slip that in under their noses -- e.g. during a
talk at a conference, or with such compeling data or with such glowing
avidafivits from people at Harvard that they have to check out the demo
CD.

My intuition tells me that you are into both the researchy and the
personal growth aspects of this CD.  However, my intuition also tells
me that you have not yet developed separate strategies for these two
aspects.

>
> > Maybe, if you replace your categories with OCC categories and/or Roseman
> > categories you will get more attention from the academic community.
>
> I certainly want to compare/constrast with OCC & Roseman.  In fact, I
> got a copy of Roseman96 today.  Sorry it took so long.  I am looking
> forward to reading it.  However, I am not willing to dump the Aleader
> model.  Perhaps I am stubborn or irrational about this, but I continue
> to believe that Aleader offers a more precise affective model than any
> other existing model.  Obviously, this belief does not rest on being
> well-read.  It rests on a long inner struggle and deep introspection.
> I will offer some comments on Roseman soon though, in any case.

Great.  I look forward to your comments.  Maybe you can show me a little
bit of your theory in these emails.

>
> In your first email, you suggested that I narrow the scope of my
> research.  I agree that I am trying to do something "too big."
> However, I don't see a problem with that.  I don't have a deadline.
> We can proceed in small, manageable steps.

Definitely.  Step one might be publish a paper about a teeny tiny problem.
Step one might be deciding to focus on just two emotions that three year
olds can reason about.  I am not against theories that cover a lot of
ground -- even if they are more complicated and ungainly than a Windows
program!  I suppose that what I am against is trying to build the entire
program all at once.  To follow the analogy, I'd rather see us write a
simple word processor.  Then maybe a spreadsheet program.  Then an
architecture to run both the word processor and the spreadsheet at the
same time.  Then, throw it all over and start over, this time with a
slightly more ambitious goal...Sadly, no one but perhaps Clark Elliot or
Eric Mueller or maybe barely just barely me, has written and validated
"word processor" yet.

> Be involved as much as
> you want.  I appreciate your feedback very much.

Good.

>
> > One important
> > question is, is there more heterogenetiy between categories than within
> > categories?  One way you could test this is ask people to view pairs of
> > scenes and rate their affective similarity on a scale of 1-5.
>
> That's a great idea!  Why didn't I think of that?  Here is an instance
> where your broad awareness of the field of cognitive psychology really
> helps me out.

Great.  I feel like there are other ideas like this if I just ponder for
a while.

>
> I have already done most of the work to automate this type of test.

Good.

> I will get busy and make a few more preparatory changes to the software.
> Perhaps within a week, we'll be ready (software-wise) to get started
> testing human subjects.

One thing that would make it *much* easier to test gobs of subjects would
be to have them simply point there browser at a website.  The CD thing can
be a pain....If you can do the website thing then you can do something
like Open Mind.  How hard would it be to do your thing via a website
rather than a linux isa cd?

>
> > o It would be interesting to apply different types of text understanding
> > and/or statistical NLP to the scripts you have provided.  If it's
> > inferences could be sync'd up to different film clips and its inferences
> > compared to human's inferences watching (reading?) the same film
> > (script), now we are getting something quite interesting.
> > Specifically, by comparing affective judgements in each of these three
> > conditions.
>
> That does sound very exciting.  It also sounds like something that
> requires big databases and more computing power than I have at
> present.

Hmm, I don't think so.  There is some simple statistical nlp involving
a program called liwc at...
http://homepage.psy.utexas.edu/homepage/faculty/pennebaker/reprints
...I doubt that "real" NLP takes up much more memory than Good Will
Hunting.  How much memory does the average feature length film
require?  Any idea off hand how much Cyc and or KM require?  (Yes,
I could check this and really should memorize it).

>
> On the other hand, I do want to collect these more ambitious research
> ideas.  Who knows, semi-automated emotional analysis of complete films
> may be of practical value to Hollywood studios.

Yes, interesting idea.  But I think the video game industry and the
clinical psychology market will be nearer term wins.

>
> Personally I am happy to begin with an investigation of the question
> you raise above, "is there more heterogeneity between categories than
> within categories?".  The test will require little effort to
> administer.  Statistical analysis is straightforward.  There is lots
> of precedent in the published literature (can you suggest a
> particularly good article which I can use as a model?).  We could even
> limit the scope of the test to the 10 easy categories in the "getting
> started" guide (maybe, this choice has pros & cons).  It should be
> relatively easy.  What do you think?

Basically yes.  But after doing my dissertation my belief is that an
idea that seems trivially simple at the start will turn into something
monumental.  I had no idea what a repeated measures design was
and that it would take me months of consultations with PhD experts
in the Statistical Services office at UT...

Nonetheless, this is
probably our best -- i.e. most well defined and tractable -- research
problem yet.  I'd like to talk to my dissertation advisor about --
she always has great advice.  Sorry, no articles I can think of
off of the bat.  It would probably be most interesting to compare
three different categorization schemes.  Maybe yours, plus Ortony's
plus some scheme that should show now sig difference.  THe number
of subjects we can get will really drive the kinds of questions
we can ask.  If we can do it via the www we'll be able to get many more
subjects IF the academic community is okay with surveys from
individuals randomly obtained from people over the web.  Else,
we *might* be able to get some subjects through my connection with
UT but, as I've said earlier, I am bottom priority given that I
am not faculty and not doing this for my dissertation.  We'll get
a feel for # of subjects available to me in the UT Subject Pool probably
by mid October.

>
> A simple article like this may be a necessary pre-cursor to a big
> NLP & inferencing project anyway.  Isn't it?

Definitely.

It's good to have a mix of starry eyed visionary long term projects
in the background and some "low hanging fruit" in the foreground.

>
> > I am still quite unfamiliar with your theory.
>
> Even so, I want to decide a course of action.  Unless you come up with
> something better, I am going to plan my time according to the aim of
> publishing an article addressing the question: "is there more
> heterogeneity between categories than within categories?"

That's a decent operating assumption.

We could also consider a conference presentation.  One can often get
both a conference and a paper out of the same piece of research -- there
are different aspects, different angles on a given project....There is
also the cynical notion of "the publicon".  The smallest publishable
unit.  People with physics backgrounds like to mention such things.

>
> I guess I need to write a research proposal now?  Perhaps a one page
> outline giving an overview of what we have, what we want to do with
> it, and what we resources we need?

Yeah, sure.  I'm a little tentative bc I feel like brainstorming more,
talking to my advisor, exploring other ideas, but my intuition tells
me that there isa at least a 55 % chance that we will not come
up with a better idea working with the same level of intensity over the
next 2 weeks.

>
> I hope you can guide me through the process and perhaps help with
> corralling human subjects.  I _may_ be able to find English speaking
> human subjects here in Nashik too.  Perhaps it would add something to
> use human subjects from two different countries?

Yes.  The cross cultural thing is quite the fad in psych nowadays.

Getting human subjects is a bitch.  You have to go through all these
committees proving that the subjects won't be harmed.  I don't know
how this works in places like India.  If one wants to publish in Cognition
and Emotion what kinds of human subjects review must one complete?
Well, I have access to Diane (my dissertation advisor) who is well
ensconsced in the academic cliques that do all this stuff.

>
> I hope you find time to explore Aleader in a bit more depth.  Here are
> my predictions about what you will find:
>
> + The categories are more robust than you thought initially.
>
> + You will appreciate the elegance and computability of the model.

Cool.  How do I learn the model.  Sorry, I'm afraid I'm asking an
extremely obvious question.  My life as an psychology intern -- just
starting this week -- can be very stressful, so I might forget stuff.

>
> + You will see how the model can be extended or scaled up.
>
> + Simultaneous with your growing appreciation of the model, you will
> begin to feel that most people will need some training to go beyond
> the ten easy categories.  I think this is a rather sticky problem.
> I look forward to hearing your opinion about it.

Yes.  Hey, I gave one the very non-techie people supervising me a 1 minute
description of your work.  She said it sounds like, get this, "Movie
Therrapy".  The idea is you turn the volume down and watch a movie
with a kid who has social skills problems.  You ask the kid what emotions
are the characters feeling.  This is especially good for Asperger's or
High Functioning Autistic kids.  Or anyone who has trouble reading
social cues.  Many of the kids with whom I will be working have just
these deficits.  You might also be interested in the CASP and the DANVA.
Both of these are instruments that neuropsychologist type people
who I work with at UT use.  And these instruments are used to
detect impairments with social cue perception.

Inferring emotion from prosody is somethign much different than
the logical cognitive models that my dissertation was about.  But
your system is capable of assessing someone's prosodic perception.
But your formal system is miles away from modeling prosodic perception.
The kind of formal system for that is more like a markov process,
a neural network...something that handles non-digital, non-symbolic
representations.

Gotta run.

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

THe above is a "good sound bite" to have near the top some of
web page.   One that focuses on the clinical dimensions of your
work.

Bill
>