Unblocking in Pavlovian appetitive conditioning.

Two experiments demonstrated unblocking in an appetitive conditioning preparation with rat subjects. One stimulus, A, was first paired with either a low-value reinforcer (one food pellet) or a high-value reinforcer (one food pellet followed by two more food pellets). A second stimulus, X, was then added to A, and the compound was reinforced with either the high- or low-value reinforcer. Conditioning to X was blocked if the same reinforcer was used in both phases of the experiment, but there was substantial conditioning to X when the reinforcer value was shifted either up or down when X was introduced. Experiment 1 demonstrated this unblocking phenomenon using a design that minimized the potential contribution of generalization decrement. Experiment 2 examined the effects of a variety of posttraining manipulations on conditioned responding to the added X cue after unblocking procedures. Responding of subjects that received downshifts in reinforcer value when X was introduced was affected by several posttraining manipulations, including changes in context value. Those manipulations had smaller effects on the responding of subjects that received upshifts in reinforcer value and no effects on responding in control conditions.