The Expression of equivalence classes influenced by distractors during DMTS test trials

Abstract

When training and testing are conducted with trials presented in a delayed matching to sample (DMTS) format, mediating behaviors may occur during the delay that separates sample offset and comparison onset. In some cases, the mediating behavior can interfere with or enhance the maintenance of correct responding. The present experiment studied how mediating behavior could have influenced responding, according to equivalence classes by manipulation of a distractor task during the delay interval in DMTS test trials for derived relations. After training the baseline relations for equivalence classes, the emergent relations tests were presented with 6-s delays. In Condition A, participants had to solve addition and subtraction tasks during the delay intervals in the emergent relations test trials. In Condition B, no mediating task was presented in the delay interval. These conditions were presented in an ABA or a BAB sequence for different individuals, thereby providing a within-subject assessment of the effects of the distractor task. Regardless of sequence, responding according to equivalence occurred when distractors were not included in the delay interval, and responding according to equivalence was suppressed when distractors were included in the delay interval. The inclusion of a distractor task did not eliminate an already formed class; rather it disrupted responding controlled by class. It also did not interfere with the formation of the equivalence classes. Processes responsible for this outcome were discussed