Abstract
Extinction of a response in the presence of one stimulus in an equivalence class can transfer to other related stimuli, but difficulties in establishing extinction can compromise analyses. The present study evaluated the transfer of avoidance extinction with two extinction procedures. In particular, avoidance or nonavoidance was always (Experiment 1) and never (Experiment 2) followed by point loss in the crucial extinction test phase. Both experiments began with the establishment of two equivalence classes with four abstract figures in each (A1-B1-C1-D1 and A2-B2-C2-D2). Clicking a button to avoid loss of points was trained with B1 and subsequently observed without direct training in the presence of C1 and D1. Extinction was then conducted with one group of participants with stimuli that underwent avoidance training (direct extinction with B1 and B2) and with another group with stimuli who did not undergo avoidance training (derived extinction with C1 and C2). Finally, the transfer of extinction was evaluated with stimuli from both classes. In Experiment 1, 10 of 14 participants met the avoidance extinction criteria, and the transfer of extinction occurred for 2 (1 in the direct and 1 in the derived extinction group). In Experiment 2, 10 of 13 participants met the avoidance extinction criteria, and the transfer of extinction occurred for 6 (5 in the direct and 1 in the derived extinction group). Overall, the transfer of extinction occurred only with the combination of an extinction procedure without aversive events and direct extinction.