[FROM: Dennis Delprato (930105)]
Subject: Behaviorism in Amer. Psychologist
From: Tom Bourbon (930101 13:58)
Dennis Delprato -- since my previous post, I have looked through
the American Psychologist issue on Skinner several more times. I
have a few questions about some things I see, and do not see.
Many of the authors, including Bryan Midgley and you, wrote about
the famous "three-term contingency" (discriminative stimulus,
response, reinforcement -- which it is said is not a S-R
formulation and does not imply cause in the traditional sense of
the word). Why did no one mention the "fourth variable" Skinner
wrote about at the start of his career -- the one he said, back
then, was necessary if the third (reinforcement) was to work? The
fourth variable of course was deprivation. Skinner dropped all
references to deprivation in later writings, apparently in an
attempt to play up the idea that food reinforcement is
intrinsically "positive.... I wonder what the two of
you think about the mysterious and long-out-of-favor fourth term.
Skinner first (1938) held that deprivation was a drive-inducing
operation as did many other neobehaviorists. In a reprinting of his
1938 Behav. of Organisms he admitted that Kantor convinced him
that drive was a dangerous construct. Subsequently, he viewed
deprivation procedures as simply controlling variables
(= variables of which behavior is a function) in his
terminology. Now drive (or motivation) had no special
place, as it did/does in most other competing frameworks.
This is why deprivation did not need mention -- from the
perspective of Skinner's psychology. For Skinner, food
deprivation increases the rate of eating, but this
manipulation is not needed. The same result could be
obtained via hypothalamic lesions, gold-thioglucose
poisoning, or selective breeding.
One "extension" of Skinner's view on deprivation that has
gained a good deal of attention is Jack Michael's proposal
of "establishing operations," which is "any change in the
environment which alters the effectiveness of some object
or event as reinforcement and simultaneously alters the
momentary frequency of the behavior that has been followed
by that reinforcement."
As far as what I think about deprivation, I think drive is
a useless construct, that general control system notions
are in the right direction, and that the specific version
associated with CSG-L has much to offer the interested
researcher as far as deprivation and other so-called
motivational phenomena. Didn't this come up a long time
ago?
I am also curious about why no one cited or discussed the Ferster
& Skinner book, Schedules of Reinforcement, in which F & S laid out
in detail the lab routines for controlling behavior -- all the way
down to deciding on the response rate you want to see from an
animal then adjusting the level of deprivation until you see that
rate. (There is one citation of Schedules of Reinforcement, in a
brief historical note at the end of the issue.) ... Is the book
out of favor, or just overlooked?
I do not follow the "behavior analysis" literature closely
enough to provide an authoritative comment on this. My
guess is that most of that ilk tend to consider S of Reinf.
as part of history, not something on which to build a
reputation as a creative researcher. Perhaps omission of
S of Reinf. is representative of movement away from Skinner's
dictates (see below) in "behavior analysis."
So this is 1993. The view out my window looks pretty much the
way it did yesterday. Wonder if anything else has changed?
I must say that skimming over the issue to which you refer led
me to be surprised by the breadth of the material. Frankly,
my experiences with the behavior analytic crowd have revealed
them to be much more resistant to broadening ideas than
revealed in the collection. They actually permitted several
references to the not-so-highly-thought-of (in my experiences),
J. R. Kantor. One paper deserving of commentary from a PCT
expert is Alessi's (pp. 1359-1370) which brings up G. Bateson
and negative feedback. Hope someone prepares a letter on this.
From my perspective, I see a good deal of change from the
psychology represented in the Skinnerian corpus. But I
do not find recognition of change in fundamental assumptions.
In my more optimistic moments, I foresee younger behavior
analysts led by data to a gradual shift in their postulates.
I find "revolutions" in science more a matter of retrospective
viewing over longer time spans than we realize than of
any sort of abrupt shift. Physics is still undergoing
change from the old physics to the new physics (as these
terms are used in a formal sense in this literature).
For some time I have detected what might be a behavior
analysis literature rather removed from Skinner's
particular views. He might have had an inkling of this,
as exhibited in certain of his later presentations.
Dennis Delprato
Dept. of Psychol.
Eastern Mich. Univ.
Ypsilanti, MI 48197