Many people suffer from various types of anxiety disorders, which are characterized by intense fear. Others who suffer from affective disorders such as melancholic depression also display intense fear and is associated with centrally mediated pituitary-adrenal activation.' These disorders may represent dysfunctional fear learning and/or fear regulation. Kalin2 has also shown in young monkeys that a correlation exists between the glucocorticoid corticosterone and fear reactions (freezing behavior). Although many studies have demonstrated an important link between glucocorticoids and stress and CRF and fear, very little is known about the role of glucocorticoids in mediating fear. The glucocorticoid, corticosterone (CORT), selectively upregulates CRF mRNA in the central nucleus of the amygdala (Ce) in rats.3 This region plays a crucial role in the learning and expression of conditioned fear reaction^.^ In the present study, we investigated whether CORT would potentiate fear responses (freezing) conditioned to acoustic stimuli. Twenty adult male Sprague-Dawley rats (250-275 g) were acclimated to the vivarium for approximately 7-10 d before conditioning began. All animals received 10 conditioning trials on one day during which the unconditioned stimulus (US; 1.0 mA footshock) was presented concurrently with the last 500 ms of the 20 s conditioned stimulus (CS; 70 dB/10,000 Hz tone). The intertrial interval was 120 s on the average. Two days after conditioning, subcutaneous hormone pellets (Innovative Research of America, Toledo, OH) were implanted under ketamine (100 mg/kg) and rompun (5 mgkg) anesthesia and distributed in the following manner: 5 animals received a low dose of CORT (5.0 mg/day/rat); 5 received a high dose of CORT (10.0 mg/day/rat); 5 received a low dose of CORT placebo (5.0 mg/day/rat); and 5 animals received a high dose of CORT placebo (10.0 mg/day/rat). Hormone pellets were left in animals for the entire duration of the experiment. Freezing was first tested on the morning of the fifth day of hormone exposure (0 h test) in a novel context. While ambulating, animals were exposed to the CS continuously for 600 s and the amount of time they spent freezing was recorded. Freezing was measured at 0 h and 4 h on the fifth day of hormone exposure, then daily until extinction was complete. A two-way analysis of variance (ANOVA) with one grouping variable (hormone