Equivalence Class Formation in Accuracy or Speed Conditions: Immediate Emergence, Adduction, and Retention

Abstract

Six pairs of participants were matched from a pool of 26 participants based on their performance in pretraining for conditional discrimination. One participant from each pair completed either a speed or an accuracy condition in the conditional discrimination training by using an linear series (LS) structure to establish potentially three, 3-member classes. Initially, the speed participants completed the training with a titrating limited hold (LH) on the comparison stimuli. The accuracy participants were yoked to the trial number of the speed participants with whom they were matched. In Phase 2, the speed participants completed matching to sample (MTS) training with a fixed 1,000-ms LH, and the accuracy participants were not yoked. In Phase 3, the probabilities of the programmed consequences were gradually reduced to 0 %. Phase 4 introduced tests for baseline and emergent relations and adduction. In Phase 5, a 2-week retention test was administered. The results indicated that 1 of 6 speed participants, compared with 3 of 6 accuracy participants, responded according to stimulus equivalence in the initial testing, whereas accuracy in the adduction test was nearly equal for all participants. In a 2-week retention test, no accuracy participants and two speed participants responded according to stimulus equivalence, whereas accuracy in the adduction test was equal between the participants. In the second retention test, 1 of 3 speed participants responded in accord with stimulus equivalence. All speed participants had a higher number of errors during the baseline trials and completed a higher number of acquisition trials. Inverse reaction time (InvRT) to comparisons did not vary as a function of trial type.