Psy 551 Learning Theory Handout on Methodological Behaviorism I. Definition of methodological behaviorism Methodological behaviorism is a generic term that refers to a variety of positions in psychology that embrace certain assumptions. It is not simply one single form of behaviorism associated with one single theorist. In many cases, it refers to positions associated with virtually all those who are called behaviorists except B. F. Skinner. In addition, it refers to a great many other positions, even though they are not called behaviorist. Methodological behaviorism has evolved over the years. As a result, there are several variations, and the variations differ from each other. The principal assumption of methodological behaviorism is that science should only include phenomena that are publicly observable in its theories and explanations. In other words, radical behaviorists argue that a position is a methodological behaviorist position when it doesn't directly appeal to anything that isn't publicly observable in its theories and explanations. The operative word in this assumption is "directly." One variation is to deny the existence of anything that is not publicly observable. The rationale is that if it doesn't exist, it shouldn't be directly included in psychological science. This variation obviously doesn't violate the definition of methodological behaviorism because it doesn't directly appeal in a theory or explanation to anything that isn't publicly observable. Another variation is to ignore the existence of anything that is not publicly observable. This variation actually grants the existence of things that are not publicly observable, including phenomena from the mental dimension, but then rules them out of bounds for direct consideration in psychological science. Even though this variation believes that mental phenomena do exist and that they do cause behavior, it believes the mental phenomena still shouldn't be directly included in psychological science because they can't be measured and agreed upon by two or more observers. If a scholar wants to consider them, the scholar would have to pursue some other mode of analysis, such as rational inquiry or philosophical analysis. This variation doesn't violate the definition of methodological behaviorism because it doesn't directly appeal in a theory or explanation to anything that isn't publicly observable, although for a different reason than the first variation. A third variation also grants the existence of things that are not publicly observable, including phenomena from the mental dimension, but it then tries to incorporate them indirectly, rather than directly. They could be incorporated indirectly as mediating "theoretical terms" or "logical constructs," which are then "operationally defined" with respect to publicly observable variables. Although the mental phenomena are assumed to exist and cause behavior, again this variation does not violate the definition of methodological behaviorism because it doesn't directly appeal in a theory or explanation to anything that isn't publicly observable. Rather, the appeal is indirect, through theoretical terms and logical constructs. The only direct appeal is to the theoretical terms and logical constructs that are operationally defined with respect to publicly observable variables. II. Historical background Attempts to create a science of psychology in the late 1800s and early 1900s were based on a fundamentally mentalistic outlook. The versions of psychology that became prominent (e.g., structuralism, functionalism) were based on the introspective analysis of the contents of consciousness ("mental life"). These attempts didn't work, if only because of the lack of agreement about the findings and theories. What seemed to be lacking were an appropriately defined subject matter and method, so that psychology could be "objective" and "empirical." That was when behaviorism became prominent, as a fresh start. Watson launched classical behaviorism around 1913. He restricted his science to stimuli and responses. Accordingly, his version of psychology classical behaviorism is also known as S - R psychology. He believed that the mental was not part of a science. At issue is why he took this position. Early on, his writing seems somewhat consistent with the second variation above, but later his writing seems more consistent with the first. By the 1930s, psychologists saw that Watson's S - R classical behaviorism wasn't really providing an adequate alternative to the study of the contents of consciousness through introspection. There was variability: Stimuli and responses weren't always correlated with each other in the one-for-one way required by classical behaviorism. In addition, the S - R model does not easily accommodate how we come to use subjective terms to describe various conditions inside our bodies. Finally, weren't other sciences making progress by hypothesizing unobservable entities? Don't unobservable entities contribute in some way (e.g., synthesizing, parsimony)? Aren't we unnecessarily limiting ourselves? Consequently, researchers began to amend Watson's S - R formulation by inserting intervening, organismic variables between stimulus and response, thereby creating an S - O - R formulation. They believed that this move would allow them to deal with the inadequacies of classical S - R behaviorism. This version may be designated mediational neobehaviorism, to distinguish it from Watson's classical S - R behaviorism. The question was how to remain scientifically respectable in the process. How was the scientific validity of the mediating organismic variables to be established? Although the entire story is quite complicated, suffice it to say at this point that the principles of operationism (as in "operational definitions"), which was developing in physics, and "logical positivism," which was developing in philosophy of science, seemed to provide the required respectability. The organismic variables were given the status of "theoretical terms" in the S - O - R model of mediational neobehaviorism, and then "operationally defined" with respect to the publicly observable variables entailed in their measurement. Theoretical terms were to be distinguished from observational terms, such as stimuli and responses. Often, however, the theoretical terms were nothing but surrogates for mental terms, but because of the restriction against direct appeal to mental terms, theorists had developed a way to circumvent the restriction. The mediational approach allowed theorists to conceive of the mental terms as theoretical, and allowed theorists to include mental terms indirectly. This approach represents the third variation described above. Radical behaviorists argue that methodological behaviorism, particularly the third variation, is the modal, "orthodox" orientation in virtually all of contemporary psychology. It is the basis for teaching, curricular design, research, and professional socialization in the discipline of psychology. In more modern language, it may be described as embracing to the following assumptions. Explanation is assumed to involve the construction of a theory. A theory in turn is assumed to involve an appeal to inferred unobservable phenomena that are presumed to underlie behavior. If a causal explanation of behavior is sought, causal explanation consists in the theoretical appeal to these inferred, "underlying" phenomena. These phenomena are unobserved, and perhaps unobservable, even in principle, by anyone. The most common examples of these inferred, underlying phenomena are acts, states, mechanisms, or processes, which are presumed to belong to a dimension that differs from the dimension in which the behavior takes place, such as the cognitive dimension (cf. mentalism). The inferred, unobserved phenomena in the theory are to be regarded as theoretical terms (or equivalently, logical constructs, theoretical constructs, inferred constructs, logical terms, etc.), and are to be operationally defined with respect to publicly observable phenomena. Data produced through experimentation according to conventional methodologies (control groups, statistical inference, etc.) with publicly observable variables are presumed to provide the only valid grounds or justification for speaking of the inferred phenomena. Even though the theorists may believe the inferred phenomena actually are mental and actually do exist, no commitment is technically necessary or is made one way or the other as to their ontology or actual existence. In this way, methodological behaviorists thought their scientific efforts and their theories would be meaningful because they were linked to the world of physical events. As this scene played out over the years, methodological behaviorists further discussed whether the theoretical terms were to be given an intervening variable interpretation, which does not permit surplus meaning, or a hypothetical construct interpretation, which does permit surplus meaning, perhaps even including the "mental." After much debate, they came to favor the latter. Again, given the latter interpretation, legitimate question may be raised as to whether methodological behaviorism really should be called a behaviorism after all, or whether it is just a disguised version of mentalism. Note that defining a cognitive construct in terms of some physiological structure or mechanism doesn't really solve the problem either, as the cause is still from another dimension. In other words, all the theorist has done is to hypothesize about the machinery that the "ghost in the machine" operates. III. Radical Behaviorist Concerns About Mentalism and Methodological Behaviorism Radical behaviorists reject methodological behaviorism, just as they reject mentalism, on pragmatic grounds. Radical behaviorists argue that a critical examination of mentalism and methodological behaviorism reveals they are based on an entire series of incorrect assumptions about the nature of verbal behavior, the role of verbal behavior in producing knowledge, and the role of theories in knowledge and explanation. These incorrect assumptions ultimately lead people to search for the wrong causes of behavior, and to accept incorrect answers to questions about the causes of behavior. That is, radical behaviorists argue that mentalism and methodological behaviorism obscure important details, they allay curiosity by getting us to accept fictitious way stations as explanatory, they impede the search for genuinely relevant variables, they misrepresent the facts to be accounted for, and they give us false assurances about the state of our knowledge. Moreover, they lead to the continued use of scientific techniques that should be abandoned, for example, because they are wasteful. Consequently, they interfere with effective prediction, control, and explanation of behavior. Note that question has already been raised as to whether these mediating theoretical terms were fundamentally mental in nature, and that methodological behaviorism was at heart a subterfuge to try to deal with mental events "scientifically." If so, then question may legitimately be raised as to whether methodological behaviorism really differs from mentalism. In principle, the mediational neobehaviorists almost always believed the theoretical terms did refer to causal mental entities, but methodological behaviorists remained silent about them and their mental nature because they believed scientists should only be talking about publicly observable variables. Thus, methodological behaviorists could indirectly incorporate mental entities in their theories, but by viewing them as theoretical terms that were operationally defined the methodological behaviorists were able to escape charges that they were returning to the days of introspection. An important point is that radical behaviorist concerns about mentalism and methodological behaviorism do not turn simply on the extent to which an approach is deemed "theoretical." Theories are verbal behavior. They are occasioned by certain antecedent conditions, and reinforced by certain other conditions. They are important as a form of discriminative stimulation that guides future action through either (a) direct manipulation of environmental events or (b) action when direct manipulation is not feasible. Radical behaviorism is not opposed to theories in principle. However, it does reject the sorts of mentalistic theories that appeal to events and entities that are somewhere else, at some other level of observation, in a different dimension ( "mental," subjective, cognitive) which must be described in different terms. It further rejects the viewpoint that causal explanation in psychology, and psychological knowledge in general, consists in framing such theories. Indeed, radical behaviorism argues that the belief such theories are necessary to psychological knowledge is a further illustration of the same mentalistic problem. Instead, radical behaviorism advocates theories and causal explanations that entail a description of the functional relation, or the contingency, between behavior and its controlling variables. Such a statement would describe uniformities in those contingencies across many different circumstances, using a minimum number of terms. The statement would be descriptively consistent, and stay within the same dimension. For example, radical behaviorism does not use "theoretical terms" (as opposed to observational terms) in its causal explanations. In addition, radical behaviorism may include private events in its explanations. Such events are accessible only to one person, and as suggested above, are identified in everyday language by such terms as "thinking," "recalling," etc. For radical behaviorists, these events are behavioral. They are not mental, and they are not given the status of inferred phenomena from another dimension. They are not necessarily part of every behavioral event, but may become so, for example by exerting discriminative control, given the right behavioral history of the organism. Thus, the subject matter, independent variables, and dependent variables for radical behaviorism are not restricted to publicly observable phenomena. Mentalists and methodological behaviorists often justify their position by claiming that science has always progressed by hypothesizing "unobservables" in theories and explanations, and that it is not appropriate to limit scientific theorizing. Radical behaviorists argue that the chief virtue of this claim is that of being precisely dead wrong. Radical behaviorists emphasize that to the contrary, science has always progressed by dispensing with fanciful explanatory fictions. Three conspicuous examples are (a) phlogiston in chemistry, (b) vital spirits in biology, and (c) the geocentric universe. At issue is the nature of the unobservables. Science progresses by ridding itself of unobservables of uncertain ontology. There is nothing limiting about this process. For radical behaviorism, mentalistic constructs are examples of things of an uncertain ontology. As such, they are irrelevant at best, and a counterproductive diversion at other times.