Schedule of Events | Search Abstracts | Invited Symposia | Symposia | Poster Sessions | Data Blitz

Using Unitization to Rescue Associative Memory

Poster Session F - Tuesday, March 10, 2026, 8:00 – 10:00 am PDT, Fairview/Kitsilano Ballroom

Catherine Carpenter¹ (), Andrew Yonelinas¹; ¹The University of California, Davis

Associative memory deficits are a common and debilitating aspect of healthy aging. While memory for individual items tends to remain relatively intact, older adults often struggle with remembering associations between items— like linking a face to a name. Research indicates that these age-related associative impairments are largely due to atrophy in the hippocampus, whereas preserved item memory is thought to rely more on the perirhinal cortex, a region less affected by aging. A growing body of evidence suggests that these deficits can be mitigated through ‘unitization’ — a process in which elements of an association are encoded as a single, cohesive unit. When unitization is successful, associative memory can be supported by item memory systems that are preserved in aging. Despite promising findings, the conditions under which unitization is most effective in rescuing associative memory in aging remain poorly understood. Identifying the factors that determine the effectiveness of unitization is critical for developing evidence-based interventions to improve associative memory in aging. The current experiments test a theory that unitized information may need to be novel to see the full benefit of familiarity-based processing that unitization can offer. Receiver operating characteristics are used to estimate the influence of familiarity and recollection for pre-experimental and experimental unitized information. The results lend initial support to the importance of novelty in familiarity-based unitized processing and will inform future work in narrowing down the neural substrates of this process in aging and in optimizing the effectiveness of this memory support.

Topic Area: LONG-TERM MEMORY: Episodic