Pavlovian-to-instrumental transfer: paradoxical effects of the Pavlovian relationship explained.

Four experiments with rats examined the origin of outcome-selective Pavlovian-to-instrumental transfer (PIT). Experiment 1 used a standard procedure, where outcomes were embedded within extended conditioned stimuli (CSs), to demonstrate the basic effect: Pavlovian stimuli augmented instrumental lever presses that had been paired with the same outcomes. Experiments 2 and 3 showed that after instrumental conditioning, whereas a conditioned stimulus (CS) trained using a backward conditioning procedure produced outcome-selective PIT, forward conditioning with a CS did not. These results are consistent with the idea that backward conditioning results in the outcome provoking its associated instrumental response during the CS and thereby allows a stimulus-response association to be acquired that directly generates outcome-selective PIT at test. Experiment 4 provided direct support for the assumptions that underlie this stimulus-response analysis. These results, and other paradoxical effects of the Pavlovian relationship, are incongruent with accounts of outcome-selective PIT that rely on a stimulus-outcome-response chain.