Abstract Although it has been reported repeatedly that retrieval-related processes decline with aging, the influence of well-documented age-related encoding deficiencies on the observed changes at retrieval have not been ruled out as a contributing factor. Here, we disentangle this confound by using a serendipitous finding reported by Nessler et al. [D. Nessler, R. Johnson Jr., M. Bersick, D. Friedman, On why the elderly have normal semantic retrieval but deficient episodic encoding: a study of left inferior frontal ERP activity, Neuroimage 30 (2006) 299–312]. In that study, age-related differences in the magnitude of left inferior frontal brain activity at encoding and subsequent recognition memory performance were eliminated when a deeper level of semantic encoding in the older adults was compared with a shallow level in the young. Based on this earlier result, the present paper is concerned with the question of whether the matched recognition performance resulting from age-equivalent ERP encoding activity was also accompanied by age-invariant retrieval-related brain activity. The results in the young were consistent with dual-process models of recognition memory due to the presence of ERP activity linked previously to familiarity (frontal EM effect) and recollection (parietal EM effect). By contrast the older adults only showed evidence of familiarity-based processes. Thus, despite age-equivalent brain activity at encoding and subsequent recognition performance, older relative to young adults appeared to base their old–new decisions on a qualitatively different pattern of retrieval processes (i.e., more on familiarity and less on recollection). Consequently, these data suggest that the age-related changes in retrieval observed here are independent of, and likely occur in addition to, any age-related changes in encoding processes.