Abstract
Successful navigation depends on one’s ability to use different visuospatial cues both to maintain a sense of orientation and to recover from disorientation. Although researchers debate the extent to which organisms integrate geometry (e.g., distance/length) with features (e.g., color), there is general consensus that even human children use geometry when reorienting in an enclosed three-dimensional environment (Cheng & Newcombe, 2005). Yet questions remain about which geometric properties support navigational processes such as reorientation. One recent proposal is that geometric properties are neurally dissociable (Spelke, Lee, & Izard, 2010) such that children reorient by relying on distance (i.e., spatial extent from the self to a boundary), not properties of the space itself (e.g., relative wall length; Lee, Sovrano, & Spelke, 2012). However, existing work confounds individual geometric properties (distance/length) with the global shape of the surrounding space. Here we address the impact of this confound. As in previous research, we used fragmented spaces to isolate distance and length. In Experiment 1, distance was available in a rectangular layout whereas length was available in a square layout. Children (Mage = 3.7 years) were better at reorienting using distance than length (p < .05) and were only above chance at localizing a target when using distance (p < .05), replicating Lee and colleagues. In Experiment 2, we dissociated distance/length from global shape by adding appendages to the ends of the walls and preventing perceptual completion. Children (Mage = 3.9 years) were now above chance at using length (p < .05) to localize a target, with no difference between length and distance conditions (p > .3). These findings demonstrate that human children are capable of using different geometric properties for the purpose of reorientation and suggest that a distance advantage, when it exists, may be a consequence of the overall shape of the environment.
Meeting abstract presented at VSS 2015