Abstract
Boundary cues crucially contribute to the spatial layout of an environment. Developmental research highlights the importance of vertically extended structure in defining the geometry of space: four-year-old children fail to reorient themselves in an arena marked by a flat black mat, whereas they can do so by a 3D curb that is only 2 cm in height (Lee & Spelke, 2011). Consistent with this finding, a recent neuroimaging study demonstrated that the parahippocampal place area (PPA) is sensitive to the presence of 3D vertical boundary structure, even if this structure is quite minimal (a "curb" boundary) (Ferrara & Park, 2014). Here, we ask whether PPA's sensitivity to the vertical height of spatial boundary cues would be affected by variation in scene geometry, specifically when manipulated by varying the amount of angular (rectilinear) information. In the current study, we used artificially created scenes that varied along 1) scene geometry (Triangle, Rectangle, Hexagon, Circle), and 2) height of boundary cue (Mat (2D boundary), Curb (~3 inches high), Wall (typical indoor space), see Supplemental Figure). Participants (N=5) viewed these conditions in blocks of 12 seconds while performing a one-back repetition task. We measured the univariate response of scene-selective regions. Consistent with the findings of Ferrara & Park (2014), right PPA shows differences in activity between the Mat and Curb across all scene geometry conditions. Moreover, neither a main effect of scene geometry nor an interaction between that and boundary height were found. These findings suggest that PPA's sensitivity to the minimal vertical structure of the Curb extends across environments of many different spatial geometries. Thus, its sensitivity to minimal vertical boundary structure is not modulated by the amount of rectilinear information that is included in a boundary. Results of extended scene-selective cortical areas (RSC and TOS) will be also discussed.
Meeting abstract presented at VSS 2016