Second-order Conditioning with and Without Unconditioned Stimulus Presentation

The effects of presenting various episodes after serial presentation of two conditioned stimuli (CS2-CS1 sequences) on second-order conditioning to CS2 were examined in three experiments using rat subjects in an appetitive conditioning situation. In Experiment 1, presentation of food unconditioned stimuli (USs) immediately after CS2-CS1 sequences interfered with second-order conditioning of CS2. In Experiment 2, postsequence presentation of a "surprising" US interfered with second-order conditioning more than did presentation of an "expected" US; similarly, less second-order conditioning of CS2 was observed when postsequence nonpresentation of a US was surprising than when US omission was expected. In Experiment 3, the interfering effect of US presentation on second-order conditioning was smaller when a brief delay was introduced between presentation of the CS2-CS-1 sequence and the US. The results are discussed in terms of an information-processing theory recently proposed by Wagner and his colleagues.