|
Behaviorism at 100: Where Are We Now? |
Sunday, May 26, 2013 |
3:00 PM–4:20 PM |
101 B-C (Convention Center) |
Area: TPC/EAB; Domain: Theory |
Chair: James S. MacDonall (Fordham University) |
Discussant: Peter Killeen (Arizona State University) |
Abstract: 100 year ago J. B. Watson published 'Psychology as the Behaviorist Views it" and fifty years ago Skinner published "Behaviorism as Fifty." The presenters in this symposium will review the current status of research areas what were limited or non-existent 50 or 100 years ago as well as discuss the future of these areas. Topics will include behavior controlled by classes of physically unrelated stimuli that function equivalently, quantitative models of behavior and resurgence. |
Keyword(s): equivalence, mathematical models, resurgence, Skinner |
|
Resurgence at 62 |
KENNON ANDY LATTAL (West Virginia University) |
Abstract: The first systematic study of resurgence was in 1951. Other did not follow until the 1970s, after which the number has increased exponentially. Far from being an isolated, esoteric by-product of the nonreinforcement of other behavior, resurgence has been suggested to have relevance for the analysis of behavioral history, a host of recurrence problems like drug-use relapse, and the post-treatment reappearance of treated behavior problems. Experimentally, the preparation involves reinforcing first one response, then discontinuing reinforcement for the former while reinforcing an alternative response. This in turn is followed by extinction of the alternative response, during which the reappearance of the nonreinforced first response constitutes resurgence of that response. Variables in all three phases of the procedure can affect the outcome. These variables include the type and extent of training in the first phase, the relation of behavior in the second phase to that in the first, and the conditions of response elimination/reduction in the third phase. Resurgence effects are transient ones, appearing soon after the extinction of responding established in the second phase appears, peaking in one or two sessions, and then declining to near-zero levels in the absence of further contingencies to shore up the dissipating responding. Future directions for resurgence include a search for a method for obtaining repeated resurgence, which could facilitate further isolation of its controlling variables. |
|
Generative Models of Behavior: A Proposal for a Behaviorism of the 21st Century |
FEDERICO SANABRIA (Arizona State University) |
Abstract: Although Watson's view on psychology has prevailed, its implementation during the 20th century as a science of behavior—or, more precisely, as a science of performance—has curbed empirical research and technological innovation. The main limitation stems from the unsuccessful attempt to reduce mental constructs that psychologists and consumers of psychology care about (knowledge, predilections, memories, affect) to attributes of behavioral performance. A behaviorism of the 21st century must, therefore, reincorporate mental constructs within the scope of psychology without compromising its focus on reproducible data. To attain this goal, it is proposed that mental constructs are identified with hidden parameters, states, and processes defined with quantitative precision in generative models. These models specify random processes that generate instances of performance, thus providing the basis for reverse engineering the principles that govern behavior. This presentation outlines the rationale behind the proposed approach and the methods involved in the estimation of model parameters. The characterization of mental constructs in terms of generative models would afford common ground between behaviorists that study the actions of individual organisms, neuroscientists seeking to unlock the mechanisms that make mental life possible, and social scientists dealing with interlocking minds. |
|
Concepts and Categories: Emergent Performances and Behavior Analysis |
KENNETH F. REEVE (Caldwell College) |
Abstract: In 1950, Keller and Schoenfeld noted that the formation of classes that consisted of stimuli that were related to each other but do not resemble each other was a very important area of study, but at that time had not received much attention. Addressing that topic from a behavior analytic perspective began in 1971 with the publication of the first experiment on equivalence class formation conducted by Sidman. Since then, the study of equivalence classes has burgeoned and has given behavior analysis a model with which to address the development of many complex human behavior such a syntax, semantics, and meaning. It has also given rise to research results that permit making substantial contact with phenomona that are typically said to reside in the domain of cognition. Some of these issues will be elaborated and consideration will be given to future directions. |
|
|