Topographic organization of eye-position dependent gain fields in human visual cortex

Abstract

The ability to move has introduced animals with the problem of sensory ambiguity: the position of an external stimulus could change over time because the stimulus moved, or because the animal moved its receptors. This ambiguity can be resolved with a change in neural response gain as a function of receptor orientation. Here, we developed an encoding model to capture gain modulation of visual responses in high field (7 T) fMRI data. We characterized population eye-position dependent gain fields (pEGF). The information contained in the pEGFs allowed us to reconstruct eye positions over time across the visual hierarchy. We discovered a systematic distribution of pEGF centers: pEGF centers shift from contra- to ipsilateral following pRF eccentricity. Such a topographical organization suggests that signals beyond pure retinotopy are accessible early in the visual hierarchy, providing the potential to solve sensory ambiguity and optimize sensory processing information for functionally relevant behavior.

A retinotopic reference frame for space throughout human visual cortex

“…Indeed, only experiments in full darkness will suffice to fully rule out the misinterpretation of such visual signals originating from the periphery. Importantly, by employing static gaze positions throughout our experimental runs and decreasing the strength of peripheral visual stimulation, we explicitly prevent such spurious signals from producing false spatiotopic patterns in our data 7 . We conceptualized experiments and analyses to be able to unequivocally adjudicate between retinotopic and spatiotopic reference frames.…”

“…Foundationally, visual locations are encoded in a common “retinotopic” reference frame 4,5 , keeping the retinal photoreceptors organization. Neural responses in this retinotopic reference frame are modulated by gaze direction 6,7 . This combination of action-related and visual sources of information is thought to allow the brain to localize objects in the outside world 8–10 .…”

Contentopic mapping in ventral and dorsal association cortex: the topographical organization of manipulable object information

“…Population receptive field analysis (pRF) and topography. The pRF analysis is a forward modelling approach aimed at accounting for the observed BOLD time series in each individual voxel by estimating parameters at the neuronal population level 7,49,[53][54][55] . pRF parameters estimates were obtained from percent BOLD time series by fitting a linear model via an iterative procedure, similar to previous applications in the domains of vision, sensorimotor integration and quantity/numerosity perception The linear model can be described as follows:…”

Peri-Saccadic Orientation Identification Performance and Visual Neural Sensitivity Are Higher in the Upper Visual Field

“…It is generally accepted that the sensitivity of the visual system to brief peri-saccadic flashes is strongly suppressed (Latour, 1962;Diamond et al, 2000;Idrees et al, 2020). However, residual visual processing still takes place at the time of saccades (De Pisapia et al, 2010;Fracasso et al, 2015;Fracasso and Melcher, 2016;Fabius et al, 2022), allowing us to ask whether such processing is more sensitive in the upper or lower visual fields. We first asked human subjects to generate horizontal saccades, and we presented upper or lower visual field peri-saccadic flashes, which were near the vertical retinotopic meridian at the time of peak saccadic suppression.…”