B.F. Skinner and Radical Behaviorism

There is much incorrect information about B.F. Skinner out there on the web, and even sometimes in psychology classrooms. In my previous blog, I gave a quick overview of his views, called radical behaviorism. Some of the things I said about B.F. Skinner are not quite accurate, and although I am not a radical behaviorist, it is only right that I set the record straight. Skinner has been the subject of much criticism, including Noam Chomsky's famous review of Verbal Behavior which helped launch cognitive psychology and Steven Pinker who dismisses radical behaviorism as a "blank slate" philosophy. He has also been heavily criticized by Daniel Dennett as a "greedy reductionist" who desired to reduce everything about human behavior to conditioning (for the record, Dan Dennett is a reductionist but uses the term greedy reductionism to describe the tendency of some to try and oversimplify a phenomenon). To what extent are these claims true?

It is true that B.F. Skinner did not consider emotions as something necessary for explaining behavior. Emotions are merely a byproduct of conditioning, and in order to understand the cause of someone's seemingly emotional behavior what we really must understand are the contingencies which led to the behavior in a situation. This was most eloquently put when he wrote that:

 As long as we conceive of the problem of emotion as one of inner states, we are not likely to advance a practical technology. It does not help in the solution of a practical problem to be told that some feature of a man's behavior is due to frustration or anxiety; we also need to be told how the frustration or anxiety has been induced and how it may be altered.¹

Skinner did concede that sometimes emotion could be used as an explanation for something, with the example he gave being the development of ulcers under stress. He wrote that:

A chronic emotional condition sometimes leads to certain forms of illness. For example, a man whose business is failing may be subject to a long series of circumstances which generate a chronic condition of frustration or anxiety. Part of the total emotion may be reflex responses in the alimentary tract, as a result of which the man may become physically ill—he may develop ulcers, for example. Here it is legitimate to attribute the illness to an "emotion" as a cause, because we define the emotion as a pattern of behavior. We might in the same way attribute a cracked skull to emotion if the injury was suffered as the result of reckless behavior. But this is very different from arguing that emotional behavior is due to an emotion. A man does not neglect his business because of anxiety or worry.²

This is actually a very important point that he makes about the scientific analysis of emotions. To what does the term "emotion" actually refer? In order to study anything, an operational definition must be provided for the abstract concept. Skinner is defining emotion as the byproduct of the stimulus-response contingencies which control behavior. So, when Skinner stated that ,"We don't say that the human subject gambles to punish himself as the Freudians might say or gambles because it feels exciting when he does so, nothing of the sort, people gamble because of the schedule of reinforcement that follows³," he did not mean that the gambler does not feel excitement from the gambling, but rather that the excitement is a byproduct of the reinforcement.

It may sound strange, but operational definitions can be rather specific. A neuroscientist might operationally define emotions as certain neurochemical and physiological states. Skinner conceived of psychology as being a separate science from neuroscience however, not to mention that in his life time advanced methods for studying the brain such as functional magnetic resonance imaging and diffusion tensor imaging had not yet been developed. He optimistically wrote that, "Eventually a science of the nervous system based upon direct observation rather than inference will describe the neural states and events which immediately precede instances of behavior.⁴"

So, it seems that Skinner's lack of much use for neuroscience stemmed more from the fact that it was so new as to not offer much insight about behavior, while the methods of behavioral analysis could show what environmental factors interact with whatever mediating neural processes there which result in behavior. However, those neurological facts are mere details. What is important are those environmental factors. He wrote that, "we do not have and may never have this sort of neurological information at the moment it is needed in order to predict a specific instance of behavior.⁵" He also praised Ivan Pavlov, saying that his greatest "achievement was the discovery, not of neural processes, but of important quantitative relations which permit us, regardless of neurological hypotheses, to give a direct account of behavior in the field of the conditioned reflex.⁶" Skinner was a practical man who wanted to understand how behavior originates and how it could be changed without having to wait around for the brain to be understood in order to do so.

Skinner has also sometimes been accused of ignoring the rest of biology and has been strongly associated with the concept of "the blank slate." Skinner was the subject of much criticism by Steven Pinker in his book The Blank Slate (which entirely misrepresents behaviorism and mostly serves to perpetuate myths about behaviorism; an analysis of this can be read here) for this reason. Skinner did not believe that we (or animals for that matter) lacked innate behaviors. One example is when Skinner referred to innate navigational instincts in animals; "But the journey of the eel from the Nile River to the Sargasso Sea, a quarter of the way around the earth, is a much more complex example of innate behavior and is much harder to explain in "cognitive" terms.⁷" Nor did Skinner deny how powerful evolution was in shaping our behavior, stating that "Anyone who has seen a slip of a plant grow into a complete plant will have no difficulty in accepting the role of natural selection in the origin of behavior, no matter how complex.⁸"  Skinner also stated from observing the offspring of rats who were good at navigating mazes being particularly talented at maze navigation, "from the observed facts alone, it was impossible to tell whether the difference was due to operant conditioning, following some such procedure as just described, or whether it was genetic.⁹" It is true that Skinner preferred explaining things in terms of operant learning due to the fact that such hypotheses are more readily testable. He stated that "Contingencies of reinforcement which shape ontogenic behavior can be arranged and studied in the laboratory. Most of the contingencies of survival responsible for phylogenic behavior observed in the field are merely inferred.¹⁰"

It is likely that Skinner would be critical of the modern field of evolutionary psychology were he currently alive, but mostly because of its strong association with computational theories in cognitive psychology. There is no clash between the study of operant conditioning and other forms of stimulus learning and sociobiological studies of phylogenetic behavior.

So, if Radical Behaviorism is not the dogmatic embrace of the blank slate, the attempt to reduce all human behavior to operant conditioning, or the denial of emotions, then what is it? Radical behaviorism is simply the systematic attempt to explain all behavior with the principles of conditioning that can be explained this way without using neurological or mentalistic terms. Radical behaviorism differs from the methodological behaviorism of John Watson because Watson's behaviorism treated the mind as an entity unreachable to science. Implicit in this is dualism, the thesis that the mind is fundamentally of a different substance and class of existence of the body, whereas Skinner was a materialist and thought that what we call thinking cannot be something unreachable to science. Skinner goes into detail about the differences between his and Watson's behaviorism in About Behaviorism. Skinner also differs from the proto-cognitivism of Edward Tolman, or the "drive theories" of Clark Hull and Kenneth Spence in that these forms of behaviorism posited hypothetical mediating structures that shaped behavior, whereas Radical Behaviorism sought to avoid them and focus only on the ways the environment shaped the behavior.

from other applications of behaviorist methods in that it is very strict about not referring to mentalistic terms, unlike the theories of other behaviorists like Clark Spence and Kenneth Hull, or the early cognitive theories of Edward Tolman. It also was unconcerned with neurology, unlike the behaviorist and neuroscientist Karl Lashley. In truth, radical behaviorism is not all that radical, but is just a pragmatic program for analyzing behavior without having to wait around for the nervous system to be figured out.

In conclusion, many of the claims about B.F. Skinner are exaggerated or a inaccurate. This is true for some of the statements in the previous blog post. It has been shown that Skinner did not deny the existence of emotions, innate or "phylogenic" behavior, or that the brain was important for the execution of behavior. Although it is impossible to know for sure, if Skinner were alive today, it is possible that he would have ended up embracing neuroscience, while being an active critic of the computational theory of mind and concepts such as the language of thought associated with it. He might even go so far as to welcome the lines of research known as embodied and embodied cognitive science (while likely rejecting the cognitive label).


Postscript note:

While modern psychology has moved past this paradigm, there is much to be learned from the approach that seems to have been forgotten. One thing that is not a myth is that it IS true that he thought the principles of operant conditioning could be used to create a modern utopia, which was expressed in his book Beyond Freedom and Dignity. At root, the book defends the idea that the principles of conditioning are so powerful that in order to make the world a better place you cannot merely express ideas in an eloquent speech to "change the minds and hearts of men and women rather than the world in which they live. ¹¹" Perhaps he does take it to a bit of an absurd extreme, but hey, no one is perfect. The biggest flaw in Skinner's approach is that it seems ill equipped to explain phenomena of a more cognitive nature, such as hallucinations for Schizophrenics. While the psychiatric models based on behaviorism for therapy are extremely effective, there's nothing you can do to condition away hallucinating pink elephants. It is only through studying neuroscience that we will learn the causes of such problems and learn ways to treat them, but given what I presented in this blog, I don't think Skinner would necessarily disagree. A neuroscientific understanding of behavior and "the mind" is a natural step forward from behaviorism that avoids the mentalism that Skinner critiqued. Just as well, while I share some of Skinner's concerns about cognitive psychology, I would not go so far as to say that “Cognitive science is the creation science of psychology.¹²" Traditional cognitive psychology treats mental functions as software which can be studied separately from the hardware of the brain, but as Patricia Churchland has put it, "The hardware-software distinction as applied to the brain is dualism in yet another disguise.¹³" Of course, many cognitive psychologists are not dualists, but nevertheless, it is doubtful that such methodological dualism will be helpful in the long run as, "the respective sciences of behavior and physiology will move forward most rapidly if their domains are correctly defined and analyzed.¹⁴" The black box has been opened, and there's no turning back now, but we must make sure our theories are truly scientific.

Bibliography and Notes:

¹ Science and Human Behavior, pg 167
²ibid, pg 168
³ https://www.youtube.com/watch?v=acDLluyl3tE
⁴ Science and Human Behavior, pg 28
⁵ ibid
⁶ ibid, pg 64
⁷  Skinner, B.F. (1984). The evolution of behavior. Journal of the Experimental Analysis of behavior, 41(2), 217-221.
⁸  ibid
⁹ Skinner, B. (1975). The shaping of phylogenic behavior. Acta Neurobiologiae Experimentalis, 35, 409-415.
¹⁰ ibid
 ¹¹ Skinner, B. (1977). Why I am not a cognitive psychologist. Behaviorism, 5(2), 1-10.
¹² Skinner, B. (1990). Can psychology be a science of mind?. American Psychologist, 45(11), 1206-1210.
¹³ Neurophilosophy, pg 408
¹⁴ Skinner, B. (1977). Why I am not a cognitive psychologist. Behaviorism, 5(2), 1-10