Reversibility of single-incentive selective associations

Blocking a selective association in pigeons

Experiment 1 demonstrated for the first time a stimulus-reinforcer interaction in pigeons trained with free-operant multiple schedules of reinforcement. Pigeons that treadle pressed in the presence of a tone-light (TL) compound for food exhibited primarily visual stimulus control on a stimulus-element test, whereas pigeons that avoided shock in TL exhibited auditory control. In Experiment 2, this selective association was blocked in pigeons pretrained with the biologically contingency-disadvantage element of the compound (i.e., tone-food or light-shock) before TL training. When this pretraining preceded compound-stimulus training, control was now auditory in pigeons that treadle pressed for food and was visual in pigeons that avoided shock. Previous attempts at blocking this selective association were unsuccessful in pigeons (LoLordo, Jacobs, & Foree, 1982) but were successful in rats (Schindler & Weiss, 1985). Experiment 2 established that selective associations can be blocked in pigeons when the procedures that were effective with rats were systematically replicated. These results further demonstrate the cross-species generality of an associative attentional mechanism involving a biological constraint on learning in species with different dominant sensory systems.