Animal Research in Psychology: An Example of Reinforced Behavior

Roger E. Ulrich


Millions of animals and enormous financial resources have been used to create a vast literature of behavioral psychology research on animals. For years, I contributed to this literature. From my work as an animal experimenter, I gained some important insights into human psychology. These insights, however, did not derive from the experiments themselves as much as from my own role as an animal experimenter.

Skinner has stated the following as the motivation for using animals to study behavior:

We study the behavior of [nonhuman] animals because it is simpler. Basic processes are revealed more easily and can be recorded over longer periods of time. Our observations are not complicated by the social relations between subject and experimenter. Conditions may be better controlled. We may arrange genetic histories to control certain variables and special life histories to control others -- for example, if we are interested in how an organism learns to see, we can raise an animal in darkness until the experiment is begun. We are also able to control current circumstances to an extent not easily realized in human behavior -- for example, we can vary states of deprivation over wide ranges. These are advantages which should not be dismissed on the a priori contention that human behavior is inevitably set apart as a separate field.(1)

I found, however, that animal experimentation offers only the illusion of control. In simplifying and segmenting the life of an "organism," we create artifacts. These artifacts -- in combination with species differences --invalidate efforts to generalize to humans.

Like Skinner, I experimented with animals and with my own life in an effort to seek truths that might help the world. The truths I discovered, however, were not what Skinner had in mind. Among these truths was the realization that my career in animal research demonstrated just how morally and scientifically impoverished a "respectable" researcher can be.

In 1961, I completed my doctoral dissertation entitled "Reflexive Fighting in Response to Aversive Stimulation." I had found that electrically shocking rats induced them to fight. This work formed the basis of a report, co-authored with N.H. Azrin, published in Journal of the Experirnental Analysis of Behavior.(2) While conducting various animal research projects, I was also studying mental patients. I was nearing completion of my Ph.D. in Clinical-Counseling Psychology in the clinical psychology program at Southern Illinois University. Southern Illinois was attempting to receive American Psychological Association recognition within the "clinician as a scientist" model, but there was a pervasive sense among both the clinical and laboratory graduate students that "real" scientists work in the lab and that the quantifiable, "objective" data derived from manipulation are inherently more scientific than clinical observations. Therefore, I was torn between my interest in clinical research and my colleagues' bias that animal research was scientifically superior. Although I recognized that the animal studies would not directly benefit humans, I had accepted the nearly universal faith that, somehow, my findings might indirectly help humans. Overall, I accepted the credo of those who would determine whether or not I would graduate from the program -- the belief that animal research elevated the field of psychology to the level of a science, like physics or biology, and that such a scientific method was the path to truth.

One attraction of a career performing animal research was that each research result provided the basis for further research. Each project followed the preceding one as a direct consequence. For example, in 1948, Miller trained rat subjects to fight by removing the shock each time the animals approximated the fighting position.(3) He concluded that this positioning was an escape reaction reinforced by the termination of electric shock. As part of the research for my dissertation thesis, we attempted to replicate Miller's procedures. We found that this positioning, in fact, was merely a natural reaction to electric shock and had nothing to do with operant conditioning. Miller's interpretation of results was incorrect, largely due to his narrow focus on "observable data" and his unfounded claim that the data linked psychoanalytic "displacement" and "stimulus-response generalization." So it was that a trivial experiment in Miller's lab (not trivial, of course, to the rats) led to our shocking still more animals.

In our subsequent review of the literature, we discovered (somewhat unhappily, I recall) that O'Kelly and Steckle had reported the same findings as ours in 1939 in the paper entitled "A long enduring emotional response in the rat."(4) Several other earlier papers had also made the same observations with rats, including one by Daniel in 1943 and one by Richter in 1950.(5,6) So, in 1962, Azrin and I "proved" once again that animals will fight when wounded. When I told my Mennonite mother what I had discovered in my dissertation, she said, "Well, we know that. Dad always told us to stay away from wounded animals on the farm because they might hurt you."

Nevertheless, for the next ten years, I would demonstrate in countless ways that animals are more likely to be aggressive when they are hurt. I wrote on aggression, researched aggression, traveled throughout the world talking about aggression, even assisted the production of movies dealing with aggression. I helped design new strategies and equipment for shocking a multitude of helpless creatures. I even convinced children to shock some rats and "watch what happens."

It was about 1972 when I seriously confronted the painful truth: My research had merely demonstrated an already well-known phenomenon. Also, the "findings" had not significantly contributed to an understanding of human aggression. They were, more often than not, completely irrelevant to such understanding, because the conditions under which the animals were studied never really mirrored the human conditions about which I presumed to generalize.

How had I become so deeply involved in this enterprise? Animal experimentation is self-reinforcing. Each study leads to more studies, no matter how trivial or irrelevant the "findings." The voluminous literature continues to grow, with no end in sight.

Recently, a colleague from my department at Western Michigan University wrote in the Newsletter of the Association for Applied Behavior Analysis:

What about aggressive reinforcers rather than aggressive behavior? Perhaps the big deal isn't the electrically-shocked monkey's biting the rubber tube. Perhaps the big deal is the possibility that pressure on the monkey's teeth will reinforce whatever arbitrary response produces that pressure.

My behaviorist colleagues criticize theorizing about non-observables as cognitive thinking. Yet, it is such theorizing by behaviorists that causes problems for animals: To find out if my behaviorist colleague is correct, someone (probably a graduate student needing a thesis topic) will conduct yet more research. At no small cost in animal suffering, money, and time, those who might pursue his question of whether the painful shock or the sensation of biting reinforces aggressive behavior perpetuate the problem.

After reading my colleague's rhetoric, I cut off a piece of garden hose, taped a fork to it, and gave it to him as the apparatus for a study in which the method would be that he (the subject) should put the hose between his teeth, jam the fork into a light socket, then write up the results. He smiled and suggested that I was being aggressive.

My early career in animal research, like that of many scientists, shows a tragic irony. While artificially inducing aggression in coerced animals, I --not my subjects -- was the true exemplar of human aggression. For many years, I tormented, injured, and killed laboratory rats, bred to be docile. I was, in fact, responsible for so much animal suffering that my research had been identified by one animal protectionist as an example of cruelty.(7) What had reinforced my repeated mistreatment of innocent creatures? The pleasant sensations that accompany acceptance of grants, publication of papers, and presentation of guest lectures served as powerful reinforcers of my behavior.

Was my research worthless? While I now deny that my data on "aggression" in rats has shed new light on human aggression, my personal experiences in the laboratory represent a useful "case study" of human aggression. Skinner, although a behaviorist, recognized the importance of self-reflection in the study of human psychology. In Walden Two, Skinner's fictional student, discussing the importance of psychology, tells his professor, "It's a job for research, but not the kind you can do in a university, or a laboratory somewhere. I mean you've got to experiment and experiment with your own life, not just sit back in an ivory tower somewhere -- as if your own life weren't all mixed up in it."(8)

Unfortunately, Skinner and many of his followers have focussed their research and writing on behavioristic assumptions, ignoring Skinner's important insight. In contrast, Don Bannister has repeatedly called on his colleagues to acknowledge the importance of self-reflection in the study of psychology: (9-11)

There is a growing recognition in psychology generally that we may have to abandon our simple mimicry of the natural scientist and recognize that we cannot usefully experiment on our subjects, we have to experiment with them. Experimenting on animals offers us a way of delaying the day of that recognition.

We can now begin to see that in psychology the ethical issues involved in experimenting on animals are not separate from the scientific issues. It is not simply being argued here that it is an unkind practice to experiment on animals and that at the same time it happens to be a not very clever practice. The unkindness and foolishness stem from the same source, that is, from a particular notion of what 'being a scientist' is about in psychology, if psychologists continue to believe that a 'scientific' psychology must be 'objective' in the manipulative and non-reflexive sense of that term, then they will use those strategies which favour that kind of 'science'. Animal experimentation is such a strategy. It allows the psychologist to ride on the back of existing cultural and ethical permissions about., and gulfs between, species.

Picture the kind of psychologist who sits, notebook in hand, watching rats drown in a water-filled glass maze as they desperately strive to find the exit, thereby increasing his or her knowledge of 'learning under stress'. He or she is not simply personally indifferent to suffering but is trapped, as surely as the rat, within a total view bf science and of his or her own nature as a scientist. Psychology has failed in that it has given such a person no psychological view of his or her own character by the nature of his or her predicament.(11)

I still believe that psychology is an important field of study. However, as Skinner observes, you have to experiment with your own life. Each human thought and action is a manifestation of human psychology, from faIling in love to shocking rats. Using ourselves and our own lives as fodder for experiment, our understanding of human psychology may grow. The hope is that, with this greater understanding, human behavior may 'improve,' for the betterment of all. For example, we know that humans are capable of extremely aggressive and selfish behavior, and our self-indulgence has created obscene inequalities of wealth, We have wreaked ecological havoc, which has caused the extinction of countless other species and now threatens our own species. Surely, the first step in reducing human aggression is to take personal responsibility and reduce it in ourselves.

References

1. Skinner BF: Beyond Freedom and Dignity. New York: Alfred A. Kraopf, 1953.

2. Ulrich RE, Azrin NH: Reflexive fighting in response to aversive stimulation. J Exp Anal Behav 1962;3:511-520.

3. Miller NE: Theory and experiment relating psychoanalytic displacement to stimulus-response generalization. J Abnormal Soc Psychol 1948;43:155-178.

4. O'Kelly LE, Steckle LC: A note on long-enduring emotional response in the rat. J Psychol 1939;8:125-131.

5. Daniel WJ: An experimental note on the O'Kelly-Steckle reaction.. J Comp Psychol 1943;35:267-268.

6. Richter CP: Domestication of the Norway rat and its implications for the problem of stress. Assoc Res Nerv Ment Dis Proc 1950;29:19.

7. Rowan AN: Of Mice, Models, & Men. Albany, SUNY Press, 1984.

8. Skinner BF: Walden Two. New York, Macmillan Pub, 1948, pp 4-5.

9. Bannister D: The myth of physiological psychology. Bull Br Psychol Soc 1968;21229-231.

10. Bannister D: Science through the looking glass, in Perspectives in Personal Construct Theory. Cambridge, Unversity Press, 1970, pp 411-418.

11. Bannister D: The fallacy of animal experimentation in psychology, in Sperlinger D (ed): Animals in Research: New Perspectives in Animal Experimentation. New York, John Wiley & Sons. 1981.

Contents