Guest post authored by William Baum
Behavior analysts, including clinical and applied behavior analysts, should know about the Molar View of Behavior, because it offers a much better conceptual framework for thinking about theory and practice than the older molecular view. Much has changed in our understanding of behavior since Skinner’s (1938) Behavior of Organisms. Back then, Skinner and others received training that centered reflexes as the model unit of behavior. Accordingly, when a stimulus appears, a response occurs, as Pavlov had stated. One of Skinner’s great innovations was the concept of operant behavior, which is behavior under control of its consequences, rather than just antecedent stimuli. Unfortunately, in inventing operant behavior, Skinner retained the reflex concept of the discrete response and the primacy of contiguity, particularly contiguity between the response and its reinforcer. As a result, behavior analysts became saddled with the notion of “operant” as the discrete response that produced the reinforcer.
Many students of behavior analysis continue to be taught this inaccurate and unnecessarily constraining model—the same one I was taught in the 1950s. I say “inaccurate” because research has established that: (1) behavior cannot be forced into the mold of discrete responses; and (2) reinforcement does not consist of mere temporal closeness between behavior and consequence. Following Herrnstein’s (1961) report of the matching relation between relative activity and relative consequences, research showed again and again that discovering order in behavioral processes required thinking in terms of rates: rates of activities and rates of consequences. In retrospect, perhaps no one should be surprised that rates would be required, because organisms must be sensitive to rates if they are to survive and reproduce. Even though Skinner wrote about rate as a dependent variable, he failed to see that rate would matter as an independent variable.
In 1969, a paper by Howard Rachlin and me came out reporting a study of choice that resulted from our speculation that the measure of activity rate should be time, rather than discrete responses. We showed that choice actually consisted of time allocation among alternative activities and that discrete responses were unnecessary. Moreover, we argued, discrete responses are an illusion created by the peculiar apparatus we had inherited from Skinner: the operation of a switch as if this were a unit of behavior. We argued that a switch operation could only indicate a certain amount of time in an activity, it was not itself a unit.
This line of thinking casts behavior as consisting of temporally extended activities. Instead of a lever press, we have lever pressing, instead of solving a math problem, we have solving problems. Moreover, many activities cannot be reduced to discrete responses. For example, watching television and running a footrace only have parts that are the same as the whole except for duration. Other activities, like playing tennis or disrupting a classroom, have parts that are themselves activities on a smaller time scale. Playing tennis has parts like serving and returning balls. Disrupting a classroom has parts like shouting, running around, and hitting others.
This way of thinking about behavior allows us to see our connection to evolution. Activities do not come from thin air. They originate in our environment, present and past. Many activities have their origin deep in the history of our species. They are with us today as responses to situations that impinged on the likelihood of surviving and reproducing. If you have not eaten for a day, you respond strongly to food, cooked, but also fruits and vegetables. If you encounter a dangerous animal, a mountain lion or grizzly bear, how do you respond? Your responses are often irrational and powerful.
Our irrational and powerful responses help to understand activities that cannot be considered operant behavior. If they are related to consequences, those were long ago in the history of our species, in phylogeny. So, they are non-operant activities. They are specific activities induced in specific situations. In the laboratory, events that might be called “reinforcers” or “punishers” in other circumstances induce patterns of activity just by their presence. A rat receiving occasional bits of food chews on anything it can get hold of and drinks inordinate quantities of water. Humans likewise engage in odd non-operant activities, for example, when waiting for an elevator or a baby. In classrooms and clinical settings, various environmental features induce seemingly intractable non-operant activities. Someone who has only the concept of “reinforcement” might call them “automatically reinforced,” or worse, “self-reinforced,” but these terms are oxymorons; reinforcement is nothing if not contingent. So, what can help?
The explanation of non-operant activities like self-harming or self-injuring behavior (SIB) lies in features of the environment that might not be immediately obvious. One recent study showed that “challenging behavior” is most often induced by a change in context—for example, change in the therapist or task (Aggarwal et al., 2026). Such activities are better understood as induced by environmental factors than as reinforced by consequences.
Here I want to emphasize the usefulness of the concept of induction, first defined by Segal (1972) and which I have extended in ways that she agreed with (personal communication). The key concept about inducers is “just by their presence.” This includes relations that might have been called elicitation in the past, but elicitation carries connotations of immediacy and discrete responses, whereas induction applies to extended patterns of behavior, activities. An organism encountering a predator may flee, freeze, or fight. A person encountering a grizzly bear may start to flee, but might freeze or fight, because grizzly bears are fast. The person who encounters her mother-in-law flees or fights and may avoid danger that way. Then that activity may shift from non-operant activities to operant avoidance. If avoidance is successful, it will afterwards by induced by such dangerous situations (Baum, 2020).
Not only may induction apply to avoidance, it also applies to what has been called “positive reinforcement.” On encountering a prey item or potential mate, an organism engages in activities that make approach and interaction more likely. A potential mate induces courtship and flirtation, which may become operant activities in the form of dating. In clinical settings, viewing videos turns out to be a powerful inducer, and when a desired activity leads to viewing, that activity is induced by the videos it produces.
What exactly is an “activity”? So far we have seen an activity as an extended pattern of behavior. More than that, an activity is what philosophers call an “individual” and a “process” (Baum, 2018, 2024). Such an entity is an integral whole with parts that work together to make the whole functional, and the parts themselves are processes on a smaller time scale. For example, serving a tennis ball requires me to position my feet, toss the ball up, raise and bring down my racquet, all to work together to move the ball at high speed over the net. Compare this to “challenging behavior,” which likewise comprises various activities that function together to challenge the demands of the therapists.
A process is a change through time that has a beginning and an end. An organism embodies many processes. Physiological processes just keep the organism intact while it is alive. Behavioral processes are the ones that have effects in the environment—that gain helpful resources and avoid harmful events. “Helpful” here means promoting survival and reproduction, for example, food, shelter, mates, offspring. “Harmful” means threatening survival and reproduction, for example, predators, illness, and injury.
Not all activities and inducers affect surviving and reproducing directly. Those inducers that need little or no experience (unconditional) to induce behavior derive their inducing power from a history of natural selection—phylogeny. Hence I call them phylogenetically important events (PIE)—food, shelter, predators and so on. PIEs induce either approach and consumption (“helpful” PIEs) or avoidance (“harmful” PIEs). Other inducers (conditional) affect surviving and reproducing indirectly. They depend on experience of correlation or covariance between a signal or stimulus and a PIE. Covariance with a PIE causes a stimulus to induce some of the activities that the PIE itself would induce. In particular, a conditional inducer or proxy induces operant activities that produce the PIE. Thus, if challenging behavior results in task avoidance, changing task will likely induce challenging behavior.
Earlier publications summarized this framework with three laws, the laws of allocation, induction, and covariance (e.g., Baum, 2018, 2024). The law of allocation states that the proportion of time taken by an activity equals its relative competitive weight. The reasoning behind it is that: (a) activities take time; (b) an organism’s activities take all the time available; (c) time is always limited; and so, (d) activities compete with one another for time. More time disrupting a classroom necessitates less time on task. The law of induction states that the competitive weight of an activity depends on how strongly PIEs and their (conditional) proxies induce the activity. If one can discover what induces an activity, desired or undesired, one may be able to increase or decrease its competitive weight. The law of covariance states that signals and activities that covary with PIEs become conditional inducers and operant activities, respectively.
I have authored three books related to this Molar View of Behavior. The first, Understanding Behaviorism: Behavior, Culture, and Evolution, gives an overview of behavioral treatments of philosophical questions and social issues. I wrote it as a textbook for an undergraduate course called “Behaviorism.” The second, Science and Philosophy of Behavior, is a collection of my theoretical papers from 1973 to 2020. This book is for anyone interested in seeing the more technical side of my thinking and its evolution.
The third book, Introduction to Behavior: An Evolutionary Perspective, is intended for all students in behavior analysis, both basic and applied, who might want to understand behavior, its origins, and its control in this molar view. The book lays out the basics of a science of behavior in the molar perspective without a tedious historical introduction. This book I recommend to all students who want a coherent framework for understanding and modifying behavior.
References
Aggarwal, I., O’Brien, M. J., Pauls, A. M., Jeglum, S. R., Franck, C. T., Martinez-Perez, C. N., & Podlesnik, C. A. (2026). Renewal of challenging behavior in an intensive outpatient clinic: Replication and extension to task changes. Journal of Applied Behavior Analysis, 59, https://doi.org/10.1002/jaba.70057.
Baum, W. M. (2017). Understanding behaviorism: Behavior, culture, and evolution (3rd. ed.). Malden, MA: Wiley Blackwell Publishing.
Baum, W.M. (2018). Multiscale behavior analysis and molar behaviorism: An overview. Journal of the Experimental Analysis of Behavior, 110, 302-322. https://doi.org/10.1002/jeab.476.
Baum, W. M. (2020). Avoidance, induction, and the illusion of reinforcement. Journal of the Experimental Analysis of Behavior, 114, 116-141. https://doi.org/10.1002/jeab.615.
Baum, W. M. (2022). Science and philosophy of behavior: Selected papers. John Wiley and Sons.
Baum, W. M. (2024). Introduction to behavior: An evolutionary perspective. John Wiley and Sons.
Baum, W. M., & Rachlin, H. C. (1969). Choice as time allocation. Journal of the Experimental Analysis of Behavior, 12, 861-874. https://doi.org/10.1901/jeab.1969.12-861.
Herrnstein, R. J. (1961). Relative and absolute strength of response as a function of frequency of reinforcement. Journal of the Experimental Analysis of Behavior, 4, 267-272. https://doi.org/10.1901/jeab.1961.4-267.
Segal, E. F. (1972). Induction and the provenance of operants. In R. M. Gilbert & J. R. Millenson (Eds.), Reinforcement: Behavioral Analyses (pp. 1-34). New York: Academic. https://doi:10.1016/B978-0-12-283150-8.50006x.
Skinner, B. F. (1938). Behavior of organisms. New York: Appleton-Century-Crofts.