Shootout

[From Bruce Abbott (950609.1335 EST)]

Rick Marken (950608.2130) --

Bruce Abbott (950608.1930 EST)

I am claiming that current reinforcement [theory] provides a
consistent and generally compelling framework for understanding
behavior (which is why it has so many adherents) and is capable of
handling the sort of data Bill P. asserts it cannot (the ratio data).

Well, I'd say it's time for you to show how reinforcement theory
handles the ratio data, pardner. I'm talking about a workin' model,
friend; no curve fitten'. So it won't do ya' no good ta draw =|;-)

Black Bart downs the shot of whiskey in one gulp, calmly places the empty
shot glass on the counter, and turns in the direction of the voice. There,
just in front of the swinging doors that mark the entrance to the Longbranch
saloon, stands yet another pimply-faced kid, his right hand poised over the
cheap six-gun protruding from a shiny new brown leather holster. Bart has
seen it all before--when you're the top gunfighter every kid who wants to
make a name for himself comes lookin' for you. Bart slowly reaches down and
releases the strap on his own holster.

"Kid," he says, "do ya really want to do this? You've got your whole life
ahead of you. Why throw it all away?"

The kid flexes his fingers over the handle of his six-gun. "Yeah, I really
want to do this. I hear you're the best, and I aim to prove I'm better."

Bart shakes his head slowly back and forth. "That's what that kid in
Tombstone said. He said I wouldn't even get off a shot, but I drilled him
between the eyes quicker n' he could say 'e. coli.' " Bart pats the
well-oiled Colt Special at his side. "Me an' the "reinforcer" here never
miss. I suggest you think it over--while you still can."

Rick:

I think Bruce is claiming that PCT is better (more accurate, simpler, etc)
than reinforcement theory as a model of behavior change. In other words,
reinforcement theory (like Ptolmeic theory) is basically OK but PCT (like
Copernican theory) is much better.

You disagreed with the "OK" part. So I will change " reinforcement... is
basically OK" to "reinforcement theory explains the data collected in
operant conditioning experiments just as Ptolomy's theory explained the
data on planetary movements. But PCT explains the operant conditioning data
better just as Copernicus' theory explained the planetary data better".

Is that more like what you are controlling for?

If so, being shown that reinforcement theory doesn't explain the data
collected in operant conditioning experiments would certainly be a
disturbance, wouldn't it?

No, it wouldn't. If I thought reinforcement theory was that good I would
never have been led to look elsewhere for a better explanation. It's not
that PCT is "more accurate" or "simpler" than reinforcement theory, it's
that PCT provides a more fundamental kind of explanation. The comparison is
more like Ptolemy versus Newton than Ptolemy versus Copernicus.

My interest in the ratio data springs from a different and more limited
concern. I believe that it is possible to come up with an explanation for
the ratio data that would be consistent with reinforcement theory as it is
currently conceived.

Beliefs range from those about which you care little to those you would
strongly defend, perhaps unto death. If I were to discover that
reinforcement theory (as _I_ conceive it) could not handle the ratio data, I
think my response would be "Hmmm. Well I'll be. I wonder if anyone in EAB
realizes this."

What does seem apparent is that people in EAB who have looked at these data
don't seem to find them particularly surprising, so I'm guessing that there
is some fairly straight-forward way reinforcement theory can handle the
data. Or perhaps no one has really examined the question closely enough to
discover that it can't. Perhaps it _seems_ that it can when thought about
on a purely verbal level. What I'm concerned about is Bill P.'s suspicion
that it is understood that reinforcement theory can't handle these data but
that this fact is being hidden in order to preserve the theory. I think
this is highly unlikely. More importantly, in terms of overall strategy I
think it destroys one's credibility to assert that a competing theory can't
handle certain data, only to be shown by the "other side" that it can. This
suggests that the wise course of action is to apply reinforcement theory to
the situation and see what it can do.

So, I'll give it a shot, but don't expect quick results, I'm very busy with
other things at the moment.

    "Now." [said Deep Thought.] "Ask what else of me you will that I may
function. Speak."
    They shrugged at each other. Fook composed himself.
    "O Deep Thought computer," he said, "the task we have designed you to
perform is this. We want you to tell us . . ." he paused, "the Answer!"
    "The Answer?" said Deep Thought. "The Answer to what?"
    "Life!" urged Fook.
    "The Universe!" said Lunkwill.
    "Everything!" they said in chorus.
    Deep Thought paused for a moment's reflection.
    "Tricky," he said finally.
    "But can you do it?"
    Again, a significant pause.
    "Yes," said Deep Thought.
    . . . .
    "But the program will take me a little while to run."
    Fook glanced impatiently at his watch.
    "How long?" he said.
    "Seven and a half million years," said Deep Thought.

    [From Douglas Adams, _The hitchiker's guide to the galaxy_.]

Regards,

Bruce