|
| International Symposium - Making the Case for Mechanicism in Behavior Analysis |
| Monday, May 28, 2007 |
| 9:00 AM–10:20 AM |
| Cunningham B |
| Area: TPC; Domain: Theory |
| Chair: N. Joseph Rodrigues (University of Nevada, Reno) |
| Discussant: Peter Killeen (Arizona State University) |
| Abstract: Behavior analysis is considered to be based on mechanicism by some and on contextualism by others. This symposium makes that case that the former is the more appropriate and useful philosphical basis for a science of behavior. The first paper argues that the methodology of mechanicism, hypothetical-deduction, is superior to that of contextualism, induction, given the goals of prediction and control. The second paper reviews the arguments made in the mechanicism/contextualism debate and suggests that behavior analysis, like other sciences, ought to be based on a mechanistic philosophy. The third paper demonstrates how a mechanistic position, namely Theoretical Behaviorism, is capable of dealing with inferred-process theories. Taken together, these papers make the case that mechanicism is an adequate, appropriate and fruitful basis for behavior analysis. |
| |
| On the Methodological Aspect of the Mechanicism/Contextualism Debate. |
| JOSE E. BURGOS (Centro de Estudios e Investigaciones en Comportamiento - University of Guadalajara) |
| Abstract: The mechanicism/contextualism debate is partly methodological. Conventional wisdom has it that mechanicists defend the hypothetico-deductive method, whereas contextualists favor the inductive method. The latter preference has an influence from Skinner’s emphasis on induction as the proper method of behavior science. This emphasis arises from assuming induction to be more effective for predicting and controlling behavior than hypothetical-deduction. However, Karl Popper pointed out that informational content and probability are inversely related. For example, the conjunction “Tomorrow will rain and the day after tomorrow will rain” is more informative than either of its conjuncts, but the conjunction is less probable than either conjunct. This result was discovered independently by Claude Shannon in his mathematical theory of communication. For example, one bit of information about an event corresponds to a probability of occurrence of that event of 1/2, while 3 bits of information correspond to a probability of 1/8. By seeking an increase in probability, induction leads to a decrease in content, which makes us more ignorant about the world. Hypothetical-deduction, in contrast, emphasizes content over probability. If content and effectiveness in prediction and control are positively related, then induction leads to less effective prediction and control than the hypothetical-deduction. |
| |
| Reviewing the Mechanicism/Contextualism Debate. |
| N. JOSEPH RODRIGUES (University of Nevada, Reno), Ramona Houmanfar (University of Nevada, Reno) |
| Abstract: It has been argued that behavior analysis is unique in that it is based on contextualism. According to these arguments mechanicism is outdated and not sophisticated enough to adequately deal with psychological phenomena. We re-examine this claim and show that a mechanistic philosophy is not only adequate but also more appropriate to the task of explaining behavior. Misunderstandings about the nature of mechanicism are clarified and the success of this position in other sciences as well as in interdisciplinary collaboration are pointed out. The majority of arguments against mechanicism are made based on Stephen Pepper's interpretation of mechanism and are targeted against discrete mechanism which he acknowledged in 1942 was already being replaced by consolidated mechanism. When these criticisms are examined in light of consolidated mechanism they lose much of their strength. Mechanicism is the only worldview that continues to evolve, thus lending itself to a better fit with the evolving nature of science in general and behavior analysis more specifically. |
| |
| Theoretical Behaviorism: Cognitive Psychology as a Behaviorist Views It. |
| JEREMIE JOZEFOWIEZ (Duke University) |
| Abstract: One of the characteristic of radical behaviorism compared to cognitivism has been its opposition to inferred-process theories. The debate on this point has been reopened by Staddon's new theoretical behaviorism which argues that inferred-process theories are necessary for the understanding of behavior. This talk discusses why inferred-process theories can indeed help achieve this goal and in which conditions the process described by such a theory are not mere fictions but might correspond to processes actually taking place inside the organism. It also tries to show that the use of inferred-process theories does not equate theoretical behaviorism with cognitivism because of radical differences between the research agenda of a behaviorist versus a mentalist psychology. |
|
| |