Reconciling the Mental and the Behavioral: An Evaluation of Behaviorism
Alissa D. Eischens Northwestern University
This paper examines Skinner's theory of radical behaviorism. The examination describes the concepts of classical conditioning and operant conditioning as well as evaluates Skinner's (1984) "Selection by Consequences." The theory of radical behaviorism is then evaluated for its greatness. Despite criticisms, the theory is shown to have merit in some respects. However, the theory is shown to have fundamental flaws that inhibit its greatness, such as the use of animal research, the "black box" concept of the mind, and the denial of thought and mental processes.
The concept of conditioning is well-established in psychological theory and practice. From Pavlov's research, the world became cognizant of classical conditioning. Through his study of the salivation habits of dogs, the concepts of stimuli and responses have been applied to psychological study. Unconditioned stimuli elicit unconditioned responses; for example, a dog salivates when presented with food. Pavlov then determined that stimuli could be conditioned to elicit conditioned responses. In the case of his dogs, he paired a bell with the presentation of food, and after time the bell itself produced salivation. Finally, higher-order conditioning is possible. This type of conditioning occurs when a conditioned stimulus is able to cause responses from other neutral stimuli by being associated with them. For example, pairing a ball with a bell can cause a dog to salivate in the presence of a ball. In humans, higher-order conditioning can be quite complex; symbols such as words can be capable of evoking emotional responses (Mischel, 1993).
Harvard psychologist B. F. Skinner proposed his own theory of conditioning. His concept, operant conditioning, is at the heart of his highly influential and controversial theory of behaviorism. Behaviorism studies behavior as the basic unit of understanding organisms, including humans. The theory observes behavior and seeks to determine the conditions that affect a given behavior. Essentially, in personality theory, according to behaviorists a person's behavior determines his or her personality.
Operant Conditioning Behaviorists, being concerned with observable behavior, treat the mind almost as the proverbial "black box." Input enters and output exits, but the processes that relate the input to the output are not examined. The "black box" idea is convenient for behaviorism because mental processes are unobservable and therefore very difficult to explain according to behavioral theory. In fact, behaviorists tend to refuse the idea that specific motivations even exist for behavior. They instead try to determine the external conditions that influence behavior and explain away motivations or drives as simply the effects of deprivations or satiations. Instead of motivation, environment is the key factor influencing behavior. Skinner's basic strategy for studying behavior involves functional analysis; the link between behavior and exact determining conditions is sought. He maintains that most of human behavior is the result of freely-given response patterns, also called operants.
In simplest terms, Skinner's operant conditioning is learning based on the consequences produced by the responses of elicited behavior. In this way, behavior can be said to be reinforced, and nearly all events can act as reinforcers. Therefore, as in the example of words being stimuli in higher-order conditioning, behaviorism would say that words that elicit positive emotional responses reinforce the speaker to utter them again in order to receive the positive response. Skinner also realized that conditioned reinforcers could be generalized when a response pattern occurred under many conditions.
Through his observation of animal behavior, primarily that of pigeons, Skinner also developed the idea of shaping to obtain a desired behavior. Shaping is the successive rewarding of closer approximations to a desired behavior. It involves schedules of reinforcement that can be continuous or partial. Partial reward of behavior tends to be more effective. Shaping can be applied to humans as well as pigeons. A common use of shaping occurs in the rearing of children as they need to be taught desirable behaviors (Mischel, 1993).
The theory of evolution holds that characteristics of organisms come about due to selection by consequences. Skinner includes behavior as a characteristic that can come about through selection by consequences. He says, "What we call behavior evolved as a set of functions furthering the interchange between organism and environment" (Skinner, 1984, p. 477). Therefore, Skinner concludes that behavior also evolves in accordance with two contingencies: the ability of an organism to be reinforced by consequences and the availability of behavior that releases stimuli.
Skinner's "Selection by Consequences" Central to Skinner's theory is the idea of consequences as related to the evolution of behavior. The first consequence, reproduction, led to the evolution of cells through natural selection and the ability of organisms to reproduce over diverse environments. The second, operant conditioning, can act in place of natural selection by influencing and shaping behavior. The third consequence, verbal behavior, occurs as humans become more social.
Skinner also designates three levels of behavioral evolution. The first level, Level i, is governed by biology. Natural selection can be seen as occurring on the genetic level, and is concerned primarily with survival of genes in an individual organism. The second level, Level ii, is governed by psychology. Level ii is very much concerned with the behavior of individual organisms. Finally, the third level, or Level iii, is governed by anthropology and is concerned with the evolution of societal behavior.
In Level i, the environment does not designate the traits that will survive; rather the species adapts to the environment, choosing traits that will optimize survival. In Level ii, the individual adjusts to the environment rather than allowing the situation to dictate behavior. Finally, in Level iii, groups solve problems as opposed to allowing circumstances to determine traits that would most benefit the group. As a final thought, Skinner asserts that individual or free will is not responsible for behavior; rather, selection by consequences is the determinant (Skinner, 1984, p. 480).
As a theory of personality, behaviorism and Skinner's idea of selection by consequences is not without merit. According to the zoologist Dawkins (1984), selection by consequences "puts a correct emphasis on the radical difference between active selection by a choosing agent on the one hand, and the blind...mechanical purposelessness of the Darwinian and quasi-Darwinian processes that Skinner lists on the other" (p. 486). Plotkin and Odling-Smee see the merit of selection as the "basis by which living systems gain knowledge of themselves and their world. Selection operates at the genetic, developmental, individual learning, and cultural levels" (Plotkin and Odling-Smee, 1984, 493). The anthropologist Harris, most concerned with Skinner's Level iii, says, "Anthropologists would benefit on behalf of selection by consequences as a metaprinciple for explaining cultural as well as biological evolution and the acquisition of individual response repertories" (Harris, 1984, p. 490). Furthermore, Skinner's theory encompasses all of psychology and provides a means to examine systematically the relations between a subject and its environment. His ideas concerning behavior as a function of its consequences (or reinforcement and punishment) are also compelling (Jensen & Burgess, 1997, p. 221).
Evaluation of Behaviorist Theory Behaviorism is, however, fundamentally flawed in a number of ways that prevent its being considered a great theory of personality. Its assumptions about human beings from observations of animals are largely questionable. In addition, the idea of the "black box" of the mind is not an acceptable explanation of mental processes.
Behavioral scientists tend to accept the idea that information provided by research with animals such as rats, monkeys, and pigeons can be used to make inferences about human behavior. However, the transferability of the knowledge gleaned from animal research is highly questionable (Cangemi & Kowalski, 1993, p. 493). Animal evidence is so problematic because humans often behave in unexpected ways, largely because they have verbal skills. Humans do not always behave in the same manner as animals because verbal input from other humans or themselves can affect their behavior (Overskeid, 1995, p. 518). Says K. E. Boulding (1984) of Swarthmore College's department of economics, "It is on the whole an inappropriate methodology in developing improved cognitive images of complex, unstable systems with changing parameters and cumulative structures, where rare events are significant. Humans are a supreme example of systems of this kind" (p. 483).
Boulding (1984) also feels that the "black box" concept of the mind is very limited in its value for understanding behavior. He cites the examples of the capacity for reflection and communication as keys to opening the "black box." Therefore, such blatant dismissals of cognitive function as Skinner's statement, "We need not suppose that events which take place within an organism's skin have special properties....A private event may be distinguished by its limited accessibility but not, so far as we know, by any special nature or structure" (Jensen & Burgess, 1997, p. 222) have little merit.
In his book Brainstorms, D. C. Dennett (1978) provides a compelling argument against behaviorism because of its denial of the mind or what Dennett terms "mentalism." Opposed to Skinner's essential claim, "behavioral science proves that people are not free, dignified, morally responsible agents" (Dennett, 1978, p. 54), Dennet illustrates behaviorism's shortcomings in explaining away mental processes as simple behaviors.
Dennett and Mentalism Skinner is vehemently opposed to the idea of mentalism. He cites several reasons for his objection including the opinions that mental processes are non-physical and therefore not directly observable, that the mental world is private, and that the analysis of mental (internal) events requires inference. Skinner inherently sees the idea of thought as negative; he is opposed to intentional idioms (verbal expressions that intimate emotion or hidden cognitive processes) such as "feel" or "believe." He also holds that intelligence or rationality cannot be presupposed; however, Dennett asserts that any type of thought must presuppose rationality. He believes that Skinner is wrong in suggesting that psychology cannot refer to intentional idioms. Skinner holds that beliefs, for example, are built as the probability of an action increases. He also shows that Skinner uses intentional idioms himself.
Dennett (1978) defines the essential mental-behavioral question as, Can intentional and scientific explanations co-exist? Whereas Skinner does not see the distinction between the two, Dennett disagrees, finding fault with behaviorism for explaining things (specifically mental processes) away rather than explaining how things work. Finally, Dennett asserts that Skinner simply fails to show that psychology without "mentalism" is possible or actual (Dennett, 1978, pp. 54-70).
B. F. Skinner's theory of radical behaviorism has made valuable contributions to the field of psychology, if for no other reason than to spark debate. His operant conditioning ideas are not wholly invalid; however, his denial of the importance of mental processes as well as the use of animal data and the "black box" idea hinder his theory from achieving greatness.
Conclusion
Peer Commentary
The Deeper and Necessary Argument Behind the Black Box Nathan C. Popkins Northwestern University Eischens reviews the theory of behaviorism very objectively and sensibly. When weighing the pros and cons of the theory, her eventual conclusion is well founded; however, the means through which she arrives at this conclusion may not be. Specifically, the idea of the "black box" is lightly skimmed over, when it seems that, by weight of the pros and cons of different aspects of this theory, a major part of Eischens' examination and evaluation of behaviorism should be based on the controversy surrounding the "black box."
First of all, Eischens' summary of the merits of behaviorism is well stated. Certainly, the levels of behavior and the predictability to which they lend themselves is behaviorism's most concrete asset. Also, she is correct in listing the extrapolation of animal behavior onto humans and the problems associated with the "black box" as the factors that keep this theory from greatness. The latter of these two factors, however, was barely touched on in Eischens' paper, when in fact the "black box" is among the most important aspects of behaviorism.
The "black box" of behaviorism is the most frequent target of criticism by non-behaviorists, and rightfully so. This aspect of behaviorism robs this theory of any hope of establishing a causal relationship within the mind (of which psychologists are so fond), and instead leaves behaviorism as nothing more than a functional input-output model.
The next possibility that needs to be examined (and that was not covered in Eischens' paper) is the question of whether or not this mostly functional input-output model is good enough, or whether it really is necessary to establish the aforementioned causation behind the model as well. To do a sufficient job of this, the author would need to cite precedents in psychology, for either argument. These precedents are not hard to find. Any "great" psychological theory, from psychoanalysis to attachment theory, will exhibit the trend in psychology to not simply look for a convenient descriptor of actions, and the reasons for them. These theories also require a fundamental explanation of why the inputs cause the outputs to which they are attributed.
For example, if behaviorism is to attain the status of "greatness," then in the famous case of the traumatized youngster who witnessed the horse fall on street, behaviorism would do more than simply state that this traumatic event was the cause of the boy's aversion to horses. Behaviorism would explain mentally what transformed the act of witnessing this event into a phobia. This is the standard to which other great theories in personality have been held, and behaviorism must also conform to this precedent to be considered a great idea in personality.
Perhaps, if this were an economic or sociological theory, then the current input-output nature of this theory would be good enough. However, psychologists do not hope for, but rather require, another level of explanation in their theories. It is because of this that the "black box" aspect keeps behaviorism from achieving greatness as a theory.
Peer Commentary
Taking a Closer Look at Behaviorism Raymond Wang Northwestern University Eischens has written a good review of Skinner's theories of behaviorism. She has mentioned many of the aspects and implications of behaviorism, as well as many different opinions on the validity of the theory. Overall, she has done well in covering a wide range of topics related to behaviorism in a short amount of space. I only hope to examine some of the ideas a little more closely.
Skinner was not concerned with the mental processes of the individual, he was concerned with how the individual acts and what conditions make him or her act that way. Skinner rejected the validity of mental processes solely because these processes are unobservable. The processes can only be determined by asking the individual, and doing so is too subjective for scientific study. Behavior, on the other hand, is observable for study. Hence, Skinner believed that behavior should be the focus of study.
Applied to personality theory, behaviorists would object to Ralph Waldo Emerson's statement, "Man is what he thinks." They would, perhaps, opt for a slightly different rendition, "Man is what he does," or perhaps even, "Man is how he reacts."
Eischens states, "operant conditioning is learning based on the consequences produced by the responses of elicited behavior." Simply put, environment affects our behavior. If what we do produces a desired consequence, we do it again. If what we do produces undesired consequences, we try not to do it again. We learn by "operating" on the environment (Carlson & Buskist, 1997). Eischens used the example of words to illustrate this point. As she states, "words that elicit positive emotional responses reinforce the speaker to utter them again in order to receive the positive response." I find this interesting because there have been several studies on the use of curse words, the emotional release caused by their utterances and the subsequent overuse of them.
What follows is Skinner's theory of "shaping." Eischens develops this idea fairly well and mentions the example of child-rearing. She did not mention, however, that shaping is used by just about everybody in acquiring skills. For instance, if we find by moving our fingers a certain way that the sound of the musical instrument becomes more desirable, then we will be more likely to move our fingers the same way again. We will gradually fine tune (no pun intended) the movement of our fingers until we have achieved the most desirable sounds we can produce. The scope of Skinner's theory is one of the reasons for its significance.
Eischens' examinations of the levels of consequences and the levels of behavior seemed rather brief. I would appreciate it if she would develop these ideas more thoroughly.
Eischens does well in examining the merits of behaviorism. She includes various quotes of all sorts dealing with biology and anthropology among many other topics. Indeed, the significance of Skinner's work has not been neglected.
However, the arguments against behaviorism seem weak. One of her arguments against behaviorism is that humans behave in largely unpredictable ways. Gavin de Becker (1997), a three time presidential appointee and author of the book The Gift of Fear, would disagree. His book focuses on predicting the violent behavior of people around us. Applied to behaviorism, he examines the behavior of individuals to determine whether or not those individuals will resort to violence. His argument is that not only are accurate behavioral predictions possible, but also everyone is capable of making them (de Becker 1997).
Next, Eischens cites Boulding (1984) and Dennett (1978) and their arguments against behaviorism. The problem is that she cites statements that are directly refuted by Skinner's ideas. She states that reflection and communication are key to unlocking the cognitive processes, whereas Skinner has already stated that these methods are too subjective. I would prefer to see more concrete examples and arguments that demonstrate the limitations of behaviorism. Eischens has not explored the recent attempts of cognitive scientists to map out the internal workings of the mind. Cognitive science aims to determine the workings of information processing, memory storage and retrieval, language processing, and many other aspects of thought. These are all internal functions that would refute Skinner's denial of mentalism.
Eischens has obviously researched her material very well. Her efforts are reflected in the scope of her work. In general, however, she may improve her paper by examining some of the topics more thoroughly.
Author Response
Behaviorism and the Mind: No Easy Answers Alissa D. Eischens Northwestern University To begin my response to the commentaries made by Popkins and Wang concerning my paper, let me say that my examination of behaviorism was indeed cursory. I did not presume to refute the theory in the short space allotted to me; I merely hoped to point out some of the shortcomings of behaviorism as a theory of personality. Both Popkins and Wang recognize the brevity of my argument, and if I truly wished to refute behavioral theory, I fully understand that further explication of my arguments would be required.
That said, in "The Deeper and Necessary Argument Behind the Black Box," Popkins makes a good point about the importance of the black box as an indicator of the shortcomings of behaviorism. I agree that recognizing the black box is integral in the criticism of behavioral theory. However, I feel that Popkins seems to be trying to make behaviorism into something that it is not. He reasons that behaviorism should attempt to explain the inner workings of the mind as do other psychological theories, and cites my failure to examine this point. However, the very premise of behaviorism relies on the denial of the inner workings of the mind; therefore, I fail to see how Popkins can say that behaviorism can only achieve greatness by doing so. Behaviorism tries to achieve greatness in ways different from most other theories of personality; it is the proverbial "black sheep" in this respect. As Popkins and I both recognize, behaviorism falls short of its goal.
In "Taking a Closer Look at Behaviorism," Wang also points out behaviorism's problems with the inner workings of the mind. I thank him for his introduction of cognitive science as a possibility for expansion upon my argument. However, I feel that findings made by cognitive research must be used carefully in the argument against behaviorism; they have the potential to be used in behaviorism's favor. He cites my examples of reflection and communication and finds fault in their subjectivity. But if the mind one day can truly be mapped and neural transmission is fully understood, could not reflection (or "thought") and communication be seen purely as physical processes? I would propose that behaviorists would jump at the chance to reduce the mind to physical functions, much like respiration or digestion. I think that a purely electrochemical mind would strengthen the case of behaviorism. Therefore, any examination of cognitive science would have to address the question with which it wrestles, namely the explanation of consciousness, which I feel could be an example of "mentalism."
Wang also finds fault with my citation of the unpredictability of human behavior. Suffice to say, I have not read de Becker's book and thus do not feel qualified to comment in that area.
I thank both Popkins and Wang for their honest commentaries concerning my paper. Should I further expound upon my argument, I would surely take their suggestions into account.
Boulding, K. E. (1984). B. F. Skinner: A dissident view. Behavioral and Brain Sciences, 7, 483-484.
References Cangemi, J. P., & Kowalski, C. J. (1993). Does a hierarchy of significance exist in psychology and the mental health disciplines? Education, 113, 489-497.
Carlson, N. R., & Buskist, W. (1997). Psychology: The science of behavior (5th ed.). Boston: Allyn and Bacon.
Dawkins, R. (1984) Replicators, consequences, and displacement activities. Behavioral and Brain Sciences, 7, 486-487.
De Becker, G. (1997). The gift of fear: Survival signals that protect us from violence. Boston: Little, Brown, and Company.
Dennett, D. C. (1978). Skinner skinned. In D. C. Dennett, Brainstorms: Philosophical essays on mind and psychology (pp. 53-70). Montgomery, VT: Bradford Books.
Harris, M. (1984). Group and individual effects in selection. Behavioral and Brain Sciences, 7, 490-491.
Jensen, R., & Burgess, H. (1997). Mythmaking: How introductory psychology texts present B. F. Skinner's analysis of cognition. The Psychological Record, 47, 221-232.
Mischel, W. (1993). Behavioral conceptions. In W. Mischel, Introduction to personality (pp. 295-316). New York: Harcourt Brace.
Overskeid, G. (1995). Cognitivist or behaviourist--who can tell the difference? The case of implicit and explicit knowledge. British Journal of Psychology, 86, 517-523.
Plotkin, H. C., & Odling-Smee, F. J. (1984). Linear and circular causal sequences. Behavioral and Brain Sciences, 7, 493-494.
Skinner, B. F. (1984). Selection by consequences. Behavioral and Brain Sciences, 7, 477-510.
Last modified August 1998
Visited times since July 2001
Comments?
Home to Personality Papers
Home to Great Ideas in Personality