Twelve children in the eight-to-nine year age range were exposed to a matching-to-sample procedure in an investigation dealing with the role of monitoring in instruction-following. On this task, two comparison stimuli were touched in the presence of a contextual stimulus. Whereas incorrect responses resulted in reinforcement loss, correct responses avoided this consequence. There were five phases in the experiment: Phases 1, 3 and 5 began with an instruction corresponding to the contingencies; contingencies in effect in Phase 1 were reversed in Phase 2, reestablished in Phase 3, reversed again in Phase 4 and renewed in Phase 5. The participants were subjected to two conditions, differing with regard to the phase where an observer was introduced in the experimental situation. During Phases 2 and 4, 10 participants did not follow instructions. This finding indicates that even when monitoring is present, instruction-following does not occur if it is not reinforced.
Rule-governed behavior; monitoring; reinforcement loss; matching-to-sample; children
