Three experiments used an autoshaping procedure in pigeons to investigate the conditioning of the context and of a discrete conditioned stimulus (CS) with a food unconditioned stimulus (US). CS-US associations were measured by directed pecking at the key light CS; context-US associations were assessed by general activity in the context. Experiment 1 investigated the influence of context-US associations on performance to a previously trained CS. The same CS produced greater key-pecking in a context of higher associative strength. Experiment 2 examined the influence of context-US associations on learning of CS-US associations. When tested in a context of fixed associative strength, a CS that had been trained in a context of high associative strength elicited less responding than one trained in a context of low associative strength. Experiment 3 found that signaling a US by a discrete CS interfered with the formation of context-US associations, as measured both in terms of general activity and ability to promote responding to another CS. These results suggest that the context and the CS compete for association with the US. They also suggest that context-US associations facilitate the exhibition of CS-US associations.
Return to reprint request page