Why do psychologists use animal models of human clinical problems? One reason researchers turn to animals is that they are prohibited from conducting certain experiments on humans; the ethics of human research purportedly dictate that animals be used. The implicit assumption is that what can’t be done to humans, can be done to animals (e.g. McKinney 1974a, p. 530). Ethical considerations regarding animal research, when raised, are considered outweighed by potential human benefits.
Another rationale for conducting animal experiments is that certain experiments are more easily conducted on animals than on humans. For example, a prospective study of the effects of early experience on adult behavior may take twenty years to complete on humans, given their slow development, but only five years on monkeys.
Both the ethical and practical constraints on human research are apparent in Stephen Suomi’s justification for animal models:
… animal models can provide investigators with experimentally rigorous research tools with which ethically delicate and practically difficult questions about human psychopathology can be addressed (Suomi 1984a, p. 228).
One can ask why experiments that are prohibited or impractical on humans should be conducted on animals. The answer often involves the experimental psychologist’s reliance on highly manipulative, controlled experiments. Indeed, these psychologists sometimes convey a sense of disdain for clinical or naturalistic observations.
The use of animal models is not without its own difficulties, however. One obvious problem is that different species do not necessarily react the same way to a given experience. Therefore, conclusions about one species may not generalize to a closely related species, much less to humans. A second problem with many animal models is the artificiality of laboratory conditions, especially in research by experimental psychologists. These conditions yield results of questionable relevance to the natural characteristics of the animal model species, let alone those of humans.
The animal modelers themselves are increasingly recognizing shortcomings of animal models. Consider again the remarks of Suomi, who has conducted numerous maternal deprivation experiments and championed the use of animal models:
… in virtually no case is an animal model a perfect . . . replica of the human disorder under study Rather, it is usually a highly simplified, theoretically biased, and incompletely generalized version, expressed in members of a nonhuman species. An animal model is almost never the “real thing”; it is, instead, only a model of the real thing (1982, p. 250).
… the primary rationale for creating most animal models lies not so much in any obvious and impressive strengths of such models as it lies in the problems inherent in conducting research with humans as subjects (p. 250).
The experimental psychologists are therefore attempting to make the best of a bad situation. They can’t conduct their reductionist experiments on humans, yet they don’t have good substitutes for human subjects. This predicament is reflected in Harlow’s adage that “you’re crazy to use animal models and you’re crazy not to” (Kornetsky 1977, p. 498).
How can animal modelers deal with the problems inherent in using animal models? McKinney and Bunney (1969) provided a set of guidelines for modelers of human mental disorders. According to these authors, the animal model and the human disorder should have similar, if not identical, (1) causes, (2) symptoms, (3) biological mechanisms, and (4) cures. The more criteria satisfied by the model, the greater its scientific validity. Unfortunately, these criteria are often overlooked in maternal deprivation research.
Please cite as:
Stephens, Martin L. (1986) Maternal Deprivation Experiments in Psychology: A Critique of Animal Models.
Retrieved from http://aavs.org/maternal-deprivation-experiments-psychology.