[This letter was sent by me to every member the University of Illinois Department of Psychology along with a preprint of a manuscript that was later to be published as: Brewer, W. F. (1974). There is no convincing evidence for operant or classical conditioning in adult humans. In W. B. Weimer & D. S. Palermo (Eds.), Cognition and the symbolic processes (pp. 1-42). Hillsdale, NJ: Erlbaum.]
[In retrospect this letter is an extraordinarily idealistic and extraordinarily naive document for a young assistant professor to hand out to his senior colleagues. Clearly, I believed that we were doing science and that in the competition of ideas may the best ideas win. I obviously did not fully appreciate the pragmatic fact that scientists are human too. This letter made my tenure decision a very close call, however, reading it over 25 years later I would not change a word.]
I am distributing copies of the attached paper partly because I think it is an interesting paper, but primarily because of its policy implications. The paper is entitled "There is no convincing evidence for conditioning in adult human beings," and if you read it I think you will be moved a good way toward accepting this unconventional thesis.
In essence, the paper shows that there is no evidence for unconscious, automatic conditioning and that the results of the traditional conditioning literature can only be accounted for in terms of a cognitive theory using concepts such as awareness, expectancy, etc. If conditioning and learning, the very foundation of S-R theory require a cognitive explanation, then the implications for S-R theory in general are obvious.
The policy implications of the paper derive from one's general views on the relationship between conditioning and learning and more complex behavior such as the higher mental processes.
(1) If one takes the position that S-R laws of conditioning and simple learning can be extended upward to handle the higher mental processes and other forms of complex behavior then:
(a) Conditioning and learning ought to be the primary focus of graduate and undergraduate course work in psychology. All students, even those in the applied areas, should master this fundamental material.
(b) A major portion of departmental resources ought to go toward ensuring that there is a strong group of faculty members in this area.
(c) A major portion of departmental research ought to go toward establishing the fundamental S-R laws of conditioning and learning and their extension to problems in cognitive psychology, child, social, clinical, etc.
(2) If one takes the position that S-R laws of conditioning and learning are appropriate for simple behavior, but that other more powerful theories are required to deal with the higher mental processes and other more complex areas then:
(a) There should be separate sets of courses, and graduate and undergraduate students should take the courses relevant to their interests.
(b) The department can choose to emphasize certain areas or try to cover all areas.
(3) If one takes the position presented in the paper that even the simplest conditioning and learning must be dealt with in cognitive terms, the implications might be taken to be like those in (1) above with the higher mental processes replacing conditioning and learning. However, I think most cognitive types are less imperialistic than were the S-R types and realize that cognitive psychology has little to offer, now or in the future, to certain areas such as motivation, community psychology, etc. (There is one implication of cognitive psychology for these areas--applying S-R theory will not be productive--these areas will just have to develop theories specific to their own problems.) Given these restrictions, evidence such as that presented in this paper would seem to suggest the higher mental processes as a central focus of a department looking to the future.
It is hard to judge the position of a department as big and diverse as ours, but I think we fall somewhere between positions (1) and (2) outlined above and it is the purpose of this letter and the attached paper to help move us somewhere between positions (2) and (3).
I would enjoy hearing your comments on these issues--printable or not.
A final point--the paper is an attempt to cover the entire relevant conditioning and learning literature. It is obviously impossible for one person to do this and so before this thing is put in final published form I would appreciate any references (pro or con) and comments on the logic and arguments it contains.
P.S. When this paper becomes public I am afraid that I will be excommunicated from the ranks of Iowa Ph.D.'s I hope no one minds having a defrocked colleague.
Back to Bill's Personal Page
Last updated May 7, 2000 by EFB