Local opposite orientation preferences in V1: fMRI sensitivity to fine-grained pattern information

Abstract

The orientation of a visual grating can be decoded from human primary visual cortex (V1) using functional magnetic resonance imaging (fMRI) at conventional resolutions (2–3 mm voxel width, 3T scanner). It is unclear to what extent this information originates from different spatial scales of neuronal selectivity, ranging from orientation columns to global areal maps. According to the global-areal-map account, fMRI orientation decoding relies exclusively on fMRI voxels in V1 exhibiting a radial or vertical preference. Here we show, by contrast, that 2-mm isotropic voxels in a small patch of V1 within a quarterfield representation exhibit reliable opposite selectivities. Sets of voxels with opposite selectivities are locally intermingled and each set can support orientation decoding. This indicates that global areal maps cannot fully account for orientation information in fMRI and demonstrates that fMRI also reflects fine-grained patterns of neuronal selectivity.

Stimulus vignetting and orientation selectivity in human visual cortex

“…In fact, decoding accuracy was unaffected (statistically indistinguishable) by shifting the slices. However, the finding that a coarse-scale bias is the source of orientation decoding remains controversial, and several recent studies have attempted to disprove it ( Alink et al, 2013 ; Pratte et al, 2016 ; Alink et al, 2017 ), in part, we believe, because the notion that fMRI is sensitive to fine-scale neural activity is highly attractive. We have no doubt that there are orientation columns in human V1.…”

“…However, in some cases, other stimulus vignetting exists, resulting in more subtle coarse-scale patterns of orientation bias ( Swisher et al, 2010 ; Larsson et al, 2017 ; Sengupta et al, 2017 ). One study ( Alink et al, 2017 ) used inner and outer circular annuli, but added additional angular edges, the result of which should be a combination of radial and tangential biases. Indeed, this study reported that voxels had a mixed pattern of selectivity, with a considerable number of voxels reliably preferring tangential gratings, and other voxels reliably favoring radial orientations.…”

“…On the other hand, we have argued that the coarse-scale orientation bias is the predominant orientation-selective signal measured with fMRI, and that multivariate decoding analysis methods are successful because of it ( Freeman et al, 2011 , 2013 ). The relative contribution of coarse- and fine-scale biases remains hotly debated ( Pratte et al, 2016 ; Alink et al, 2017 ).…”

Multivoxel Pattern of Blood Oxygen Level Dependent Activity can be sensitive to stimulus specific fine scale responses

“…As such, the findings reported here provide intuitions into the sensitivity of multivoxel decoding to high and low spatial frequency information when the input stimuli contain both low and high spatial frequencies, which is not guaranteed to be the same as when low or high spatial frequency stimuli are presented in isolation. A recent study by Alink et al 36 implemented an analysis strategy comparable to the one adopted here to assess the spatial scale of information exploited by MVPA during orientation decoding in V1. They spatially offset the testing relative to the training set by 1, 2, 4, and 6 mm and measured the impact on decoding accuracy.…”

“…We then tested the performance of 3 classifiers: linear SVM; Linear Discriminant Analysis (LDA); and Naïve Bayes Classifier (NBC) on real data. It should be noted that comparable, albeit slightly different approaches have previously been implemented to assess the scale of information exploited by MVPA during orientation decoding in V1 at 3 T. Alink et al 36 , for example, implemented an analysis strategy comparable to the one proposed here. They spatially offset the testing relative to the training set by 1, 2, 4, and 6 mm and measured the impact on decoding.…”