The Equivalence in Equivalence-Based Instruction
Equivalence-based instruction (EBI) is an approach to instructional design that is based on the laboratory model used in research on stimulus equivalence (SE). Its defining feature is that a minimal subset of all possible relations among the instructional stimuli are selected for teaching, and the remaining relations left to emerge without instruction as verified through testing. The major premise of EBI is that it saves effort relative to the alternative of teaching all target relations directly; however, this premise has been minimally subjected to empirical testing. In addition, a gap exists between basic research on SE and applied EBI research, in that the former has not examined potential differences between equivalence classes and stimulus classes in which all relations have been acquired through direct experience. I will describe a series of laboratory experiments that aimed to begin bridging this gap. In the context of establishing arbitrary stimulus classes, each experiment compared the efficiency of one or more EBI arrangements with a comprehensive instruction (CI) control condition in which all relations within target stimulus classes were taught directly. Additionally, we conducted tests for class expansion, transfer of function, class reorganization, and longer-term retention, in order to document potential effects of training history on the properties of the resulting stimulus classes. Overall, the results suggest EBI is more efficient than CI under some circumstances, but relative efficiency depends on EBI training structure and the relative stringency of mastery criteria. Both EBI and CI appear to produce equivalence classes with similar properties, but relative class flexibility may be influenced by training structure. Discussion will focus on claims that can and cannot be made about EBI at this point based on the strength of available evidence.
Anna Ingeborg Petursdottir received her PhD from Western Michigan University. She is currently an associate professor of psychology and chair of the psychology department at Texas Christian University (TCU). She also holds an appointment as a part-time lecturer at Reykjavik University. Anna is a previous editor of The Analysis of Verbal Behavior, and a previous associate editor of both JABA and JEAB. She is past president of Division 25 of the America Psychological Association and the Texas Association for Behavior Analysis. Anna’s research encompasses both basic and applied interests and focuses primarily on verbal behavior acquisition, stimulus equivalence, and other derived stimulus relations.