Operant Learning and Selectionism: Risks and Benefits of Seeking Interdisciplinary Parallels

Behavior Analysis Program
Department of Psychology
Western Michigan University

Note: This article originally appeared in the Behavioral and Brain Sciences

Abstract: Seeking parallels among disciplines can have both risks and benefits. Finding parallels may be a vacuous exercise in categorization, generating no new insights. And pointing to analogous functions may cause us to treat them as homologous. Hull, Langman, and Glenn, (2001) have provided a basis for the generation of insights in different selectionist areas, without confusing analogy with homology.

Those of us concerned with operant behavior are generally called behavior analysts, or informally,Skinnerians. And, though most of us are psychologists, we are somewhat outside the mainstream of psychology because of our reluctance to make much use of intervening variables, hypothetical constructs, reifications, or more derogatorily, explanatory fictions, the fundamental building blocks of most psychology. So, we behavior analysts have taken great intellectual and perhaps spiritual comfort in finding a home on the selectionist continuum between Charles Darwin’s evolutionary biology and Marvin Harris’ cultural materialism (Harris argues that cultural practices survive if they contribute to the survival of the groups that practice them).

Most of us behavior analysts have found sufficient comfort based on the popular writing of Steven Jay Gould (e.g., 1983) concerning biological evolution and Marvin Harris (e.g.. 1984) concerning cultural materialism, that we have not studied the technical writings in either field. Sigrid Glenn, is among the few behavior analysts who are not so easily comforted; therefore, she has she has studied the selectionist continuum in greater depth and more thoughtfully, than most of the rest of us behavior analysts. This sophistication is apparent in her contribution to the article on which I am commenting (Hull, Langman, & Glenn, 2001).

It is a fascinating exercise to find the parallels between one approach or discipline and another. And behavior analysts have been previously fascinated in that manner, as exemplified by their demonstrations of the parallels between Freudian psychoanalysis and Skinnerian behavior analysis or between economic analysis and behavior analysis, as well as between evolutionary biology and the acquisition and maintenance of response classes.

But this exercise in parallelisms is not without its risks. First, we risk not going far enough with the parallels, not exploiting them sufficiently. Given that we have demonstrated a parallel between two disciplines, what new insights does that give us into either discipline. Does the parallel suggest new causes or new functional relationships? Does it suggest new categories or classifications? Does it point to new dimensions of either the independent or dependent variables we have ignored or to dimensions we have considered important but now should recognize as trivial? The parallels drawn by Hull, Langman, and Glenn, (2001), do not yet give us these sorts of new insights into behavior analysis; but they may have delineated a model that may be productive of such insights in the future.

Second, there is the opposite risk, the risk of taking the parallels too literally. The authors attenuate this risk by saying, “The processes by which operant adaptation occurs are viewed here as analogous to the processes by which biological evolution occurs,” and “When operant behavior is seen as the figure, against organism as ground, the elements involved in selection processes are analogous to (not the same as) those involved in gene-based biological evolution.” Unfortunately, such disclaimers can be quickly forgotten, in the heat of intellectual discourse or in the heat of practical applications. The transient nature of disclaimers is frequently illustrated with disclaimers in the form of operational definitions; for example psychologists often operationally define “intelligence” as “the score obtained on an IQ test”; and then, within two or three sentences, they have reified intelligence into an internal, causal agent.

To prevent taking parallels to literally, I find it helpful to distinguish between “analogous” and a somewhat extended notion of “homologous,” where two events are analogous, if they serve the same function of have the same effect; and where they are homologous, if they are based on the same underlying mechanisms or processes. For example, a rat’s lever pressing increases in frequency, if that pressing is followed immediately by a drop of water, the directly underlying process being reinforcement. And our commentary writing might increase in frequency, if that writing is followed within a few months by publication, the directly underlying process being a rule-governed analog to reinforcement—it looks like simple reinforcement; it acts like simple reinforcement (same behavior-increasing function), but it is much more complex and requires sophisticated language skills (here I am not talking about the language skills needed for actually writing the commentary).

Similarly, the selectionism of operant learning is only analogous to the selectionism of evolutionary biology, it is not homologous; the behavioral and biological processes underlying operant learning differ from those underlying biological evolution. However, it makes sense that both biological evolution and operant learning should be selectionistic. Something like selectionism seems almost required, in both cases; otherwise, life would be much more chaotic than it is. But it also makes sense that these two examples of selectionism have evolved (been selected) as independent reactions to the demands of survival in the same environment. Biological evolution and operant learning have evolved analogously, not homogonously.

In conclusion and as a behavior analyst, I think the authors have provided a valuable service in separating the essential features of a selectionist model (i.e., replication, variation, and environmental selection) from the unessential features of specific instances of that model (e.g., replication of organisms across space, in the case of biological evolution, vs. replication of responses across time, in the case of operant learning).

References

Gould. S. J. (1983) Hen’s Teeth and Horse’s Toes. W.W. Norton.

Harris, M. (1974) Cows, pigs, wars and witches. Vintage Books.

Hull, D. L., Langman, R. E., & Glenn, S.S., (2001) A general account of selection: Biology, immunology and behavior. Behavioral and Brain Sciences. 24