In Behavior Analysis, several studies have been carried out in order to understand how
behavior can be productively controlled by arbitrarily related events, through
equivalence class formation. The inclusion of reinforcer stimuli in the classes has been
pointed as a possible facilitator of their formation. The present study aimed to evaluate
the formation of equivalence classes mediated by reinforcer stimuli with children with mild learning difficulties. By using normally developing children and in a larger number
we intended to obtain data with less across subject variability, which is commonly found in this kind of research. It was used specific reinforcement with fourteen children that showed mild learning difficulties at school (Experiment 1). The experiment comprised 10 phases. In all phases, there were specific reinforcers (fruit or toys) to each of the potential classes. Initially we carried out an identity matching-to-sample
procedure with the stimuli of the sets A (A1 and A2), B (B1 and B2) and C (C1 and C2) in a continuous reinforcement schedule, followed by an intermittent reinforcement schedule. Then, tests for emergent relations ABBA, ACCA, and BCCB were carried out. Before each test, we returned to the baseline training. Data from Experiment 1 show great across subject variability. The Experiment 2 was to investigate the effect of two different kinds of pre-training on the performance in tests for class formation. Six children participated and they were divided into two groups. Group 1 was trained with identity matching-to-sample and Group 2 with arbitrary matching-to-sample. The
results confirm partly the hypothesis that arbitrary MTS as pre-training may reduce across subject variability in this kind of research, because high rates of variability were found in this study. Further research will explore this possibility more systematically.
|
