[From Rick Marken (951208.2130)]y
Bill Powers (951208.1730 MST) --
What we need to look at is the model that explains steady-state
behavior, not the process by which the correct behavior is stumbled
across nor the final trimming that leads to the steady state.
In other words, we need to investigate the simple claim that
"reinforcement maintains behavior."
I agree. Now try to get an EAB (or an ex-EAB) type to agree.
I don't care what the actual experiment is, but it should include a
reinforcer that is provided according to a schedule when a person
presses a key at some rate. After the final rates are observed, I
propose adding to the reinforcer a slowly time-varying amount that is
some fraction of the asymptotic rate, as a disturbance.
This was my hope.
How a reinforcement model could make predictions about this
situation I do not know.
I don't know either. I've been hoping that the ex-EABers could provide
such a model. I presented an example of a POSSIBLE PCT model (I have
already written a simpler one) to show the ex-reinforcement theorists
what I'm looking for; some computer code for a reinforcement model
that will maintain pressing behavior in an operant experiment. I was
really hoping to get the code from Bruce Abbott, who says that PCT
provides a better model of operant behavior than reinforcement
theory. I don't see how he can tell that this is true unless he is able to
compare the predictions of the reinforcement model to that of the
control model.
Alas, as you see, there has been no reinforcement model forthcoming
from ANYONE, let alone the ex (or current) reinforcement theorists.
I guess it's not that important to them; or, more likely, it's VERY
important to them;-)
Samuel Saunders (951208:21:57:10 EST) suggests playing rope-a-dope
with the reinforcement theorists rather than going for the knock out
punch. It looks to me like reinforcement theory has already lost by
TKO simply by not showing up. An experiment to test reinforcement theory
seems irrelevant to me now, since there is apparently no such thing as
reinforcement theory. I guess we have been tilting at windmills, Senor
Quixote.
Samuel Saunders (951208:21:57:10 EST) says:
I am afraid that the effect of ccomparing PCT to reinforcement models
in a restricted domain, then asserting the global superiority of PCT
will more likely have the effect of shutting the door to PCT.
When was it open?
The immediate response is likely to be something like "You have a
nice account of responding on FR schedules; how do you account for
matching (stimulus control) ( x, y or z) ?
Yes. we know. Pathetic, eh?
If PCTers then persist in asserting that the only logical choice is to
reject reinforcement and take up PCT, I suspect PCTers will get the
reputation of being flakes, and any future efforts will be tuned out.
What's our reputation now?
Imagine that PCT accounts for the detailed data in a schedule
experiment, while a reinforcement model accounts for the same data,
but not nearly as well, or in as much detail. Bill Powers and Rick
Marken will argue that it is clear evidence that PCT should be
adopted, and reinforcement approaches dropped.
If reinforcement theory could account for the same data as PCT I would
be AMAZED -- and happy, because at least I WILL HAVE SEEN A REAL WORKING
REINFORCEMENT MODEL. At the moment, the main problem is that PCT
has no model to beat -- and there are apparently a lot of people out
there who think PCT does have a model to beat -- the mythical
reinforcement model.
Most EABers are likely to conclude that PCT has something
Interesting to say about schedules, so it should be analyzed from a
reinforcement point of view to determine what it has to contribute.
You'll have to parse this one for me. Are you saying that the PCT
model should be analyzed from a reinforcement point of view? This
makes no sense to me. Or are you saying that we should continue to
analyze the data using the reinforcement model? This makes more
sense. And I'd agree. Sure, if the reinforcement model can account
for the data (even if it doesn't do it quite as well as PCT) it's
still in the running.
On the other hand, if we present a nice set of data and models
accounting for schedule effects, and just say "here is a nice model for
schedule effects" there may be some efforts to disprove the model, but
the stakes won't be as high.
We've done it; the EABers paid no attention. They certainly haven't
tried to disprove the PCT model. Why should they? They already have
the right model -- the one that can't be wrong -- the Emperor's new
model -- reinforcement theory!
I wonder if having people accept a little PCT, then a little more, until
the ideas begin to seem natural, may not work better than demanding
that where PCT is concerned, "you have to love it or leave it."
Sounds great. How do you gradually move people from the idea that, say,
food is a reinforcer to the idea that it is a controlled variable?
Best
Rick