Social Evolution Forum
FIND sef:
Welcoming Behaviorism Back into Evolutionary Psychology

This post refers to an ongoing series in This View of Life covering evolutionary psychology.

In my previous SEF post I commented on the unfortunate intrusion of us/them thinking in discussions about Evolution Psychology (EP). One example is the distinction that Leda Cosmides and John Tooby made between EP and what they called the Standard Social Science Model (SSSM), represented by figures such as B.F. Skinner in psychology and Clifford Geertz in anthropology. The sin committed by the SSSM was to treat human nature as a blank slate that anything could be written upon.

In some respects, Cosmides and Tooby were beating a dead horse with their critique of behaviorism (the tradition that Skinner represented). It had already been deposed in academic psychology by the so-called cognitive revolution. However, Cosmides and Tooby were also going beyond the cognitive revolution by positing a massively modular mind rather than the mind as a single domain-general computer.

Sign up for our newsletters

I wish to receive updates from:

BFSkinnerBehaviorism’s decline in academic psychology happened for a reason. Skinner and his associates tried to explain too much in terms of operant conditioning (e.g., language). In addition, their insistence that the mind should remain a black box that can only be studied in terms of input-output relationships was unsustainable. But behaviorism did not fade into the sunset. It remained alive and well in the applied behavioral sciences, including branches that go by names such as behavioral therapy, applied behavioral analysis, prevention science, and contextual behavioral science. When cognitive science started to be applied to practical contexts, it was added to behaviorism rather than replacing it, as the well-known term Cognitive Behavioral Therapy (CBT) implies. More recently, mindfulness-based techniques have been added to CBT in what is often called third wave behavior therapy.

The fact that behaviorism remains at the core of the applied behavioral sciences tells us something important. A colleague of mine once quipped that academic science is where the rubber meets the sky. Applied science is where the rubber meets the road. If behaviorism remains at the core of the applied behavioral sciences, it is because it works and we can’t accomplish positive behavioral change in real-world situations without it.

Triple PAnother conceit of academic science is that applied science is methodologically inferior. I have become involved in applied science enough to learn that, methodologically, it includes the best and the worst. The worst is very bad but the best is breathtakingly good. One of my favorite examples is Triple P (for Positive Parenting Program), a method for improving parenting practices at the population scale implemented around the world. In one experiment,1 Triple P was implemented in 9 counties in South Carolina selected from 18 counties in a randomized design. The outcome variables were public health statistics such as foster care placements and emergency room visits by children. Triple P demonstrably improved parenting practices according to these metrics at a cost of approximately eleven dollars per child, a pittance compared to the savings from preventing the fallout of bad parenting practices. Triple P relies heavily on good old-fashioned operant conditioning.

The core of truth in behaviorism was clearly articulated by B.F. Skinner in a 1981 Science article titled “Selection by Consequences.”2 The abstract is worth quoting in full.

Selection by consequences is a causal mode found only in living things, or in machines made by living things. It was first recognized in natural selection, but it also accounts for the shaping and maintenance of the behavior of individuals and the evolution of cultures. In all three of these fields, it replaces explanations based on the causal modes of classical mechanics. The replacement is strongly resisted. Natural selection has now made its case, but similar delays in recognizing the role of selection in other fields could deprive us of valuable help in solving the problems which confront us.

desert camoflageTo unpack this statement, let’s think about the adaptationist program in evolutionary biology, which plays a central role in Cosmides and Tooby’s vision of EP. To the degree that phenotypic variation is heritable, the phenotypic properties of organisms can be predicted by the environmental forces that molded them, not from their physical makeup. Many desert species will be sandy-colored to avoid being seen by their predators and/or prey. They will be light if the sand is light and dark if the sand is dark. The prediction can be made for any desert species (snails, insects, reptiles, mammals, birds) even though these species have different genes and physical exteriors. Assuming the existence of heritable variation is equivalent to assuming that organisms are blank slates with respect to the molding action of natural selection. Whenever evolutionary biologists think in adaptation mode, they are employing a blank slate assumption of their own.

Skinner was saying that the open-ended learning of single organisms and the open-ended capacity of cultures to change their properties over time can be regarded as evolutionary processes, similar to genetic evolution. If so, then their properties need to be understood in terms of a past history of selection, not their physical makeup, or “the causal modes of classical mechanics” as he put it.

operant conditioningIn operant conditioning, behaviors are selected by reinforcers such as pleasure, pain, or social approval. Skinner posited that reinforcers evolved by genetic evolution to select behaviors over the short term that are genetically advantageous over the long term, at least on average. This means that operant conditioning is a fast-paced evolutionary process that evolved by genetic evolution. If we want to know why an organism behaves as it does, we need to understand its past history of reinforcement, just as we need to know the past history of genetic evolution to explain the anatomical traits of a species. This is why Skinner said that selection by consequences was already accepted for genetic evolution but still needed to make its case for individual learning and cultural change.

The blank slate assumption for heritable genetic variation, individual learning, and cultural change is only a partial truth. To pick genetic evolution as an example, not all phenotypic variation is heritable: single traits are connected to other traits during development, a trait can be a byproduct of another trait, and so on. Evolutionists debate these issues all the time but at the end of the day, to the degree that organisms are adapted to their environments, we must understand their properties in terms of selection by consequences. In the same fashion, we can back away from the blank slate assumption while still recognizing the importance of selection by consequences for operant conditioning and cultural change.

Against this background, Cosmides and Tooby’s us/them distinction between EP and the SSSM was a major wrong turn. There is simply no warrant for rejecting “selection by consequences” thinking for individual learning and cultural change while indulging in it to the hilt for genetic evolution. Worse, Cosmides and Tooby are still racing off in the wrong direction after all these years. Tooby’s most recent statement is an essay written for a book titled This Idea Must Die: Scientific Theories that are Blocking Progress. His essay, which is also available on the internet site us vs, is titled “Learning and Culture”. It is a fine specimen of us/them thinking and an ode to neural and cognitive mechanisms without hinting that “selection by consequences” thinking can play the same role for the study of individual and cultural change that it does for genetic evolution. If you want to understand why many psychologists (broadly defined) distance themselves from the EP label, look no further than Tooby’s essay.

Recognizing the valid elements of behaviorism and especially Skinner’s key insight about selection by consequences goes a long way toward establishing the face-value definition of EP as “the study of psychology from an evolutionary perspective” and detaching it from any given school of thought, such as that espoused by Cosmides and Tooby. It is important to stress that I do not reject all aspects of Cosmides and Tooby’s school of thought, any more than I accept all aspects of behaviorism (see my previous post). The point is to overcome us/them thinking, to examine the elements of any particular school of thought on a case-by-case basis, and let the chips fall where they may.

Nurture effectAn article titled “Evolving the Future: Toward a Science of Intentional Change3 that I co-authored with three applied behavioral scientists, Steven C. Hayes, Anthony Biglan, and Dennis Embry, welcomes behaviorism back into Evolutionary Psychology. It is published in Behavioral and Brain Sciences, a journal that features commentaries on each target article so that readers can learn the opinions of several dozen of our peers along with our own views. Anthony Biglan’s new book, The Nurture Effect: How the Science of Human Behavior Can Improve our Lives and Our World, is another way to learn about the importance of “selection by consequences” thinking for understanding and positively influencing behavioral and cultural change. He summarizes his book in a TVOL article titled “Selection by Consequences: Recovering Skinner’s Key Insight about Learning as an Evolutionary Process” that adds to this post and richly deserves to be part of TVOL’s theme on Evolutionary Psychology.

At the forthcoming annual meeting of the Northeastern Evolutionary Psychology Society (NEEPS), which will be held in Boston during April 9-12, I will be taking part in a symposium that is designed to integrate two societies for applied science. The first is the Society for Applied Evolutionary Psychology (AEPS), which numbers a few dozen individuals. The second is the Association for Contextual Behavioral Science (ACBS), which numbers over 7000 individuals. The relative sizes of the two societies indicates the degree to which ACBS has proven itself where the rubber meets the road and the degree to which AEPS is just starting out. The two societies and their perspectives have much to offer each other, but the self-described evolutionary psychologists must be prepared to function in the role of learners in addition to that of teachers in the creation of a single intellectual community—an “us” in which the only “them” is the ideas that have been falsified by the scientific process.


  1. Prinz, R.J., Sanders, M.R., Shapiro, C.J., Whitaker, D.J., & Lutzker, J.R. (2009). Population-based prevention of child maltreatment: The U.S. Triple P System Population Trial. Prevention Science, 10, 1-13.
  2. Skinner, B. F. (1981). Selection by Consequences. Science, 213, 501–504.
  3. Wilson, D. S., Hayes, S. C., Biglan, A., & Embry, D. (2014). Evolving the Future: Toward a Science of Intentional Change. Behavioral and Brain Sciences, 37, 395–460.









Join the discussion


  1. Anon says:

    “One of my favorite examples is Triple P (for Positive Parenting Program), a method for improving parenting practices at the population scale implemented around the world.”
    Unfortunately, Triple P may not be one of the best examples of successful applied behaviorism in action. See, for example:
    link to

  2. Fred says:

    Welcome back? We were here first 😉

    In seriousness, I don’t think you’ll find any behaviour analysts who would argue against studying psychology from an evolutionary perspective (even if we’re typically more interested in demonstrating functional relations in the here and now) but some of the conceptualisations of adaptations as modules and other mentalistic terms would be regarded as problematic by most behaviour analysts.

  3. […] in my friends (Laureate Education, n.d.). And while most educational theories (like cognitivism, behaviorism, and constructivism) are based on the premise that learning takes place within a person, Siemens […]

  4. Aaron says:

    Thanks for this article. It’s disappointing that the views of different disciplines and scientists are so often misrepresented and disparaged by those who have not even read the relevant primary source documents. Science discourse should be conducted with more discipline than often occurs, people should recognize that just because one focuses his/her research on determining how one set of influences on behavior (e.g., the history of relationships between an organism’s behavior and environmental contexts) operates does not mean that he/she denies that other influences (e.g., genes) are important, and scientists should work hard to understand what another discipline is actually saying while recognizing it’s unlikely that all of the people saying these things are complete idiots. Yes, if something sounds incredibly idiotic, it of course may be, but there is a better chance you’re just misunderstanding what is actually being said and lacking the requisite background knowledge. I applaud your efforts towards consilience.
    “No reputable student of animal behavior has ever taken the position that the animal comes to the laboratory as a virtual tabula rasa, that species differences are insignificant, and that all responses are about equally conditionable to all stimuli” – Skinner, 1966. link to

  5. Justin E. Lane says:

    I have to say, I originally thought this was a spoof article. Starting with Chomsky it was shown that behaviorism was subsumed by a more powerful approach (cognitive science). What do we gain from attempting to resuscitate this lesser approach? It seems that the study of human behavior gains nothing, but those who wish to state that behavior is the result of selection by consequences now have a way of dealing with human psychology-this is a return to an axiomatic claim of some but seems to be underspecified overall. The issue with behaviorism has already been addressed by Tooby and Cosmides in the integrated causal model presented in 1992 and has been addressed in other areas of human behavior (such as Robinson 2009 dealing with religion). The cognitive approach, generally, has been shown to be better able to explain a wider range of psychological and behavioral phenomena than those who attempt to state that it is all selection and variation. I just don’t understand why walking backwards to behaviorism and social Darwinism is at all a good turn for the study of human behavior. All it seems to do is attempt to provide some theoretical congruence between behavioral studies and cultural evolution; possibly, it is better if we just reject cultural evolution since it is consistently underspecified in its predictions.

    • curious says:

      There is no question research in the area of cognitive science has produced many important findings and findings that behavior analysis simply wasn’t and may still not be set up to examine even in its modern iterations. However, cognitive science is also too often plagued by descriptions masquerading as explanations, and efforts at reproducing results have proven difficult or futile likely in no small part due to poorly operationally defined, highly variable terms. On the whole, you will find stronger effect sizes in behavior analytic approaches to identify functional relationships. If you work in an applied setting with real people (or other organisms) and want to actually explain, predict AND INFLUENCE individual behavior with precision, scope and depth, science with roots in behaviorism is, currently, at least, more likely to get the job done than cognitive science. Of course, no science dealing with behavior gets all of these jobs done by itself. A strong case can be made that “cognitive science” is most effective when it, like behaviorism, at least attempts to act like a natural science link to , and it’s questionable at best whether Chomsky’s (basically essentialist) account of language was any better than Skinner’s link to I am not an expert in the area of cultural evolution and as such don’t believe I’m in a position to evaluate your claim that it is “consistently underspecified” in its predictions. This may well be true. However, what predictions does the approach you endorse do a better job of?