Abstract
Neuroimaging studies have revealed strong selectivity for scenes in the parahippocampal place area (PPA). However, the mechanism of this selectivity remains unknown. One possibility is that scene-selectivity reflects tuning for particular low-level stimulus features characteristic of scenes but not other non-preferred categories. Supporting this view, Nasr and colleagues (2014) recently observed that stimuli that were shown to strongly activate the PPA in previous studies tended to contain more rectilinear edges than non-preferred stimuli. Moreover, they demonstrated that PPA response is modulated by rectilinearlity for a range of non-scene images. Motivated by these results, we tested whether rectilinearity suffices to completely explain PPA selectivity to scenes. We scanned participants with fMRI while they viewed grayscale photographs of natural scenes and faces matched on rectilinearity, presented in a block design. To equate scene and face rectilinearity, 940 scene and face images were ranked according to their relative rectilinearity, and stimuli (52 unique images per category) were then drawn from this set so that average rectilinearity was matched across categories. Faces and scene categories were also matched on a number of other low-level image properties, including mean high spatial frequency content, luminance, and contrast. Despite matched rectilinearity, we found that the PPA – as well as the other scene-selective regions, the retrosplenial complex (RSC) and occipital place area (OPA) – still exhibited a significantly higher response to scenes than faces. Thus, the preference for low-level rectilinear edges does not suffice to completely explain selectivity for scenes in the PPA.
Meeting abstract presented at VSS 2015