Considering some controversies about the role of behavioral variation on the sensibilility of rule-following to programmed contingencies of reinforcement, the present study investigated whether a history of behavioral variation generated by different instructions would produce performances that are sensitive to signaled changes in environmental contingencies. Fourteen children, between ages of eight and nine years old were exposed to a matching-to-sample procedure. During each trial, a sample stimulus and two comparison stimuli were presented and then a light was switched on. In the presence of those stimuli, the participant was required to touch one of the comparison stimuli. Correct and incorrect responses were differencial consequences. The experiment consisted of two conditions: Condition 1: Only One instruction and Condition 2, Multiple Instructions. Both conditions consisted of three phases each. Phase 1 of the Condition with Only One Instruction (UI) was initiated by the presentation of instructions that corresponded to the contingencies. In this phase, selecting the comparison stimulus that was the same as the sample, was reinforced in the presence of a green light. Also, selecting the comparison stimulus that was different from the sample was reinforced in the presence of a red light. The contingencies in Phase 1 were reversed in Phase 2 and reestablished in Phase 3. Phase 1 of the Multiple Instructions Condition (MI) had three steps. Each step was initiated with a corresponding instruction. During Step 1, selecting the same stimulus was reinforced in the presence of a green light and selecting the different one was reinforced in the presence of a yellow light. During Step 2, selecting the same stimulus was reinforced in the presence of a yellow light and selecting the different one was reinforced in the presence of a red light. During Phase 3, selecting the same stimulus was reinforced only in the presence of a green light and selecting the different comparison was only reinforced in the presence of a red light. The contingencies in Step 3 were reversed in Phase 2 and reestablished in Phase 3. In both conditions, the transitions from one phase to another were signalized by the presentation of an instruction specifying that the participant should discover the best way to gain tokens. In each phase the participants were asked what they should do to gain tokens. Verbal responses were never reinforced. The results showed that the six participants of the IU Condition followed the instruction in Phase 1. In Phases 2 and 3, five participants continued following instructions, regardless of the changes in the contingencies. The eight participants of the MI Condition followed instructions in Steps 1, 2 and 3 in Phase 1. During Phases 2 and 3, four participants continued following instructions and four changed their performances in accordance with the reinforcement contingencies. The verbal behavior of all participants corresponded to the nonverbal during all phases and conditions of the experiment. Results suggest that the variability in the instructions as well as in the contingencies, before modifications in contingencies, together with the signalization of this alteration, may contribute to make the instructed behavior more sensitive to changes in contingencies.
|
