Abstract
Visuospatial difficulties in Williams syndrome (WS) are well documented. Recently, research has shown that spatial difficulties in WS extend to large-scale space, particularly in coding space using an allocentric frame of reference. Typically developing (TD) children and adults predominantly rely on the use of a sequential egocentric strategy to navigate a large-scale route (retracing a sequence of left–right body turns). The aim of this study was to examine whether individuals with WS are able to employ a sequential egocentric strategy to guide learning and the retracing of a route. Forty-eight TD children, aged 5, 7, and 9 years and 18 participants with WS were examined on their ability to learn and retrace routes in two (6-turn) virtual environment mazes (with and without landmarks). The ability to successfully retrace a route following the removal of landmarks (use of sequential egocentric coding) was also examined. Although in line with TD 5-year-olds when learning a route with landmarks, individuals with WS showed significantly greater detriment when these landmarks were removed, relative to all TD groups. Moreover, the WS group made significantly more errors than all TD groups when learning a route that never contained landmarks. On a perceptual view-matching task, results revealed a high level of performance across groups, indicative of an ability to use this visual information to potentially aid navigation. These findings suggest that individuals with WS rely on landmarks to a greater extent than TD children, both for learning a route and for retracing a recently learned route. TD children, but not individuals with WS, were able to fall back on the use of a sequential egocentric strategy to navigate when landmarks were not present. Only TD children therefore coded sequential route information simultaneously with landmark information. The results are discussed in relation to known atypical cortical development and perceptual-matching abilities in WS.