Learning symmetrical relations along conditional discrimination training

Abstract

In order to avoid testing emergent conditional relations in extinction, Velasco and Tomanari (2009) developed a procedure to evaluate symmetry during the acquisition of conditional discriminations. Human adults were trained on baseline and their corresponding symmetrical relations, in addition to control relations that were trained in the absence of their corresponding symmetrical relations. All participants learned the symmetrical relations faster than the control relations. However, due to the baseline training, symmetricalrelation stimuli were displayed more frequently than control-relation stimuli whose baseline had not been trained. To control for this difference, two experiments were conducted with human adults as participants. The first balanced the number of training trials of each symmetrical and control relations, although stimuli of the former relations were displayed more frequently than those of the latter. The second experiment balanced the frequency of stimuli presentation on both groups of relations. In general, symmetrical relations were acquired more accurately and faster than control relations, regardless the stimuli were presented with higher (Experiment 1) or equal frequency (Experiment 2) during training. These results are in line with former findings, as they confirm and expand methodological alternatives to testing emergent relations in the absence of reinforcement.