Abstract
There is a rift in the human scene processing literature. While several neuroimaging studies have argued that the parahippocampal place area (PPA) is involved in landmark recognition (i.e., recognizing a particular place or stable object in the environment), evidence from many other studies suggests otherwise. Based on results from the latter studies, we propose that PPA is not well suited to recognize landmarks in the environment, but rather is involved in recognizing the category membership of scenes (e.g., recognizing a scene as a coffee shop). We used fMRI multi-voxel pattern analysis to test this hypothesis. We scanned participants after they learned the layout of a virtual town that consisted of a park square surrounded by eight buildings. There were two buildings on each corner of the town. Each building belonged to a particular category: two coffee shops, two hardware stores, two gyms, and two dentist offices. Importantly, the locations of any two buildings belonging to the same category were dissociable from the category information (e.g., one gym was in the northeast corner of the town, while the other was in the southwest corner). If PPA represents landmark information, then it must be able to discriminate between two places of the same category, but in different locations of town. By contrast, if PPA represents general category information, then it will not represent the location of a particular place, but only the category of the place. As predicted, we found that PPA represents two buildings from the same category, but in different locations, as more similar than two buildings from different categories, but in the same location, while another scene-selective region of cortex, the retrosplenial complex (RSC), showed the opposite pattern of results. Such a double dissociation suggests distinct neural systems selectively involved in navigation and categorization of scenes.
Meeting abstract presented at VSS 2018