Bad Reviews

[From Bruce Abbott (941105.1145 EST)]

Rick Marken (941101.1410)

I tried to publish the results of my studies of the E. coli effect in
Science, as an answer to Skinner's article. The paper went through at least
two harrowing reviews and revisions but, ultimately, it left the reviewers
nonplussed. I, of course, have saved all the reviews; when PCT finally makes
it they should make very amusing reading, if they can still be read on the
decaying paper.

It is in these reviews that I learned of all the different ways that
reinforcement theory can handle the E. coli effect. The first approach was to
say that the results of a bar press were not really random. A press would
tend to occur when the dot was moving away from the target so the chances
were that it would be moving in a "better" direction after the press. So the
result of a press after a "bad" direction was likely to be a "reinforcement"
(movement in a good direction) and, in the future it would be more likely
that the subject would press when the dot was moving in a bad direction. In
fact, the results of a press are random -- that's it. And, when I implemented
this verbal explanation as a model it produced a random walk, as expected.
The reviewers were not interested.

Rick, if it's any consolation, I have similar stories to tell. A few years
ago I ran a study designed to pit two competing explanations of preference for
signaled over unsignaled shock schedules against one another. While this
study was in progress, another researcher came out with a third explanation,
which he said was firmly based on the Rescorla-Wagner model of classical
conditioning. When I submitted my results for publication, the editor stated
that the method was sound and the results interesting, but that my discussion
contained no reference to this newly proposed Rescorla-Wagner explanation and
would have to be modified to take it into account. (Did I mention that the
researcher who proposed this explanation was a former graduate student of the

The problem with the editor's suggestion what that my study had manipulated
signal length, and Rescorla-Wagner includes no parameters for signal length.
So I tried to construct different interpretations of signal length under
Rescorla-Wagner, e.g., by dividing time into intervals and calculating the
probability that the signal would be on, which is then a function of signal
length. I wrote a computer simulation to implement each of these
interpretations as they would apply within the context of my experiment.
Every one of these models predicted results exactly opposite to those I
actually obtained in my study.

I added these conclusions to my discussion section and returned the revised
manuscript to the editor. After about two months it came back rejected. It
seems that on further consideration, the editor had decided that the method
was flawed after all and that therefore the study did not merit publication.
And, of course, he completely disagreed that my simulations could have
produced the results I said they did. [The original manuscript sans Rescorla-
Wagner was subsequently published in another journal, whose editor and
reviewers apparently were not as good at spotting these serious methodological

On another occasion I received a letter from a colleague, also claiming that
Rescorla-Wagner could explain preference for signaled over unsignaled shock.
I wrote his explanation into a simulation and, lo and behold, it predicted
that subjects would be indifferent to the two schedules. He said he couldn't
see how this could be. I sent him the code. He ran it. He looked at the
code. He said that he couldn't see any flaws in the program, but that it just
couldn't be right. That was the last time I heard from him.

Years ago I worked in industry as a laboratory technician. A fellow
technician by the name of John and I sometimes got into friendly arguments,
and on those occasions when I would begin to win him over by citing empirical
evidence, he would grin and shout, "Don't confuse me with the facts! I've
made up my mind!" There are a lot of Johns out there. It has been my
experinece, however, that persistence pays. Eventually you will find people
who are willing to listen with an open mind. Some of them even write
simulations. (:->]