Academic journal article The Psychological Record

Equivalence Class Formation Via Common Reinforcers among Preschool Children

Academic journal article The Psychological Record

Equivalence Class Formation Via Common Reinforcers among Preschool Children

Article excerpt

Equivalence formation, expansion, and reversal were investigated as a result of arbitrary matching associated with specific reinforcers. Four 4- and 5-year-old normal children were taught identity matching with stimuli, A, B, C, and D, and stimulusspecific reinforcement. Then the children were taught two conditional discriminations AB and BC. All subjects showed formation of the ABC stimulus classes; one subject, however, did not show expansion to ABOD classes. This subject was taught to name the D stimuli, he then demonstrated the expanded class. Next, 2 subjects who showed expanded classes were taught identity matching with the reinforcers reversed for the D stimuli. In tests that followed, their matching responses remained consistent with the original equivalence classes. These subjects were then taught to reverse the names for the D stimuli. As a result, the children reversed the classes. The results suggest that class expansion and subsequent reformation of classes may be facilitated when stimulus withi n each class controls a common naming response.

Most of the recently published studies concerning the development and the expansion of equivalence stimulus classes have focused on stimulus-stimulus relations (Devany, Hayes, & Nelson, 1986; Saunders, Saunders, Kirby, & Spradlin, 1988; Sidman, Kirk, & Willson-Morris, 1985; Sidman & Tailby, 1982). However, research by Dube, Mcllvane, Maguire, Mackay, & Stoddard (1989), and Schenk (1994), has demonstrated that stimulus equivalence classes may be expanded as a function of stimulusreinforcer relations. Dube et al. (1989) taught identity matching and arbitrary matching to retarded subjects using two different reinforcers, one for each potential class of stimuli. Throughout training, correct selections of Al, B1, C1, and Dl were always followed by food (R2), and correct selections of A2, B2, C2, and D2 were followed by drink (R2). On identity-matching trials, in the presence of Sample A1, selections of Al were reinforced with R2, and in the presence of A2, selections of A2 were reinforced with R2, and similarly, for Stimulus Sets B, C, and D. On arbitrary-matching trials, in the presence of Sample Al, selections of B1 were reinforced with R1 and, similarly, in the presence of B1, selections of C1 were also followed by R1. In the presence of A2, responses to B2 were reinforced with R2, as were responses to C2 when B2 was the sample.

In the tests that followed, both subjects showed formation of the class comprising A, B, and C stimuli, and also expansion to encompass the D stimuli. That is, they were able to match Dl to Al, B1, and C1 and D2 to A2, B2, and B2 as well as Al, B1, and C1 to Dl, and A2, B2, and C2 to D2 demonstrating two classes of stimuli: Al-B1-Cl-Dl and A2-B2-C2-D2.

Dube et al. (1989) noted that the classes entailing the stimuli A, B, and C could have resulted purely from the AB and BC arbitrary-matching training but, because the D stimuli never appeared on arbitrary-matching trials, their inclusion in the classes must have been based on relations with the common reinforcers during the identity-matching training.

The role of the common reinforcers was later confirmed in a second stage of the experiment. In that stage, Dube et al. (1989) reversed the reinforcement contingencies on the identity-matching trials with the D stimuli. Matching Dl to Dl now produced R2, drink, and matching D2 to D2 now produced R1, food. This reversal in the reinforcement contingencies also produced a corresponding reversal in the stimulus class membership of the D stimuli. Now the classes consisted of Al-B1-C1-D2 (all the stimuli now associated with one particular type of reinforcer, food, R1), and A2-B2-C2-Dl (all the stimuli now associated with the other type of reinforcer, drink, R2). The rearrangement of classes with the D1 stimulus moving into the A2, B2, and C2 class and the D2 stimulus into the Al, B1, and C1 class demonstrated that stimulus class expansion can occur on the basis of stimulus reinforcer relations. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.