Psy 551 Learning Theory Handout on Radical Behaviorism I. Radical behaviorism as a thoroughgoing behavioral philosophy of science Radical behaviorism is the philosophy of science that underlies behavior analysis. Sometimes radical behaviorism is taken to imply an "extreme" or even "fanatical" form of behaviorism, wherein talk of only publicly observable stimulus and response relations is allowed, and direct consideration of a purported internal phenomenon is rejected because it is unobservable and cannot be agreed upon by two or more people. Some early versions of behavioral psychology did in fact adopt this perspective, but radical behaviorism does not. In particular, radical behaviorism is not extreme in the limiting or restricting sense of the word radical. Rather, a more apt synonym for radical is "thoroughgoing." Radical behaviorism argues for a thoroughgoing, comprehensive explanation of behavior at the descriptively consistent level of behavior. The variables may be inside or outside the skin, but they are all in the behavioral dimension. It does not accept as literally valid the conceptual schemes derived from everyday linguistic practices, within or across the behavior of the research scientist, subject, service provider, or client. Radical behaviorism also has a lot to say about the way other orientations approach a science of behavior and its application. In particular, radical behaviorism has a lot to say about an orientation called "mentalism," and about an orientation to doing psychology called "methodological behaviorism." Those matters are the subjects of independent discussions. Summary points 1. Critical examination of subject matter and methods of behavioral science, as well as the application of its findings 2. Theoretical, philosophical, or conceptual questions asked by radical behaviorists: (a) Why do scientists examine and explore a given subject? (b) What rate of discovery will sustain their behavior in doing so? (c) What precurrent behaviors improve their chances of success and extend the adequacy and scope of their descriptions? (d) What steps do they take in moving from protocol to general statement? (e) What aspects of behavior are significant? (f) Of what variables are changes in these aspects a function? (g) How are the relations among behavior and its controlling variables to be brought together in characterizing the organism as a system? (h) What methods are appropriate in studying such a system experimentally? (i) Under what conditions does such an analysis yield a technology of behavior and what issues arise in its application to socially significant behavior? II. Definition of Radical Behaviorism A. Basic Statement A person is first of all an organism, a member of a species and a subspecies, possessing a genetic endowment of anatomical and physiological characteristics, which are the product of the contingencies of survival to which the species has been exposed in the process of evolution. The organism becomes a person as it acquires a repertoire of behavior under the contingencies of reinforcement to which it is exposed in its lifetime. The behavior it exhibits at any moment is under the control of a current setting. It is able to acquire such a repertoire because of processes of conditioning which are also part of its genetic endowment. (Adapted from Skinner, B. F. [1974]. About behaviorism. New York: Knopf, p. 213.) B. Important Characteristics of Radical Behaviorism 1. The study of behavior is concerned with the interaction between organism and environment. Behavior is important as a subject matter in its own right, and may be dealt with at its own level. Two implications follow from this characteristic. The first implication is that behavior is not reducible to physiology. Physiology is an independent science concerned with how the parts of the body work when an organism interacts with the world. Psychology is an independent science concerned with the relation between the behaving organism and the parts of the world with which the organism interacts. The second implication is that behavior is not important as simply the basis for making inferences about mental causes. There is no mental dimension, so there can be no mental causes. Some of the mental phenomena to which traditional explanations of behavior appeal are simply fanciful explanatory fictions, endowed with just the right set of powers and forces to explain the behavior in question. Radical behaviorism argues that we need not be concerned with these sorts of phenomena. Other mental phenomena may be understood as really behavioral in nature. Their special characteristic is that they are covert or private, and not accessible to anyone else. The lack of accessibility does not mean they are actually mental, or that a special kind of analysis is required. 2. Behavior is a function of selection by consequences at the phylogenic, ontogenic, and cultural levels. At the phylogenic level, evolution selects certain innate behavioral characteristics over the lifetime of the species. At the ontogenic level, such environmental factors as contingencies of reinforcement select the operant behavior of the individual organism during its lifetime. At the cultural level, social/cultural contingencies in the environment select broader practices affecting the welfare and ultimately survival of the social group of which the individual organism is a member. At the phylogenic level, a science of behavior is complementary with behavioral genetics. At the ontogenic level, a science of behavior is concerned with respondent and operant relations. At the cultural level, a science of behavior is complementary with social or cultural anthropology. The application of selection by consequences means that psychology is actually a part of the broader field of biology. Behavior is adaptive, and the principles and concepts of biology may be usefully applied, beginning with behavior as a product of evolution. Some forms of behavior are innate or phylogenic, and develop during the natural history of the species. Other forms of behavior are ontogenic, and develop during the lifetime of the individual organism. Still other forms of behavior are also ontogenic and develop during the lifetime of the organism, but they are a function of being in a social group and would not develop otherwise. This third form contributes to the welfare and ultimately survival of the group, or more broadly, the culture. In any event, the theme of "selection by consequences," beginning with Darwinian natural selection, may be applied at three levels to understand the context in which behavior occurs. 3. Some behavior is elicited, in the sense that its probability is attributable to the presentation of an eliciting stimulus. Behavior of this nature is called respondent behavior. When a formerly neutral stimulus signals the presentation of an eliciting stimulus, and behavior changes to the formerly neutral stimulus in specified ways, the process is called respondent conditioning (also, classical conditioning, Pavlovian conditioning). To the extent a contingency is involved, the contingency is between the formerly neutral stimulus, called a CS, and the neutral stimulus, called the US. 4. Other behavior is emitted, in the sense that its probability is attributable to the response having certain consequences. Behavior of this nature is called operant behavior. When the consequences increase the probability of behavior, the consequence is called a reinforcer. When the consequence decreases the probability of behavior, the consequence is called a punisher. Punishment is an effective means of changing behavior. However, punishment is often not recommended, because it has undesirable side-effects. When a formerly neutral stimulus signals that a response will have a characteristic consequence, and the probability of behavior changes in the presence of the stimulus, the process is called operant conditioning (also, instrumental conditioning, Thorndikian/Skinnerian conditioning). The systematic relation among the antecedent circumstance, the response, and a reinforcing consequence is called a contingency of reinforcement. The contingency of reinforcement is fundamental unit of analysis for operant behavior. A contingency may be schematically depicted as follows: S D : R ==> S R + This notation suggests that a discriminative stimulus ( S D) sets the occasion ( : ) for a response (R) to produce ( ==> ) a reinforcer ( S R + ). Most forms of behavior that are of interest to us are forms of operant behavior. Thus, the analysis of those forms of behavior is concerned with an analysis of the contingencies responsible for them. 5. Verbal behavior is regarded as operant behavior, and is to be given the same analysis as any other form of operant behavior. In particular, it is not to be regarded as giving special evidence of underlying mental processes or mental phenomena. Verbal behavior is also not amenable to the early associationistic, S - R model of classical behaviorism. Rather, as operant behavior, verbal behavior is analyzed in terms of the underlying contingencies. 6. Private events are important, but they are important as events in the behavioral dimension, not in the mental dimension. These private behavioral events are the sorts of phenomena identified in everyday language by such terms as "thinking," "problem solving," "recalling," "imagining," etc. They should not be regarded as "mental" simply because they are not accessible to anyone else. The important questions are (a) how they develop, and (b) how they enter into the contingencies that influence subsequent behavior. 7. Responses and stimuli are generic in nature and are functionally defined. That is, responses and stimuli are members of classes of similar phenomena, and are not individual entities. They are defined with respect to the environmental relations in which they participate, rather than in terms of their physical properties as such. Classes of operant behavior are created by differential reinforcement with respect to classes of discriminative stimuli. An operant is not necessarily a single movement. It may in fact be a series of movements. The nature and topography of an operant is determined, which is to say selected, by the environmental contingencies. 8. Matters of truth, value, and ethics are matters of pragmatism, or successful working in everyday life. These matters are assessed in terms of contingencies. These contingencies operate across time, for individuals, for the social/cultural group to which the individual belongs, and for the species to which the individual belongs. 9. Our society/culture should actively promote practices that will enhance, and should actively discourage practices that decrease, the quality of life for its citizens. These steps should be carried out according to known behavioral principles. III. Radical behaviorism and the "Inside Story" An important feature of radical behaviorism is the way it incorporates variables that are inside the skin in some sense (e.g., as mentioned in paragraphs II-1 and II-6 above). Some of these variables are physiological, and some are behavioral. In no sense does radical behaviorism regard the organism as "empty," or does radical behaviorism deny or ignore variables inside the skin that are nevertheless relevant to an understanding of a given instance of behavior as an act in context, even though they are unobservable to someone else. In the case of physiological variables, an organism has a certain genetic endowment that determines its basic characteristics, such as its basic sensitivity to environmental stimuli and its ability to engage in motor responses. In addition, an organism's nervous system is relevant in two ways. First, the nervous system provides continuity within a behavioral event, such as between (a) contact with environmental stimulation and (b) the resulting response. Second, it provides continuity between behavioral events, such as between (a) a behavioral event at one moment in time and (b) the effects of that event measured at a later time. Behavior can be predicted and controlled on the basis of knowledge of an organism's history, but if knowledge of environmental history is lacking, knowledge of how the physiological state of the organism has been modified by interactions with the environment can be used to predict and control behavior. In addition, physiological knowledge (e.g., or mechanisms, structures, or pathways) is not necessary for the "explanation" of behavior. To so assume is reductionism, and is one of the hallmarks of methodological behaviorism. In the case of behavioral variables inside the skin, individuals sometimes behave covertly. For example, their past history may have led them to be able to "visualize" the solution to a problem. Reading aloud may have been punished, so individuals learn to read silently. They may engage in some activity that is accessible to themselves alone, as when they are passengers in a car and find themselves tensing the same leg muscles that would be involved in driving a car. These responses do not necessarily occur in all cases, but when they do, they may be viewed as links in the chain, whose function is not different from behaving overtly. Individuals also learn to describe their internal feelings and sensations through interactions with the environment. They learn to report a pain caused by a sharp object as a sharp pain, and by a dull object as a dull pain. Excruciating pains come from being crucified, as the etymology clearly suggests. In sum, variables inside the skin and unobservable to others are relevant for radical behaviorism. However, they are regarded as behavioral, rather than mental. Although they may be internal and unobservable in a certain sense, they are not from a different dimension. Thus, radical behaviorism does not mandate theories and explanations that only include observable variables.