Anxiety-like behaviors emerge with repeated exposure to and short-term withdrawal from cocaine. The stress-related neuropeptide, corticotropin-releasing factor (CRF), has been implicated in the anxiogenic effects of cocaine withdrawal, as well as in some of the long-lasting effects of cocaine. One objective of the present experiments was to determine whether repeated exposures to cocaine, under conditions that induce anxiety in the initial withdrawal period, would induce longer-lasting anxiogenic responses. A second objective was to determine whether any such effects would be potentiated by CRF. In Experiment 1, animals were injected once daily for 7 days with cocaine (30 mg/kg, i.p.) or saline in the home cages and, after a 10-day drug-free period, were given an i.c.v. injection of CRF (0.5 or 5.0 micro g) or vehicle, followed by a 5-min test for anxiety in the elevated plus maze or light-dark transition apparatus. In Experiment 2, animals were given the cocaine or saline injections in a distinct environment. At test, they were placed in the distinct environment after the CRF (0.5 micro g) or vehicle injection and were subsequently tested for anxiety. Cocaine produced enhanced levels of anxiety when pre-exposures were given in a distinct environment, but not when they were given in the home cage. In neither case did cocaine differentially alter anxiety-like responses to CRF. The results suggest that a "reminder" of the drug experience, such as re-exposure to cocaine-paired contextual cues, may be necessary to induce elevated levels of anxiety after the initial withdrawal period. In addition, although the results do not rule out a role for endogenous CRF in lasting cocaine-induced anxiogenic responses, they suggest that an increased sensitivity of CRF receptors to the peptide is not responsible for the effect.