Search
Browse By Day
Browse By Time
Browse By Panel
Browse By Session Type
Browse By Topic Area
Search Tips
Register for SRCD21
Personal Schedule
Change Preferences / Time Zone
Sign In
X (Twitter)
Introduction. Remembering specific episodes requires the ability to retain the unique spatio-temporal features of events. Memory for item-space, item-time, and item-item exhibit distinct trajectories of behavioral change. Although contextual binding mechanisms are supported by the hippocampus, it is not yet established whether hippocampal subregions contribute to distinct spatio-temporal features that might explain these trajectories.
Hypotheses. We hypothesized that anterior hippocampus would respond more to fast-changing item-specific contextual features and the posterior hippocampus to slower-changing features of global context (e.g., whether the experiment took place, the overall temporal context of the experience).
Sample and Procedure. Participants included 8-year-olds (N=14), 10-year-olds (N=21), and young adults (N=23). Participants were scanned while they completed an associative recognition task in which they were asked to recognize whether pairs of items had been originally presented together. During the test, studied pairs either appeared in their initial location (same-location) or they appeared in the location of their paired item (flipped-location). The test also included pairs of completely novel items or rearranged pairs of old items presented either in their initial location (rearranged same-location) or flipped location (rearranged flipped-location) as new pair conditions. Functional magnetic resonance imaging data were collected during 4 consecutive runs of the task with different sorts of items.
Results. Behaviorally, we observed the expected age-related improvement in memory ( F_8,208=2.57,p=0.011), and 8-year-olds’ performance was reduced for flipped pairs ( p<0.001). We conducted Representational Similarity Analysis in the hippocampus and found a significant 3-way interaction between age, trial, and sub-region (F_12,330=1.81,p=0.046), such that in the adult anterior hippocampus, same-location old pairs and flipped-location old pairs exhibited significantly greater similarity than same-location old pairs and novel pairs (p=0.003; Fig.1). This was not the case in the two children groups. This finding suggests that the anterior hippocampus increasingly respond to salient, item-level changes from one trial to the next. In contrast, presentational similarity was not different across these trials in the posterior hippocampus, but the level of similarity was overall greater in adults compared to 10-year-olds, which in turn was greater than that in 8-year-olds (ps<.05;Figure 1 ). The pattern of representational similarity across trials in the posterior hippocampus suggests that this region responds to high-overlap, slow-changing contextual features, which help identify event boundaries in episodic memories. To gain evidence for this role, we compared representational similarity between different scanning runs as a way to indicate slow changes in temporal context. In the adult posterior hippocampus, we found higher degree of representational similarity between first and second runs and between first and fourth runs compared to first and third run only (ps<.001, Figure 2). This pattern was not found in children. This result suggests that the posterior hippocampus may contribute to encoding of slow changing features.
Conclusions. Overall, these results provide support for the idea that the hippocampus is functionally heterogeneous along the anterior and posterior axis. Our discussion will focus on whether these differences may reflect degrees of resolution of the contextual binding representation during development.