from [ Marc Abrams (990724.1228) ]
[From Bruce Gregory (990724.1227 EDT)]
I'm not interested in what else be is controlling for _unless_ it is
necessary to explain what is happening as far as this particular
interaction
is concerned.
I believe it is.
If I drive by at 90 mph and he ignores me, then I must
consider what other perceptions he might have been controlling that
distracted him from his "duty".
How do you know he is "ignoring" you. What does that mean?
> It's not what is stored in memory ( everything _in principle_ goes into
> memory ) It is how the stored material is utilized that makes all the
> difference. References that are set from memory will have dramatically
> different effects if the imagination mode is being utilized in the
> "relationship level" versus the "event Level". or "sequence level".
All that may be perfectly true, but how does it effect my analysis? What
is
wrong with my analysis that would be rectified by considering these
issues?
Because your analysis has no justification unless you can do the test. Since
you can't. It's simply your interpretation versus someone else's. That's
science? Where is your model of this "simple" analysis.
I could indeed. One way to test this would be to drive on the road at a
time
when there was very little traffic and see how I behave.
Maybe, but you still have no idea what the trooper might be controlling for
and you would have a _very_ difficult time driving and focusing your
attention on introspection. Again, this is science?
No, I am assuming the simplest possible model that explains the data.
That's
what we usually do in science.
Sorry, I disagree. First of all you have no model. You have a verbal
description that _might_ describe a single CV at a single level. What
"data" are you explaining? What happens at one level is affected by what
happens both below it ( perceptions ) and above it ( setting the
reference ). _How_ those perceptions and reference levels come to be, at
_each_ level is fundamental and important..
Specifically, what's the point of your example? If you are trying to
describe how or why a driver or trooper might be controlling for a specific
variable ( speed, possibly ) then you don't have enough "data". Nor do you
have a model to test your assumptions with. Sorry Bruce, this isn't
science.
I disagree. Only if the simplest model _fails_ to explain the data,
Again, you have no data, you have no model, you have no explanation.
do we
need to worry about complications. There is insufficient data to either
warrant or to make possible the kind of model you are proposing to build.
Your right about the data. But I _can_ build a model and I can use the same
"data" you use to do _Your_ science. I can make certain assumptions about
the data. Except in my case I'll be able to test the validity of it with the
model. _If_ I can't build a model, then it falls into the same bucket as
your descriptions. _Pure Speculation_. Nothing wrong with that, but it is
what it is.
You are quite correct that my model ignores many things that may be
occurring, but that is its strength, not its weakness.
You don't have a model Bruce. You have a _very_ generalized description of
something that _might_ take place. It doesn't _explain_ anything. It's not
that it ignores many things that are happening. It ignores the things you
_need_ to "explain" what you want to explain.
A powerful model explains what you are seeking to understand _without_
needless
complications.
Agreed.
Think of all the things that Galileo _might_ have worried
about with regard to the two cannon balls dropped from the leaning tower.
He
ignored these possibilities and settled on the simplest possible
interpretation, even arguing that the experiment would be totally
convincing
if it could be conducted in a vacuum. It is not difficult to imagine
complications. Resist the temptation! Now go, and sin no more, my son.
I appreciate the advice Bruce. But I am a glutton for punishment 
Marc
