Neurocognitive mechanisms of spatial pattern separation in older adults with and without subclinical memory impairment

Abstract

Background Decline in episodic memory is a hallmark feature of Alzheimer’s disease (AD). Episodic memories can be operationalized as detailed conjunctions of items and their spatiotemporal context. Our past work has uncovered deficits in pattern separation of similar object memories in nondemented older adults. However, separation of similar spatial locations was only slightly impaired in this population. We hypothesized that these deficits may be more readily observable in individuals with subclinical memory impairment (assessed by verbal list recall). We also sought to determine the neural mechanisms by which spatial pattern separation deficits may manifest in older adults, focusing on the posterior parietal network which is involved in spatial processing and implicated in AD. Methods Two cohorts were included in this analysis. The first is a group of older adults with subclinical memory impairment (aged impaired – AI, age range: 61‐91; mean MMSE: 28.3 ± 1.50; mean RAVLT delayed recall: 9.66 ± 4.05). The second is a group of non‐demented older adults without memory impairment (aged unimpaired – AU; age range: 61‐86; mean MMSE: 28.1 ± 1.59; RAVLT delayed recall: 11.4 ± 3.01). Participants underwent whole brain high‐resolution resting state functional magnetic resonance imaging (1.5 x 1.5 x 2 mm resolution). They were also administered a battery of neuropsychological tests including tests of object and spatial pattern separation. We used performance on the RAVLT as the outcome predicted by performance on the discrimination tests. Specifically, we used delayed recall (DR), retroactive interference (RI), and percent forgetting (PF). Results Behaviorally, spatial, but not item, pattern separation performance predicted RAVLT DR, RI and PF scores in the AI group, but not the AU group. Functional connectivity analyses revealed that the connectivity of regions involved in spatial processing was significantly related to spatial pattern separation performance in the AI group, but not the AU group. Conclusion These results validate the neurobiological basis of spatial pattern separation and suggest that such tasks can be used to detect early stages of AD, which has important implications for future diagnostics and treatment approaches.