Describing Interactions After Multi-Element Conditional Discrimination Training: Learning to Write Without Writing

Abstract

Interpreting and describing complex information are essential skills for effective functioning in many academic and occupational settings. For example, undergraduate psychology students must often interpret and describe the complex interactive effects of at least two variables on behavior. This experiment described a new procedure that enhanced the emergence of this difficult-to-learn skill: the prior training of selection-based conditional discriminations between graphs of interactions and their printed descriptions. The accuracy and completeness of written descriptions for students who received standard conditional discrimination training procedures were very low and no better than those obtained from students who were in a test–retest control group. Two other specially designed conditional discrimination conditions involved training that established explicit control by many elements in the graphs and their corresponding textual elements in the printed descriptions of the graphs. These groups showed notable improvements in their written descriptions. The behavioral mechanisms responsible for these effects are discussed. Finally, these findings might inform the development of instructional packages to reliably induce this complex communicational repertoire in a variety of content areas.