The present research explored the effects of restraint stress on two situations involving incentive downshift: consummatory successive negative contrast (cSNC) and extinction of escape behavior in the Barnes maze. First, Experiment 1 confirmed that the restraint stress procedure used in these experiments increased levels of circulating corticosterone. Second, prior exposure to restraint stress enhanced the cSNC effect whether stress was administered before the first downshift trial (Experiment 2) or before the second downshift trial (Experiment 3). In none of these experiments did restraint stress affect the consummatory behavior of unshifted controls. In Experiment 4, animals received training to escape into a target hole in the Barnes maze and were then exposed to eight extinction trials in which the escape box was absent. Restraint stress before extinction did not affect the latency to reach the target hole, but it increased the distance traveled and approach to nontarget holes. In Experiment 5, restraint stress before a post-extinction test a day later reduced spontaneous recovery in approach to the goal hole without affecting exploratory behavior. The results were interpreted in terms of the aversive summation hypothesis according to which two sources of stress (i.e., restraint and incentive downshift) can affect behavior and enhance the retrieval of aversive memory. (Contains 7 figures.)