 Dr. Barry Lowenkron is Professor of Psychology and director of the ABA-accredited MS program in applied behavior analysis at California State University Los Angeles. Although his doctoral training was in cognitive psychology with a focus on the role of language in mediating complex performances, he rather quickly became dissatisfied with the ambiguous theories that characterize cognitive psychology. He was equally dissatisfied with psychology’s dogmatic insistence on group research designs and the statistical analysis they required in order to see if an experimental effect had occurred. It seemed that this analysis moved the experimenter away from, rather than closer to, the critical characteristics of the performance under study. After several stimulating discussions with Derek Hendry and a close reading of Sidman’s Tactics of Scientific Research, along with attendance to several early meetings of ABA, Dr. Lowenkron saw the light and switched to a radical behaviorist orientation. Since that time, he has pursued his original research topic; no longer by studying the role of language in mediating complex performances, but rather, for the last 20 years, by studying the role of Skinner’s verbal operants, especially the role of joint control, in complex generalized performances. This research program has resulted in the publication of a steady stream of research and theory articles in the behavior-analytic journals. |
| Abstract: Certainly the biggest impediment to the development of a complete behavior analysis of human behavior lies in the problem of accounting for generalized responding: that is, accounting for responding appropriately to hitherto untrained stimuli. While behavior analysis can account for primary or concrete generalization, that is, generalization along a physical dimension, it has not yet developed a parsimonious and complete account of generalized responding based on abstractions; that is generalization responding based on relations between stimuli. These relations consists of two sorts. The first sort involves relations between words and objects, while the second consists of relationships between objects themselves. Thus the first sort involves questions such as how words only trained to be emitted as names (objectword responding) can, without further training, serve to select objects (word-object responding). The second sort of relation involves performances as abstract as seeking an object in response to a given description, and then, upon failing to find the specified object, modifying the description in a systematic fashion so as to select some other object bearing the same relation as the original (e.g., find the next largest object). In this paper we shall see how the simple notion of joint control serves to provide a parsimonious behavioral account of these and other seemingly conceptual performances.. |