The present study investigates potential age differences in the self-reference effect. Young and older adults incidentally encoded adjectives by deciding whether the adjective described them, described another person (Experiments 1 & 2), was a trait they found desirable (Experiment 3), or was presented in upper case. Like young adults, older adults exhibited superior recognition for self-referenced items relative to the items encoded with the alternate orienting tasks, but self-referencing did not restore their memory to the level of young adults. Furthermore, the self-reference effect was more limited for older adults. Amount of cognitive resource influenced how much older adults benefit from self-referencing, and older adults appeared to extend the strategy less flexibly than young adults. Self-referencing improves older adults' memory, but its benefits are circumscribed despite the social and personally relevant nature of the task.