Granularity of hippocampal long-axis representations with repeated encoding
Sam Audrain1, Jude Baffoe-Bonnie1, Jenna M. Wilson1, Alex Martin1; 1NIMH
Repeated exposure to an experience can either strengthen the neural representation of that unique experience or integrate it with overlapping memories to the detriment of memory specificity. The hippocampus is thought to play a critical role in maintaining unique mnemonic representations that can be differentiated from similar experiences. However, it is unclear how repeated encoding affects representation of overlapping experiences along the long-axis, which differs in representational granularity. We investigated how the hippocampus represents granular content along its long-axis with repeated encoding. We scanned healthy young adults with 7 Tesla fMRI as they repeatedly encoded unique objects paired with one of four scenes: two visually similar beaches, and two visually similar kitchens. At retrieval, they were presented with each object and asked to retrieve which specific scene it had been paired with. We examined pattern similarity across the long-axis of the hippocampus as a function of repeated encoding and degree of content overlap between object-scene pairs. We found that patterns in the posterior hippocampus changed more than in the anterior hippocampus with repeated encoding, becoming less similar to other memories with each exposure regardless of how much scene content overlapped. Across the long axis, while individual trials maintained representational specificity with repetition (e.g. apple-beach1 vs apple-beach1), they became differentiated from overlapping trials (e.g. apple-beach1 vs pylon-beach1). Thus, we found evidence of both content-agnostic and content-specific changes in pattern similarity with repetition, suggesting multiple scales of representational change within the hippocampus with experience that serves to preserve unique, granular memory traces.
Topic Area: LONG-TERM MEMORY: Episodic
CNS Account Login
April 13–16 | 2024