Association for Behavior Analysis International

The first paper by Arntzen includes an experiment with college students focusing on yields (number of participants who form equivalence classes) as function of training trials. The main finding was that yields increase as function of number training trials in a linear series training structure combined with a simultaneous training and testing protocol. The second paper by Fields, Wakim, and Foxe studied event-related potential (ERP) in college students who formed 2-node 4-member classes. Some of findings were that the baseline in contrast derived relations generated activation 400 ms after target onset. The third paper by dos Santos, Carvalho, and de Rose includes an experiment in which they have used a game, labyrinth, to investigate equivalence class formation and also with the use of remote data collection. The main findings were that the game was a viable option to conduct remote data collection, and that the nature and magnitude of consequences can influence formation and maintenance of equivalence classes. The last paper by Vaidya discusses a review of the Stroop literature to ask if the main findings in that literature are interpretable in terms of stimulus equivalence and the notions of class cohesion and class conflict.

Basic knowledge about emergent relations

Eleven of 11 participants formed 2-node 4-member classes w using priming trials for training and testing. The ERPs were then measured during D-based baseline and derived relations. In Figure 1, the within- vs cross-class contrasts (W-C) generated activations 350-450 ms after target onset recorded from electrodes arrayed in a symmetrical U-shaped pattern above the borders of the left to right temporal, left and right parietal, and central occipital lobes. At 450 ms, activation ceased for the rest of the 700 ms measurement interval. The baseline vs derived relations contrasts (B-D) generated activation 400 ms after target onset, which remained active for the rest of the measurement interval. Activation was recorded from electrodes above the parietal and occipital regions describing a bilateral, symmetrical, M-shaped pattern with peaks in the medial parietal lobes. The time between 400-450 ms was the only one generating activity by both W-C and B-D contrasts at the same loci: right parietal and occipital regions. In that time, the W-C contrast also activated the lateral temporal and parietal regions, but the B-D contrast generate none. Finally, in Figure 2, the three islands of B-D activation between 450-700 ms might reflect the neural correlates of 1-3 node relations.

We developed a game to investigate equivalence class formation with remote data collection. The game simulates a labyrinth, with “sample rooms” with a picture (sample) on the only exit room, leading to “comparison rooms”, with pictures on three exit doors (comparisons). The player found diamonds after correct choices, and fell into a pit, losing diamonds, after incorrect choices. 56 adults played the game at home, with remote access to the experimenter’s computer. Group More-Reinforcement (MR) won four diamonds when correct and lost one when incorrect. Group Balanced (B) won one diamond when correct and lost one when incorrect. Group More-Punishment (MP) won one diamond when correct and lost four when incorrect. Participants learned conditional relations which could lead to three equivalence classes, each comprising three abstract pictures. Data collection ended after 10 participants formed equivalence in each group. Group MP showed higher attrition rate and poorer performance in a maintenance test, a week later. Results indicate that the game is a viable option to conduct remote data collection, and that the nature and magnitude of consequences can influence formation and maintenance of equivalence classes.

The original demonstration of the Stroop Effect found, in part, that participants’ reaction times to name a color (e.g., GREEN) were slower when the color of the ink and the name of the color were incongruent than when they were congruent. The robust effect has been documented numerous times across many laboratories since its original demonstration. Our lab has recently presented some data showing that compounds consisting of elements from different equivalence classes were reacted to more slowly than compounds consisting of elements from the same equivalence classes. These data suggest that the congruent and incongruent compound stimuli might be fruitfully interpreted in terms of class-based conflict or class-based cohesion. In this presentation, we present a brief review of the Stroop literature to ask if the main findings in that literature are interpretable in terms of stimulus equivalence and the notions of class cohesion and class conflict.