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.


49th Annual Convention; Denver, CO; 2023

Event Details

Previous Page


Symposium #191
CE Offered: BACB
Some Important Variables for the Formation of Stimulus Equivalence Classes
Sunday, May 28, 2023
5:00 PM–6:50 PM
Hyatt Regency, Centennial Ballroom A
Area: EAB/VRB; Domain: Basic Research
Chair: Erik Arntzen (Oslo Metropolitan University)
Discussant: Karen M. Lionello-DeNolf (Assumption University)
CE Instructor: Erik Arntzen, Ph.D.

The main purpose of the present symposium is to present some results on variables influencing emergent relations. The first paper by Arntzen and Thomassen presents an experiment studying (1) how simple discrimination training can produce stimuli for testing of expansion of equivalence classes documented by sorting tests and (2) self-talk during training and testing. In the second paper, Viela and Tomanari present an experiment focusing on time as a stimulus in procedures involving conditional discrimination. The authors employed a One-to-Many (AB/AC) training structure in which A was different time intervals, and B and C stimuli were colored images. The third paper by Fields analyzes one of the four parameters (Fields & Verhave, 1987) that could influence responding in accordance with stimulus equivalence, that is, the distribution of singles on equivalence class formation. The last paper by Silguero and Vaidya presents results from an experiment on differential outcomes effect when establishing conditional discriminations, and they propose mechanisms for partitioning of subclasses.

Instruction Level: Intermediate
Keyword(s): self-talk, stimulus equivalence, time, variables
Target Audience:


Learning Objectives: At the conclusion of the presentation, participants will be able to: (1) know how time could be a part of equivalence classes; (2) the four different structural parameters influencing equivalence class formation; (3) the use of self-talk in sorting tests
Equivalence Class Formation: Sorting, Class Expansion, and Self-Talk
ERIK ARNTZEN (Oslo Metropolitan University), Linn Thomassen (Oslo Metropolitan University)
Abstract: The purpose of the experiment was twofold in studying: (1) how simple discrimination training can produce stimuli for testing of expansion of equivalence classes documented by sorting tests and (2) self-talk during training and testing. Twenty-four adults were exposed to three experimental phases. In the first phase, the participants were trained on 12 conditional discriminations arranged as a linear series training structure, with A, B, D, and E stimuli being abstract shapes, and C stimuli being meaningful pictures, and a sorting test. The second phase included simple discrimination training of C stimuli, training two, four, and six key presses in the presence of C1, C2, and C3. Half of the participants were exposed to an expansion test (F1, F2, and F3 stimuli were 2, 4, and 6, respectively), including AF, BF, DF, and EF relations. The final phase contained sorting and MTS tests. A talk-aloud procedure was implemented in different parts of the experiment. The main findings were that all participants talked aloud during the simple discrimination training, 92% during the expansion test, and 79% during the second sorting test. Furthermore, with and without exposure to the expansion test did not influence the percentage who formed three 6-member classes.

Time as Part of Equivalence Classes

Eduardo Cunha Vilela (University of São Paulo, Brazil), GERSON YUKIO TOMANARI (Universidade de Sao Paulo)

Time is a fundamental part of operant behavior, as studies on schedules of reinforcement, timing and temporal discrimination have shown. Data on how time may function as stimulus in a conditional discrimination and how different intervals may be part of equivalence classes are the main contributions to the symposium. Participants underwent a matching-to-sample (MTS) training procedure that established the equivalence classes A1B1C1D1E1F1 and A2B2C2D2E2F2. A one-to-many training structure was used, in which set A were time intervals (300 ms and 1200 ms) and sets B and C were colored images. Class formation was evaluated by three distinct measures: the emergence of conditional relations among B, C, D, E and F stimuli; sorting the stimuli according to the derived equivalence classes; and assigning durations to the visual stimuli by pressing a key on the keyboard according to the corresponding Set A. Results from all measures demonstrated equivalence classes; however, a series of procedural adjustments had to be experimentally controlled, what highlights some particulars of time as a source of discriminative control.


CANCELLED: Distribution of Singles, Nodal Density and The Structure of Equivalence Classes

Lanny Fields (Queens College, City University of New York)

The structure of any equivalence class is defined by four parameters (a) number of class members, (b) the number of nodes in a class, (c) training directionality, and (d) the distribution of singles attached by training to a node. To date, many studies have shown that variation in the values of the first three parameters influences the likelihood of class formation and/or the relatedness of stimuli in a class. To date, the effects of the fourth, the distribution of singles, on equivalence-based performances have received minimal attention. The only exception is Nedelcu, Fields, and Arntzen (2015). This presentation will consider the fourth parameter in detail, review the findings presented by Nedelcu, et al (2015), indicate how some other published studies might also reflect the effects of density of singles on class formation, suggest how singles density can account for the enhancement of class formation by meaningful stimuli, and finally, suggest some new studies that might show additional effects of density of singles on equivalence class formation.


What Happens When a Stimulus “Drops Out” of an Equivalence Class? Revisiting Sidman’s (2000)

RUSSELL SILGUERO (University of North Texas), Manish Vaidya (University of North Texas)

Sidman’s (2000) theory of stimulus equivalence states that when an individual satisfies a reinforcement contingency, all positive members of the contingency will become members of an equivalence class. When two reinforcement contingencies contain elements from the same equivalence class, there is a conflict between the equivalence class and the contingencies. Sidman proposed that under these conditions, the class members that conflict with the contingency will “drop out” of the equivalence class. A study by Minster et al. (2006) tested Sidman’s “dropping out” hypothesis and found evidence that the conflicting member does not drop out of the class and that it continues to mediate derived performances. Their results can be interpreted in terms of an equivalence class partitioned into subsets of functional classes. However, their procedures do not clearly show the process by which these subclasses develop. We employed a variation of their procedures by first establishing equivalence classes before conditional discrimination training. We then provide a clear picture of the process of subclass differentiation, observed as differential outcomes effect in conditional discrimination learning. We discuss how Sidman’s “dropping out” hypothesis relates to this and other empirical work.




Back to Top
Modifed by Eddie Soh