STUDENT PAPER WINNER
THE IMPACT OF CLASS-SPECIFIC REINFORCERS ON CONDITIONAL DISCRIMINATION ACQUISITION AND EQUIVALENCE CLASS PERFORMANCES
Holly Smith Schomer
UNIVERSITY OF NORTH CAROLINA, WILMINGTON
Over the past decade, many researchers have attempted to identify variables that affect the emergence and stability of equivalence classes (e.g. Sidman and Tailby, 1982). This laboratory has used reversal studies to examine emergent relational stimulus control (Pilgrim, Chambers, & Galizio, 1995; Pilgrim & Galizio 1990; 1995). In these reversal studies, match-to-sample procedures were used to train a series of arbitrary conditional discriminations and test for the emergence of equivalence classes. The original reinforcement contingencies were then altered and effects on the emergent properties of equivalence were assessed.
In order to examine how equivalence relations are formed, changed, and maintained, at least two conditional discriminations must be acquired and stable prior to any baseline manipulations. However, several researchers have reported difficulties in teaching arbitrary conditional discriminations to young children (e.g. Lipkens, Hayes, & Hayes, 1993; Zygmont, Lazar, Dube, McIlvane, 1992; Pilgrim, Jackson, & Galizio, 2000). As a result, a variety of special training procedures have been employed in effort to facilitate conditional discrimination acquisition (e.g. stimulus-control shaping, instructions, sample-naming, pre-training sequences). When researchers have successfully established the original conditional discriminations, findings for post-reversal probe performances have been mixed. In the only two published studies with children, participants have demonstrated erratic responding where patterns were consistent with neither the original nor the reversed contingencies (Pilgrim, Chambers, & Galizio, 1995), and results that were mostly, although not perfectly, consistent with the original contingencies (Saunders, Drake, & Spradlin, 1999).
In the majority of stimulus-equivalence studies, a single reinforcer has been used to reinforce all responses deemed correct. According to recent theoretical treatments of equivalence-class formation (Sidman, 2000), this may be problematic. Several studies have demonstrated that the reinforcer can actually become a member of the equivalence class (e.g. Dube & McIlvane, 1995). If each of several equivalence classes shares a common reinforcer, and the reinforcer is actually a member of the classes, there may be a class collapse or merger. What should be separate classes may become one large class via the relation between each stimulus and the common reinforcer (Sidman, 2000). Such a collapse could potentially interfere with successful conditional discrimination and equivalence-class formation. To avoid such difficulties, Sidman suggested that training procedures use class-specific reinforcers.
The present research examined the effect of class-specific reinforcement (CSR) on children's conditional discrimination acquisition and equivalence probe performances. Eight children ages 4-10 years learned two arbitrary conditional discriminations (AB and AC) using match-to-sample procedures. Given A1, A2, or A3 selections of B1, B2, or B3 produced Reinforcer 1, Reinforcer 2, or Reinforcer 3, respectively. Likewise, given A1, A2, or A3, selections of C1, C2, or C3 produced R1, R2, or R3, respectively. Following mastery of baseline training, unreinforced probe trials were introduced to test for the properties of symmetry, transitivity/equivalence, and reflexivity. Once three three-member equivalence classes were demonstrated, reinforcer-probe sessions were given to determine whether or not the reinforcer had become a member of the equivalence classes. The reinforcement contingencies for the AC conditional discriminations were then reversed (given A1, A2, or A3, selections of C2, C3, or C1 produced R1, R2, or R3, respectively), and probe trials were presented again to test for class flexibility. A control group of six children was trained using procedures identical to those used with the experimental group, except that during the initial conditional discrimination training control participants were presented with a single reinforcer for each correct response. If no acquisition trend was demonstrated following 10 sessions with single-reinforcer training, class-specific reinforcers were introduced.
Unlike participants in past studies, seven out of eight children in the CSR condition readily acquired the baseline discriminations without the aid of any other special training techniques (i.e., pre-training, naming, instructions). In contrast, none of six control-group children showed signs of acquisition, but all mastered the conditional discrimination readily when CSR was introduced. After mastering the conditional discriminations, six of six participants quickly demonstrated the emergence of equivalence classes (A1B1C1, A2B2C2, A3B3C3). For all but one of the participants who completed reinforcer probes, reinforcer-probe response patterns indicated that each reinforcer had become a member of its respective equivalence class.
Upon reversal of the AC contingencies, five children once again mastered the baseline conditional discriminations. For the three children who completed the post-reversal probe condition, performances were unlike those seen in previous studies involving children (Pilgrim, Chambers, Galizio, 1995), in that patterns were indicative of three distinct equivalence classes. Two of the participants required the minimum number of sessions to meet stability on the reversal probes; however, their probe response patterns were quite different. The youngest (age 5) of the two participants demonstrated reversal-probe and reinforcer-probe response patterns consistent with the original contingencies, despite the reversed AC conditional discrimination. The oldest participant (age 10), however, showed reversal-probe and reinforcer-probe performances in accordance with the reversed contingencies. The third participant (age 7) required more sessions to meet reversal-probe stability criterion. Her initial response patterns were largely consistent with the original contingencies. However, following several cycles of testing, modified (reversed) patterns of responding emerged. Although the final post-reversal pattern differed across individuals, all three children showed evidence of three distinct equivalence classes. These data suggest that class-specific reinforcement enhances the acquisition of arbitrary conditional discriminations and provides the basis for distinct equivalence classes.
REFERENCES
Dube, W.V., & McIlvane, W.J. (1995). Stimulus-reinforcer relations and emergent matching to sample. The Psychological Record, 45, 591-612.
Lipkens, R., Hayes, S.C., & Hayes, L.J. (1993). Longitudinal study of the development of the derived relations in an infant. Journal of Experimental Child Psychology, 56, 201-239.
Pilgrim, C., Chambers, L., & Galizio, M. (1995). Reversal of baseline relations and stimulus equivalence: II. Children. Journal of the Experimental Analysis of Behavior, 63, 239-254.
Pilgrim, C., & Galizio, M. (1990). Relations between baseline contingencies and equivalence probe performances. Journal of the Experimental Analysis of Behavior, 54, 213-224.
Pilgrim, C., Jackson, J., & Galizio, M. (2000). Acquisition of arbitrary conditional discriminations by young normally developing children. Journal of the Experimental Analysis of Behavior, 73, 177-193.
Saunders, R.R., Drake, K.M., & Spradlin, J.E. (1999). Equivalence class establishment, expansion, and modification in preschool children. Journal of the Experimental Analysis of Behavior, 71, 195-214.
Sidman, M., & Tailby, W. (1982). Conditional discrimination vs. matching to sample: An expansion of the testing paradigm. Journal of the Experimental Analysis of Behavior, 37, 5-22.
Sidman, M. (2000). Equivalence relations and the reinforcement contingency. Journal of the Experimental Analysis of Behavior, 74, 127-146.
Zygmont, D.M., Lazar, R.M., Dube, W.V., & McIlvane, W.J. (1992). Teaching arbitrary matching via sample stimulus-control shaping to young children and mentally retarded individuals: A methodological note. Journal of the Experimental Analysis of Behavior, 57, 109-117.