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.


43rd Annual Convention; Denver, CO; 2017

Event Details

Previous Page


Symposium #87
Stimulus Control Questions and Equivalence Classes
Saturday, May 27, 2017
3:00 PM–4:50 PM
Hyatt Regency, Centennial Ballroom B/C
Area: EAB/VRB; Domain: Basic Research
Chair: Erik Arntzen (Oslo and Akershus University College)
Discussant: Francis Mechner (The Mechner Foundation)
Abstract: Questions about stimulus classes and the analytic units have been discussed quite much within behavior analysis. The purpose of the present symposium is to present research which is going enlighten about stimulus control and equivalence classes. Results from experiments with both humans and nonhumans will be presented. In the first paper by Fields, presents an experiment on Errorless transfer of stimulus control, fading without reinforcement, and the induction of resemblance based stimulus classes (Figure 1). In the second paper by Arntzen, Nartey, and Fields, present an experiment on Malleability of Equivalence Classes: How Reorganization of Classes Influences Formation of New Classes (Figure 2). In the third paper Tomanari and Grisante, present an experiment on Sample/S- relations acquisition in a three-choice Matching-to sample procedure that requires Observing Responses (Figure 3). In the fourth paper by Vaidya, Stancato, and Condon, present an experiment on further explorations of interactions between the development of analytic units and equivalence relation (Figure 4).
Instruction Level: Advanced
Keyword(s): stimulus equivalence

Errorless Transfer of Stimulus Control, Fading Without Reinforcement, and the Induction of Resemblance Based Stimulus Classes

LANNY FIELDS (Queens College, City University of New York)

Errorless transfer of stimulus control in the absence of reinforcement (i. e., stimulus fading) was studied with eight pigeons. First, participants learning to respond left to red and right to green by use of differential reinforcement. In contrast, white horizontal and white vertical lines do not control these responses. Then, low salience lines were superimposed on the colors with horizontal on red and vertical on green. After gradually increasing the salience of the lines, color salience was gradually attenuated in the absence of reinforcement. Eventually, the line stimuli in the absence of the colors- occasioned the responding trained to the colors: left to horizontal and right to vertical. This occurred for six of the eight pigeons in less than 30 minutes. Thus, errorless transfer of stimulus control occurred without the reinforcement. Errorless transfer then was produced by stimulus-stimulus pairings. Since fading entailed many fading levels, the procedure was a form of multiple exemplar training. Thus, fading may have involved the induction of resemblance based classes, which influenced errorless learning.

Malleability of Equivalence Classes: How Reorganization of Classes Influences Formation of New Classes
ERIK ARNTZEN (Oslo and Akershus University College), Richard Nartey (Akershus University College), Lanny Fields (Queens College, City University of New York)
Abstract: In Condition 1, adults attempted to form equivalence classes A1-B1-C1-D1-E1, A2-B2-C2-D2-E2, and A3-B3-C3-D3-E3. In the ABS and four preliminary training groups, classes contained abstract shapes. Classes in the PIC group contained one picture and four abstract shapes. Before class formation, each preliminary training group learned either identity (CC) or arbitrary (CX) relations established in a simultaneous or delayed matching format. Thereafter, participants in all group were exposed to Condition 2 that involved the training of BC and CD relations that contained C stimuli linked to other class than that established in Condition: B1-C2 and C2-D1, B2-C3 and C3-D2, along with B3-C1 and C1-D3. Follow-up derived relations tests assessed the emergence of the reorganized equivalence classes A1-B1-C2-D1-E1, A2-B2-C3-D2-E2, and A3-B3-C1-D3-E3. Without preliminary training, the same likelihoods of class formation occurred in Conditions 1 and 2: low and high in the ABS and PIC groups, respectively. In Condition 1, preliminary training of either ID and ARB relations in the simultaneous format produced modest increases in class formation. In contrast, large increases in class formation occurred when these relations were formed with delays; indeed, these yields matched those in the PIC group. In Condition 2, all four forms of preliminary training led to the formation of reorganized classes with likelihoods equaling that in the PIC group. Changes in group based yields from Condition 1 to 2 were driven by unique performances of individual participants. Derived relations test performances showed how preliminary training influenced the speed and types of errors made when classes were not formed. Thus, the content of an equivalence class is determined by the current relations that have been learning, while the likelihood of class formation is influenced by the prior learning of other relations. These results also clarify how the class enhancing effects of a meaningful stimulus may reflect the prior acquisition of many relational functions in the normal course of events.

CANCELED: Sample/S- Relations Acquisition in a Three-Choice Matching-to-Sample Procedure That Requires Observing Responses

GERSON YUKIO TOMANARI (Universidade de Sao Paulo), Priscila Crespilho Grisante (Universidade de São Paulo)

A relatively well-accepted hypothesis about equivalence class formation suggests that emergence of programmed equivalence relations depends on the nature of controlling relations learned during training. We investigated the establishment of Sample/S- controlling relations in a modified three-choice MTS procedure. Black squares covered stimuli during experiment. Participants could briefly observe one stimulus at a time contingent to observing responses (OR) to a button below each square. OR were not mandatory. A software presented S+ as the last stimulus displayed to participants in 80% of trials to promote Sample/S- relations (choosing the square after observing both S-; i. e. without observing S+). Five children and three adults were participants. All adults and four children met learning criterion and presented equivalence class formation. Two adults and four children presented responding under Sample/S- controlling relations at the end of training phase. Children and adults showed similar patterns of controlling relations acquisition, which is, starting to respond under control of Sample/S- relations after occurrences of correct choices that included S+ observation. Responding controlled by Sample/S- relations became preponderant at the end of training. Results suggests that sample/S- relations acquisition not necessarily disrupts equivalence class. Additionally, allowed to provide a description of sample/S- relations acquisition process.


Further Explorations of Interactions Between the Development of Analytic Units and Equivalence Relations

MANISH VAIDYA (University of North Texas), Stefanie S. Stancato (University of North Texas), Manish Vaidya (University of North Texas)

Sidmans (2000) theory of stimulus equivalence suggests that equivalence relations arise out of the contingencies that also develop analytic units. As such, the theory predicts an interaction between the development of analytic units and the development of equivalence relations. Vaidya & Brackney (2014) documented one such type of interaction for groups of stimuli, simple discriminative functions were acquired more rapidly when the groups were drawn from within the an equivalence class than when the groups were drawn from different equivalence classes. Thus, equivalence relations facilitated the development of analytic units. The current set of studies further explore these interactions. Can equivalence relations also retard the development of analytic units? Can existing analytic units (such as simple discriminations) influence the development of equivalence relations? This address will present data addressing each of these questions. The implications of these analyses for developing stimulus control in applied settings will be discussed.




Back to Top
Modifed by Eddie Soh