Latent encoding of movement in primary visual cortex
4 affiliations
- 1Department of Mathematics, Brandeis University, Waltham, Massachusetts, United States of America
- 2Department of Biology, Brandeis University, Waltham, Massachusetts, United States of America
- 3Volen National Center for Complex Systems, Brandeis University, Waltham, Massachusetts, United States of America
- 4Neuroglial Interactions in Cerebral Physiology and Pathologies, Center for Interdisciplinary Research in Biology, Collège de France, CNRS, INSERM, Labex Memolife, Université PSL, Paris, France
Abstract
Neurons in the primary visual cortex (V1) are classically thought to encode spatial features of visual stimuli through simple population codes: each neuron exhibits a preferred orientation and preferred spatial frequency, both of which remaining invariant to other aspects of the visual stimulus. Here, we show that this simple rule does not apply to the representation of major features of stimulus motion, including stimulus direction and temporal frequency (TF). We collected an extensive dataset of cat V1 responses to stimuli covarying in orientation, direction, spatial frequency, and TF to assess the extent of motion selectivity. We show that preferred TF is mostly uniform across the cortical surface. Yet, in over half of V1, the preferred direction is reversed with changing stimulus TF, revealing four distinct map motifs embedded in V1’s functional architecture. Similarly, despite the lack of spatial modulation for the preferred TF map and the lack of invariance for the preferred direction map, we found using convolutional neural networks that direction, TF and stimulus speed can be accurately decoded from V1 responses at all cortical locations. These findings suggest a latent encoding of information where subtle, spatially distributed modulations of V1 activity may convey fine information about stimulus motion, pointing to a novel primary sensory encoding mechanism despite complex co-variation of responses to multiple attributes across V1 neurons. Author summary: Understanding how the external world is represented in the brain has long been a central endeavor in neuroscience. In the mammalian primary visual cortex (V1), groups of neurons encode stimulus properties through changes in activity, with “preferred” stimuli eliciting maximal responses, as demonstrated for orientation and spatial frequency. We collected an extensive dataset of neuronal responses to investigate whether V1 also encodes motion speed. We found that cortical response organization remains largely invariant, in the sense that neurons uniformly exhibit a preference to the same temporal frequency. Yet, machine learning algorithms decode motion speed with remarkable precision, revealing that subtle, spatially distributed modulations of activity underlie speed encoding. Moreover, we show that preferred direction flips with speed across 80% of cortex, uncovering a novel co-organization of motion and direction information in V1.
Related
- Anterior cingulate feedback facilitates rapid global context modulation in primary visual cortex
- Cross-attention encoding models reveal dynamic spatiotemporal routing across human higher visual cortex
- Feedback to the primary visual cortex is highly concentrated on the central visual field representation
- A cognitive representation in primary visual cortex modulated by vision
- Selective collateralization of transcriptomically distinct neurons organizes visual-stream output from the primary visual cortex
BibTeX
@article{germaine2026latent,
title = {{Latent encoding of movement in primary visual cortex}},
author = {Hannah Germaine and Isaiah Choi and Stephen D. Van Hooser and Jérôme Ribot and Jonathan David Touboul},
journal = {PLOS Computational Biology},
year = {2026},
month = oct,
doi = {10.1371/journal.pcbi.1014826},
url = {https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1014826},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Latent encoding of movement in primary visual cortex
AU - Hannah Germaine
AU - Isaiah Choi
AU - Stephen D. Van Hooser
AU - Jérôme Ribot
AU - Jonathan David Touboul
PY - 2026
DA - 2026/10/12
JO - PLOS Computational Biology
DO - 10.1371/journal.pcbi.1014826
UR - https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1014826
AB - Neurons in the primary visual cortex (V1) are classically thought to encode spatial features of visual stimuli through simple population codes: each neuron exhibits a preferred orientation and preferred spatial frequency, both of which remaining invariant to other aspects of the visual stimulus. Here, we show that this simple rule does not apply to the representation of major features of stimulus motion, including stimulus direction and temporal frequency (TF). We collected an extensive dataset of cat V1 responses to stimuli covarying in orientation, direction, spatial frequency, and TF to assess the extent of motion selectivity. We show that preferred TF is mostly uniform across the cortical surface. Yet, in over half of V1, the preferred direction is reversed with changing stimulus TF, revealing four distinct map motifs embedded in V1’s functional architecture. Similarly, despite the lack of spatial modulation for the preferred TF map and the lack of invariance for the preferred direction map, we found using convolutional neural networks that direction, TF and stimulus speed can be accurately decoded from V1 responses at all cortical locations. These findings suggest a latent encoding of information where subtle, spatially distributed modulations of V1 activity may convey fine information about stimulus motion, pointing to a novel primary sensory encoding mechanism despite complex co-variation of responses to multiple attributes across V1 neurons. Author summary: Understanding how the external world is represented in the brain has long been a central endeavor in neuroscience. In the mammalian primary visual cortex (V1), groups of neurons encode stimulus properties through changes in activity, with “preferred” stimuli eliciting maximal responses, as demonstrated for orientation and spatial frequency. We collected an extensive dataset of neuronal responses to investigate whether V1 also encodes motion speed. We found that cortical response organization remains largely invariant, in the sense that neurons uniformly exhibit a preference to the same temporal frequency. Yet, machine learning algorithms decode motion speed with remarkable precision, revealing that subtle, spatially distributed modulations of activity underlie speed encoding. Moreover, we show that preferred direction flips with speed across 80% of cortex, uncovering a novel co-organization of motion and direction information in V1.
SN - 1553-7358
SN - 1553-734X
ER -