Association for Behavior Analysis International

The Association for Behavior Analysis International® (ABAI) is a nonprofit membership organization with the mission to contribute to the well-being of society by developing, enhancing, and supporting the growth and vitality of the science of behavior analysis through research, education, and practice.


45th Annual Convention; Chicago, IL; 2019

Event Details

Previous Page


Symposium #34
CE Offered: BACB
Equivalence Class Formation and Errorless Learning: Theory and Application
Saturday, May 25, 2019
10:00 AM–11:50 AM
Hyatt Regency West, Lobby Level, Crystal Ballroom A
Area: DDA/EAB; Domain: Translational
Chair: Russell W. Maguire (Simmons College)
Discussant: Julian C. Leslie (Ulster University)
CE Instructor: Julian C. Leslie, Ph.D.

This symposium will address advanced topics in the stimulus control of behavior in the experimental and applied domains. All presentations, two experimental and two applied, investigated how a variety of stimulus conditions impacted the formation of stimulus classes under. The first presentation manipulated the physical properties of stimuli to form generalized equivalence classes (money equivalences) to induce new applied repertoires (purchasing skills) that emerged in novel settings. The second presentation compared contingent and non-contingent reinforcement during errorless instruction and determined that the influence of a “hidden contingency” accounted for errorless learning in the absence of contingent reinforcement. The third presentation evaluated different training modalities on the subsequent formation of equivalence classes. The results suggested that stimulus control topography coherence theory may require revision. Finally, the fourth presentation investigated whether or not prompts used during training entered into equivalence classes and expanded those classes. The results indicated that class-specific prompts became members of relevant classes and could be used to efficiently expand those classes. The outcomes of these studies are discussed in terms of the development of novel forms of stimulus control and improving the efficacy of instruction of complex behavior.

Instruction Level: Intermediate
Keyword(s): Errorless Instruction, Stimulus Equivalence
Target Audience:

Graduate students conducting research in the areas of complex stimulus control and derived relational responding; Individuals designing instruction for children with and without autism and developmental disabilities,

Learning Objectives: 1. Describe the defining properties of stimulus equivalence; 2. Define stimulus control topography coherence theory; 3. Explain how stimulus equivalence and stimulus control topography coherence theory can be applied to instruction.
Using A Generalized Equivalence Class Strategy to Teach Functional Money Skills
(Applied Research)
MEGAN BREAULT (Realizing Children's Strengths Learning Center; Simmons University), Russell W. Maguire (Simmons College), Christina M. King (Realizing Children's Strengths Learning Center; Simmons University), Colleen Yorlets (Realizing Children's Strengths, Behavioral & Educational Consulting; Simmons University)
Abstract: A generalized equivalence class is demonstrated when reflexive, symmetrical, and transitive relations emerge among stimuli that are perceptually disparate, and others that are perceptually similar to the members of the base equivalence class. In this study, a participant with autism learned to match coins (B) and items to be purchased (C) and to price tags (A). After three-member classes were established, tests determined whether untrained but physically similar items (C’) entered into the established class, thus expanding the class beyond the original three members. While equivalence classes formed, we found incomplete inclusion of the novel, physically similar items (C’). A second systematic replication will seek to produce complete inclusion of the novel but physically similar items (C’) into the established classes by systematically manipulating those physical properties. The results of these studies are discussed in terms of procedures that may yield a variety of generalized equivalence classes (i.e., minimally, partially and fully elaborated).
Learning in Stimulus Fading by Response-Contingent Reinforcement and by Response-Contingent Stimulus Change
(Basic Research)
MARGOT BERTOLINO (University of Lille), Vinca Riviere (University of Lille ), Lanny Fields (Queens College, City University of New York)
Abstract: This experiment explored the influence of the reinforcement contingency on the acquisition of all discriminations in a stimulus fading (SF) protocol by studying two conditions. In the stimulus fading condition (SF) contingent reinforcement was used throughout the SF protocol. In the yoked control condition (YC) non-contingent reinforcers were presented through the protocol. In both protocols, a participant had to learn up to eight increasing difficult discriminations arrayed along a dimension of luminance difference. All eight discriminations were acquired by 18 of 20 in the SF protocol and 2 of 20 in the YC protocol. Of the 18 in the YC protocol who did not learn all eight discriminations, many of intervening discriminations were acquired even though no contingency of reinforcement was active. When errors were considered, very few occurred during the SF protocol (errorless learning) while many more occurred during the YC protocol. Thus, while the contingency of reinforcement played a significant role in learning in stimulus fading, a “hidden” contingency– response produced stimulus change - was responsible for the learning of the discriminations in the absence of the contingency of reinforcement, and non-contingent reinforcement impeded the acquisition of the discriminations. Since response-produced stimulus change is present in all fading protocols, it could also influence discrimination learning in fading, regardless of procedural variation. Finally, failures in stimulus fading might also define a participant’s differential threshold for luminance differences.

Training Modality and Equivalence Class Formation: A Test Of Stimulus Control Topography Coherence Theory

(Basic Research)
Lanny Fields (Queens College, City University of New York), DEBRA PAONE (Douglass Developmental Disabilities Center)

This experiment explored how training influenced the formation of 3-node 5-member equivalence classes during the simultaneous protocol. The baseline relations were established concurrently (CONC) or serially (SER) on a trial and error basis, or serially and “errorlessly” with a constructed response matching to sample procedure (CRMTS). After training, test blocks were administered to measure class formation. Test blocks trials contained all baseline relations and probes for symmetry, transitivity, and equivalence. The error percentages while acquiring the baseline relations were highest during concurrent training, lower during serial training, and lowest during constructed response training. Yet, similar percentages of participants formed classes in each training condition. Thus, the likelihood of equivalence class formation under the simultaneous protocol was not influenced by training modality or prevalence of errors during baseline acquisition. In addition, transient stimulus control topographies that emerged during training did not subsequently resurge during testing, thus, their resurgence did not account for failed class formation. Because the error and resurgence findings were not consistent with stimulus control topography coherence theory, it might have to be revised to accommodate to the data reported in this experiment.

The Inclusion of Prompts in Equivalence Classes
(Applied Research)
SIMONE VILAS BOAS PALMER (Simmons College and Crossroads School), Russell W. Maguire (Simmons College), Karen M. Lionello-DeNolf (Assumption College), Paula Ribeiro Braga Kenyon (Trumpet Behavioral Health)
Abstract: Sidman (2000) posited that equivalence relations may include all elements of a conditional discrimination (e.g., sample and comparison stimuli, responses, and reinforcers). Research has verified this outcome. However, establishing conditional discriminations may involve the use of supplementary stimuli, called prompts, to occasion the correct responding. To date the question of whether or not prompts may enter into the relevant equivalence class has not be answered. Experiment 1 taught graduate students visual-visual matching-to-sample relations with arbitrary stimuli. Initially, the S+ stimulus on each trial was highlighted using a class-specific prompt (e.g., colors: class 1 = blue; class 2 = red; class 3 = yellow). Contingent on correct responding the prompt was systematically faded until six conditional discriminations were acquired, in the absence of the color prompt (A1-B1; A2-B2; A3-B3; A1-C1; A2-C2; and A3-C3). Following this training, testing documented the formation of 3-three member equivalence classes. Identity matching-to-sample training was then conducted, again using the color prompts (D-D). Subsequent testing revealed that the class-specific prompts (colors) became members of relevant equivalence class, established during training and testing and expanded the classes to four members. Implications for teaching students with developmental disabilities and increased efficacy of instruction are discussed.



Back to Top
Modifed by Eddie Soh