Being scientific; random behavior

[From Bill Powers (951108.middle of the night)]

NOTE: Mary is alive and well. I posted that excerpt at her _behest_, not
in the process of executing her _bequest_.

Bruce Abbott (951107.2025 EST) --

I understand the point you're trying to make, which is that even though
psychologists have been using an inadequate theory, they have conducted
themselves as scientists while doing so. I have to agree with you,
insofar as science consists only of honest observation, accurate
calculation, and consistent use of theory and methodology.

Where I depart from agreement is roughly at the same point where Skinner
and his disciples departed from their traditional colleagues, accusing
them of being unscientific and laying claim to having the only _real_
science of behavior. The basis of this claim was not that other
psychologists were dishonest or couldn't apply statistics properly or
were inconsistent in their uses of theory and methodology. It was that
other psychologists were using improper and untestable models of
behavior. If you read Skinner while eliminating all his positive
statements about good science and how behavior DOES work, what is left
is a continuing attack on the way other psychologists have tried to
explain behavior. I even have a letter from him in which he rejected an
invitation to discuss control theory, indicating that he saw nothing it
could add to our understanding of behavior. Skinner, his adherents, and
his predecessors have behaved toward other schools of psychology far
worse than any PCTer has behaved toward EABers.

If we stipulate that most scientists are honest, competent, and
consistent, what is left to discuss but the substance of the different
points of view? One can be honest, competent, and consistent and still
be on entirely the wrong track. That's the hard truth of scientific
life: one can in fact waste a career. Even worse; in the effort to avoid
reaching that shattering conclusion, one can honestly, competently, and
consistently refuse to consider any evidence that this may be true for
oneself. The result is to erect a wall to defend against outside ideas,
and this result can be seen everywhere in psychology. Not only do the
dozens of schools of thought ignore or disparage each other; they defend
themselves against each other and reject any attempts to remove the
walls. Each school considers itself the center of the intellectual
universe; all progress in psychology is progress in one's own field. I
won't say that such attitudes are unscientific, since they are obviously
demonstrated by numerous scientists, but I will say that I think this
attitude is stupid. It's the best possible way to prevent psychology
from ever becoming a unified scientific discipline.

We in PCT have always tried to be open to criticisms from any other
school of psychology. But with only a few outstanding exceptions, this
openness has gone only one way. Many people who have expressed an
interest in PCT have done so primarily to evaluate it as a threat to
positions they already hold. They are generous in offering difficult
problems for us to try to solve, but stingy when it comes to considering
problems we pose for them -- they are satisfied with their solutions,
and instead of discussing the matter they want only to show that they
have already solved the problems. To my mind, this sort of attitude
counts in evaluating whether the other person's approach is scientific.
There is more to science than honesty, competence, and consistency.


Chris Cherpas (951107.1734 PT) --

As far as I know, nobody has
ever compared cumulative records obtained when (a) an organism
is operating the bar or key, and (b) an artificial device
programmed to behave arbitrarily is operating the bar or key.

      Not true. In the lab I worked in, we routinely ran programs to
     operate the devices used in operant conditioning experiments to
     test the equipment. Your statement sounds as if someone from the
     EAB camp would really be caught with their pants down on this one
     -- that you thinks it's amazing that nobody would ever think of
     doing this. Think again.

I wasn't lying. As far as I knew, what I said was true. Now I know more,
and evidently it wasn't true. I would really like to see the results you
got; can you cite some papers? I really haven't seen anything published
where behavioral results have been compared against a baseline produced
by a random behavior generator.

     I see a lot of "nobody understands me" coming from CSG, but it's
     hard to take this kind of comment seriously. Would you expect to
     see scallops on FIs or post-reinforcement pauses on FRs with random

No. I would expect to see them in a control-system model of behavior,
and did, but not with random responding. I was talking, however, about
matching. Incidentally, any luck getting my operant behavior model

     Duh. Random responding on Concurrent VIs would not produce
     matching, my friend. It would produce the worse case of
     undermatching that anybody had ever seen.

Are you saying that it did, or that it would if anyone tried it? If it
has been tried, I'd like to see the results, and if not, shouldn't it be
done to get a baseline against which to evaluate the degree of matching
actually obtained? I can't tell whether you're speaking hypothetically
or reporting on actual results.

     In general, I'm really getting interested in PCT because it does
     offer the promise of something more. But don't BS me, OK?

Looks as though that would be pretty hard to get away with.

     If it's EAB models you want to see, a somewhat recent source is
     JEAB, Volume 57, Number 3, May, 1992: "Special Issue on Behavior
     Dynamics." I can't say I'm satisfied with what you'll find,
     otherwise I wouldn't be so interested in PCT, but at least the
     models of the organism are getting more explicit.

Love to see it. I'll start the machinery of interlibrary loan going
(i.e., Mary).

     Meanwhile, the "establishing operation" (e.g., Michael, 1982,
     JEAB), might be worth checking out, if only because it is a concept
     that is aimed toward better defining environmental events which
     dynamically set reference levels. You heard me.

How about an essay on the subject? I'd be very interested in knowing how
environmental events can set reference levels. I have some ideas, but
I'd like to hear yours first.
Best to all,

Bill P.