Mapping Spatial Frequency Preferences Across Human Primary Visual Cortex

Abstract

Neurons in primate visual cortex (area V1) are tuned for spatial frequency, in a manner that depends on their position in the visual field. Several studies have examined this dependency using fMRI, reporting preferred spatial frequencies (tuning curve peaks) of V1 voxels as a function of eccentricity, but their results differ by as much as two octaves, presumably due to differences in stimuli, measurements, and analysis methodology. Here, we characterize spatial frequency tuning at a millimeter resolution within human primary visual cortex, across stimulus orientation and visual field locations. We measured fMRI responses to a novel set of stimuli, constructed as sinusoidal gratings in log-polar coordinates, which include circular, radial, and spiral geometries. For each individual stimulus, the local spatial frequency varies inversely with eccentricity, and for any given location in the visual field, the full set of stimuli span a broad range of spatial frequencies and orientations. Over the measured range of eccentricities, the preferred spatial frequency is well-fit by a function that varies as the inverse of the eccentricity plus a small constant. We also find small but systematic effects of local stimulus orientation, defined in both absolute coordinates and relative to visual field location. Specifically, peak spatial frequency is higher for tangential than radial orientations and for horizontal than vertical orientations.

ConvNets Develop Organizational Principles of the Visual Cortex when using Ganglion Cell-Based Sampling

“…Our findings indicate that the primate visual system may similarly exhibit receptive fields with constant effective spatial frequency tuning: Like in the RSL-CNN, in the primate visual system the receptive field size grows linearly as eccentricity increases (Dumoulin & Wandell, 2008; Smith et al, 2001), and the preferred spatial frequency grows linearly with eccentricity (Broderick et al, 2021; Henriksson et al, 2008; Sasaki et al, 2001). To translate our findings to primate visual systems, V1 cells with similar inherent orientation and spatial frequency selectivity should be present throughout the visual area, independent of the corresponding visual field location.…”

“…CNNs also lack a relationship between spatial frequency preference and eccentricity. Such a relationship has been observed in primate visual cortex, where small receptive fields exhibiting high spatial frequency preferences are concentrated foveally, whereas, with increasing eccentricity, receptive fields get progressively larger and tuned towards lower spatial frequencies (Broderick, Simoncelli, & Winawer, 2021;Dumoulin & Wandell, 2008;Sasaki et al, 2001;Smith, Singh, Williams, & Greenlee, 2001). Lastly, in contrast to CNNs, orientation tuning in the early visual cortex exhibits radial bias; the finding that radially oriented lines (and gratings) elicit a larger neuronal response than other orientations (Freeman, Brouwer, Heeger, & Merriam, 2011;Sasaki et al, 2006;Westheimer, 2003).…”

“…In the visual system, not only receptive field size is eccentricity-dependent but also spatial frequency preference (Broderick et al, 2021; Henriksson et al, 2008; Sasaki et al, 2001). We investigated whether such a relationship also emerges in the RSL-CNN but not the CNN by presenting sinring apertures revealing various spatial frequencies at a range of eccentricities to both neural networks (see Figure 9).…”

“…The RSL-CNN further exhibited eccentricity-dependent spatial frequency preferences. Like in primate visual systems (Broderick et al, 2021;Henriksson et al, 2008;Sasaki et al, 2001), the RSL-CNN exhibits a negative relationship between eccentricity and spatial frequency preference. This can be accounted for by higher spacing between retinal ganglion cells which results in a lower sampling rate at higher eccentricities, and hence, a compression of the input.…”

“…To translate our findings to primate visual systems, V1 cells with similar inherent orientation and spatial frequency selectivity should be present throughout the visual area, independent of the corresponding visual field location. This would imply that the empirically observed eccentricity-related tuning characteristics of visual cells (Broderick et al, 2021; Henriksson et al, 2008; Sasaki et al, 2001) are in fact not inherent tuning properties of the cells themselves. Rather, the observed differences in overall preferences may be caused by the non-linear retinal distortion of the visual field.…”