A METHODOLOGICAL ANALYSIS OF SHIFT PARADIGMS IN HUMAN DISCRIMINATION LEARNING

EVALUATES THE DEVELOPMENT OF SHIFT PARADIGMS FOR TESTING MEDIATION THEORY VS. SINGLE-STAGE THEORY. THE BASIC DESIGN, THE OPTIONAL SHIFT, AND THE PARTIAL CHANGE DESIGN ARE ALL SHOWN TO HARBOR POSSIBLE BIASES WHICH COULD CONFOUND THE RESULTS. THESE SOURCES OF BIASES ARE INTERMITTENT REINFORCEMENT, SHIFT DETECTION, OBVIOUSNESS OF SOLUTION, STIMULUS NOVELTY, AND NEGATIVE TRANSFER. IT IS SUGGESTED THAT THE STRAIGHT REVERSAL SHIFT BE SUPPLANTED IN FAVOR OF ANOTHER TYPE OF INTRADIMENSIONAL SHIFT. THE TOTAL CHANGE DESIGN, WHICH USES NEW VALUES ON ALL DIMENSIONS FOR ALL GROUPS, IS PREFERABLE, PARTICULARLY IF SYMBOLIC STIMULI ARE EMPLOYED. (48 REF.) (PsycINFO Database Record (c) 2006 APA, all rights reserved). © 1968 American Psychological Association.