Associative learning can be observed from the neonatal period onward, providing opportunities to examine changes in basic learning and memory abilities. One method that is suitable to study associative learning is classical eyeblink conditioning (EBC) which is dependent on the cerebellum. Extinction learning can be systematically investigated in this paradigm by varying the context during learning and extinction. Because of methodological difficulties and ethical challenges, no studies have compared extinction learning using EBC across human development. Our goal was to test feasibility of a 3-day delay EBC paradigm that can be used from infancy to adulthood. Acceptance/safety was tested especially for infancy by investigating attrition rates and parental report on infant wellbeing. On a paradigm side, we tested if the paradigm leads to successful acquisition and extinction. An air puff served as unconditional stimulus (US) and a tone as conditional stimulus (CS). On day 1 during acquisition, participants received 36 US–CS pairings in context A. On day 2, participants received 12 acquisition trials in context A to consolidate association learning, followed by 48 extinction trials (tone alone presentations) in context B. Renewal was assessed on day 3 and incorporated 12 CS alone trials presented in both the acquisition context and the extinction context. Eyeblink responses were videotaped and coded offline. The protocol was tested with 12–36-months-old infants (N = 72), adolescents (N = 8), and adults (N = 8). Concerning the acceptance/safety side, attrition ranged from 21 to 58% in infant samples due to the complex preparation of the children for the paradigm. However, attrition is equal to or lower than other infant learning paradigms. Parents of infant samples were very interested in the paradigm and reported low levels of infant stress, exhaustion, and negative feelings during the sessions. Data quality was very high, and no participant had to be excluded because of insufficient data. Concerning the paradigm side, participants showed successful acquisition and extinction as a group. The procedure is ethically sound, feasible, tolerated by many infants, and acceptable among parents. The data show successful acquisition and extinction rates, making the paradigm a valuable tool for investigating developmental changes in extinction learning over the lifespan.