Clases de quivalencia de estímulos y actitudes

Clases de quivalencias y actitudes

STIMULUS EQUIVALENCE AND ATTITUDES

IAN M. GREY
University of Wales

DERMOT BARNES
University Col/ege Cork

Two studies examine the contribution of stimulus equivalence procedures to the study of attitude formation and attitude change. In Experiment 1, using a matching-ta-sample procedure, subjects were trained to form three, three-member equivalenceclasses (A1, B1, C1; A2, B2, C2; A3, B3, C3) using nonsense syllables as stimuli. One member from two of these classes (B1 and B2) was placed on one of two video cassettes in the form of a label. Each video contained scenes of either a romantic or religious nature which subjects viewed. Subjects were then presented with four new videos. These videos were labeled with the remaining nonsensesyllables that were used in the equivalence training (i.e., A1, C1, A2, C2). SUbjects were then asked to categorize these unseen videos as "good" or "bad:' This stage represented our attempt to investigate attitude formation to objects never directly experienced. In the next stage of the experiment which examined attitude change, subjects were exposed to a violent sexual video [which was labeled withone of the nonsense syllables from the remaining class (B3) from the equivalence training] and were again exposed to the categorization tasks outlined above. In the second experiment, contextual control through equivalence relations was incorporated into the procedure to investigate whether performance on the categorization tasks could be manipulated. Results from both experiments indicate thatsubjects' behavior was in accordance with the expected equivalence relations, and that a stimulus equivalence model of attitude formation and change has practical benefits for the study of attitudes. In recent times, the concept of stimulus equivalence is being called on more and more frequently to support behavioral accounts of cognition and language (Barnes & Holmes, 1991; Hayes, 1989; Sidman,1990).
The experimental work was conducted as part of Ian Grey's doctoral research program at University College, Cork. He thanks Dermot, Janelle, Martina, Ann, Robert "M.M" Jones, and, Siadjana for help and encouragement. An appendix giving the details of the films used in both experiments can be obtained from Ian Grey on request. Reprint requests should be sent to Ian M. Grey, School of Psychology,University of Wales, Bangor, Gwynedd LL57 2DG, United Kingdom or to Dermot Barnes, Department of Applied Psychology, University College, Cork, Cork, Ireland.

244

GREY AND BARNES

EQUIVALENCE AND AlTITUDES

245

The concept of equivalence relations was developed by Sidman (1971) to describe human subjects' untrained performance on the matching-to­ sample procedure. In its most basicform, stimulus equivalence involves training a subject in at least two conditional discriminations (e.g., given A pick B, and given B pick C). After this training, subjects presented with C as a sample often choose A from a range of comparison stimuli. This is a new relation that often "emerges" (i.e., C to A) in the absence of explicit training and the relation is normally called an equivalencerelation. Closely related to emergent relations is the property of transfer of functions and "it is this that provides the general psychological importance to stimulus equivalence" (S. C. Hayes & L. J. Hayes, 1992, p. 1389). Suppose that a stimulus A has a specific behavioral function (e.g., the word "Slow"is discriminative for decelerating in a driving context). If a derived relation is thenestablished between A and another stimulus B (e.g.• the Welsh word "Araf" which means "Slow"), in certain contexts (e.g., driving in Wales) B may acquire a new behavioral function (e.g., I slow down when I see the word "Araf" when driving). This behavioral outcome is not based on a history of experiencing the aversive consequences of ignoring the sign "Araf," rather, it may involve a transfer of...