Abstract
In complex real-world scenes, image content is conveyed by a large collection of intertwined visual features. The visual system disentangles these features in order to extract information about image content. Here, we investigate the role of one integral component: the content of spatial frequencies in an image. Specifically, we measure the amount of image content carried by low versus high spatial frequencies for the representation of real-world scenes in scene-selective regions of human visual cortex. To this end, we attempted to decode scene categories from the brain activity patterns of participants viewing scene images that contained the full spatial frequency spectrum, only low spatial frequencies, or only high spatial frequencies, all carefully controlled for contrast and luminance. Contrary to the findings from numerous behavioral studies and computational models that have highlighted how low spatial frequencies preferentially encode image content, decoding of scene categories from the scene-selective brain regions, including the parahippocampal place area (PPA), was significantly more accurate for high than low spatial frequency images. In fact, decoding accuracy was just as high for high spatial frequency images as for images containing the full spatial frequency spectrum in scene-selective areas PPA, RSC, OPA and object selective area LOC. We also found an interesting dissociation between the posterior and anterior subdivisions of PPA: categories were decodable from both high and low spatial frequency scenes in posterior PPA but only from high spatial frequency scenes in anterior PPA; and spatial frequency was explicitly decodable from posterior but not anterior PPA. Our results are consistent with recent findings that line drawings, which consist almost entirely of high spatial frequencies, elicit a neural representation of scene categories that is equivalent to that of full-spectrum color photographs. Collectively, these findings demonstrate the importance of high spatial frequencies for conveying the content of complex real-world scenes.
Highlights
Everything that our visual system processes begins as collections of low-level features, such as contrasts, colors, orientations, and spatial frequencies
It has previously been shown that scene-selective areas of cortex (PPA, retrosplenial cortex (RSC) and occipital place area (OPA)) show greater relative levels of activation in response to places and buildings compared with faces and objects [29,30,31,32], but that the distributed patterns of activity elicited by different types of scenes contain scene content information, such that an image’s specific scene category can be decoded from these regions [17, 26]
It was possible to decode the categories of the high spatial frequency (HSF) scenes above chance for all of the tested scene-selective Regions of interest (ROI) (PPA: t(9) = 2.68, p = 0.013, d = 0.85; RSC: t(9) = 4.67, p < 0.001, d = 1.48; OPA: t(9) = 3.31, p = 0.005, d = 1.05), as well as object-selective area lateral occipital cortex (LOC) (t(9) = 4.40, p < 0.001, d = 1.39)
Summary
Everything that our visual system processes begins as collections of low-level features, such as contrasts, colors, orientations, and spatial frequencies. Watson et al (2016) [10] reported that the neural activity of a large visual ROI that included scene-selective regions could discriminate between high and low spatial frequencies in an image better than between natural and man-made image content. Behavioral studies have shown that when viewing a scene, low spatial frequency content of the image is processed prior to higher spatial frequencies [11] This phenomenon is referred to as the coarse-to-fine hypothesis of visual perception [12]. To address the unresolved question “How much scene content is conveyed by different spatial frequencies in scene-selective visual cortex?” we presented participants with images of frequency-filtered scenes and compared how well scene categories could be decoded from fMRI activation patterns in each of the frequency conditions. Even though the filtered noise stimuli do not activate scene-selective brain regions to the extent that scenes do, they serve as an important control for explicit neural representations of spatial frequencies
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
