Abstract Major learning events are typically followed by a period during which the number and/or duration of rapid-eye movement sleep episodes is increased. Processes critical to memory formation are thought to take place during this interval of ‘enhanced’ rapid-eye movement sleep. We therefore compared the capacity for long-term potentiation during rapid-eye movement sleep and alert wakefulness after learning. Rats were chronically implanted with electrodes for stimulation of the perforant path and recording of evoked potentials and EEG in the dentate gyrus. After obtaining baseline recordings, rats were trained on a 40-trial two-way active avoidance task. Conditioned rats exhibited a two-fold increase in the mean duration of rapid-eye movement sleep episodes, as reflected by a prolongation of the hippocampal theta rhythm. There was no change in the sleep pattern of pseudoconditioned controls, which received explicitly unpaired tones and foot shocks in a yoked design. High-frequency stimulation was applied during the second, third, and fourth major rapid-eye movement sleep episodes after active avoidance training. Another group was tetanized at matching time points during alert wakefulness. After pseudoconditioning, tetanus applied during wakefulness or rapid-eye movement sleep readily induced long-term potentiation, and there was no difference between groups in the magnitude of increase for the population excitatory postsynaptic potential slope or the population spike height as measured 1 h, 24 h, and 5 days post-tetanus. By contrast, in conditioned rats, tetanus applied during wakefulness failed to elicit long-term potentiation of the excitatory postsynaptic potential slope (7.6% increase 1 h posttetanus), while the group stimulated during ‘enhanced’ rapid-eye movement sleep exhibited a 32% potentiation, equivalent to that obtained after pseudoconditioning. Analysis of individual experiments revealed a marked reduction in the probability of inducing long-term potentiation during post-learning wakefulness. We conclude that avoidance learning affects the induction of long-term potentiation in the dentate gyrus, suppressing induction during alert wakefulness while releasing the potential for synaptic modification during episodes of rapid-eye movement sleep. The effect is compatible with the hypothesis that memory consolidation involves a dynamic regulation of events such as long-term potentiation during sleep and wakefulness.