This study investigated developmental differences in aversive conditioning, extinction, and reinstatement (i.e., the recovery of conditioned aversive associations following reexposure to the unconditioned stimulus [US] post-extinction). This study examined these mechanisms in children (Mage = 8.8 years), adolescents (Mage = 16.1 years), and adults (Mage = 32.3 years) using differential aversive conditioning with a geometric shape conditional stimulus (CS+) paired with an aversive sound US and another shape (CS−) presented alone. Following an extinction phase in which both CSs were presented alone, half of the participants in each age group received three US exposures (reinstatement condition) and the other half did not (control condition), followed by all participants completing an extinction retest phase on the same day. Findings indicated (a) significant differences in generalizing aversive expectancies to safe stimuli during conditioning and extinction that persisted during retest in children relative to adults and adolescents, (b) significantly less positive CS reevaluations during extinction that persisted during retest in adolescents relative to adults and children, and (c) reinstatement of US expectancies to the CS+ relative to the CS− in all age groups. Results suggest important differences in stimulus safety learning in children and stimulus valence reevaluation in adolescents relative to adults.