Purchase this article with an account.
Emily J. Ward, Whitney E. Parker, Alana M. Feiler, Russell A. Epstein; Adaptation for individual places but not for place categories in scene-selective cortical regions. Journal of Vision 2008;8(6):333. doi: https://doi.org/10.1167/8.6.333.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
An image of a real-world location can be identified in at least two ways: as a specific place in the world (“VanPelt Library”), or an exemplar of a more general place category (“library”). Previous fMRI studies (Epstein and Higgins 2007) indicate that scene-selective cortical regions such as the parahippocampal place area (PPA) and retrosplenial cortex (RSC) are sensitive to this distinction, responding much more strongly when subjects identify scenes as specific places than when they classify them into place categories. Although these results may indicate that the PPA/RSC are uninvolved in scene categorization, an alternative possibility is that information about scene category is encoded but using a submaximal response. Here we use fMRI adaptation to resolve this issue. In different blocks, subjects viewed images depicting: (1) prominent landmarks from a familiar college campus, (2) prominent landmarks from an unfamiliar college campus, (3) indoor scenes of various categories (e.g. kitchen, bedroom), (4) outdoor scenes of various categories (e.g. playground, beach). Within each block, the category or campus landmark was either fixed (different images of the same category/landmark) or varied (different categories/landmarks). Adaptation was indicated by reduced response in the fixed condition compared to the varied condition for each of the 4 stimulus classes. In RSC, adaptation was observed for familiar landmarks but not unfamiliar landmarks, indoor categories, or outdoor categories. No adaptation was observed in the PPA for any stimulus class. These results support the contention that the PPA/RSC encode representations useful for identification of specific places: a viewpoint-specific scene snapshot in the PPA, and a more viewpoint-invariant representation of familiar places in RSC. In contrast, neither PPA nor RSC appear to encode information about scene/place category.
This PDF is available to Subscribers Only