Human imagination is a complex system that allows us to form images or concepts in the mind that are not present to the senses. Research on imagination has been heavily influenced by the idea that humans store two distinct types of long-term memory: episodic and semantic memory. This theoretical distinction is particularly important in the context of aging, where older adults show reduced episodic memory compared to semantic memory (Levine et al., 2002). However, recent work has shown that these two memories are not as distinct as once thought (Renoult et al., 2019; Irish & Vatansever, 2020), suggesting a need to either refine the relationship between these concepts, or the concepts themselves.

Here, we apply a broader framework for imagination to the autobiographical memories of older adults. Introduced by Andrews-Hanna & Grilli (2021), memory and future thoughts can be understood as the outcome of the collaboration between two representational forms of imagination: the mind’s mind and the mind’s eye. The mind’s mind is described as a high-level, abstract form of imagination accompanied by a verbal representational form, and the mind’s eye is described as a contextually-specific, image-based form of imagination. In the present study, we examine whether this broader framework for understanding imaginative thought can a) explain some of the established age-related changes in episodic and semantic memory, and b) extend beyond existing research to offer new ways to conceptualize autobiographical memory in aging.

In this study, we introduce a novel scoring protocol distinguishing mind’s eye from mind’s mind forms of imagination and apply this protocol to the autobiographical memories of eighty-two cognitively normal older adults. Participants were instructed to retrieve unique autobiographical events, and to focus on describing event-specific details. All data were scored both with our new scoring protocol as well as the Autobiographical Interview scoring protocol from Levine et al. (2002).

Our novel scoring protocol demonstrated high inter-rater reliability across two raters for both mind’s mind (0.95) and mind’s eye (0.96) details. First, we show that the proportion of mind’s mind and mind’s eye details on average are significantly different, with an increased proportion of mind’s eye details. Second, we find that both mind’s eye detail production and mind’s mind detail production is significantly reduced with age, whereas only internal details decline across age when scored with the Autobiographical Interview scoring procedure.

The new scoring protocol suggests that both mind’s mind and mind’s eye details undergo change with age, a finding that shares similarities and differences with results from the Autobiographical Interview scoring technique. Taken together, our results hint at a more elaborate set of detail types forming autobiographical memories that change with age, with implications for understanding episodic and semantic memory.