Abstract
Prior work has revealed that spatial memory is supported by two distinct systems: one in the striatum for encoding locations relative to landmarks, and the other in the hippocampus for encoding locations relative to environmental boundaries. Although abundant research supports this distinction in adult animals and humans, less is known about its development–whether its foundations can be detected early in typical development, whether it undergoes significant developmental change, and whether the two systems show differential impairment in cases of atypical development. Here we examined landmark- and boundary-based memory in typically developing (TD) children (6, 8, and 10 y/o), TD adults, and adults with Williams syndrome (WS)–a rare developmental disorder with hippocampal abnormalities. Participants performed a virtual navigation task that required them to learn the locations of objects within an arena comprising a landmark, a circular boundary, and distal cues (mountains) for orientation. The objects were tethered to either the landmark or boundary, and the relative position of the landmark and boundary was intermittently changed, allowing learning of landmark and boundary object location to be independently assessed. In typical development, we found increasing performance with age for both landmark-tethered and boundary-tethered objects; however, landmark-tethered object performance was similar to adults by 10 y/o, while boundary-tethered object performance in 10 y/o's was still below that of adults. In atypical development, we found that WS adults performed remarkably similar to TD adults on landmark-tethered objects, but significantly worse on boundary-tethered objects—in fact, similar to 6 y/o's. Taken together, these findings demonstrate that the cognitive systems mediating landmark- and boundary-based spatial memory i) develop under different trajectories, with the boundary-based system lagging behind the landmark-based system, and ii) are differentially susceptible to breakdown, with the boundary-based system more vulnerable than the landmark-based system, providing causal evidence for this distinction in humans.
Meeting abstract presented at VSS 2016