A unifying perspective on neural manifolds and circuits for cognition

Langdon, Christopher; Genkin, Mikhail; Engel, Tatiana A.

doi:10.1038/s41583-023-00693-x

Cited by 75 publications

(51 citation statements)

References 135 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Critically, the geometric transformation from the sensory manifold to the perceptual manifold was realized through collaborative work combining specialized neuronal properties of nonlinear mixed selectivity with heterogeneous connectivity pattern among these neurons. In summary, our study demonstrated an unambiguous line drawn between sensation and perception from a neural geometry perspective, enriching our understanding to the dictum that “structure determines function” in neural networks (Langdon et al, 2023; Sporns, 2014).…”

Section: Discussionsupporting

confidence: 58%

“…While necessary, NMS neurons alone seem insufficient for dimension expansion, as it has been suggested that the dimensionality of neural manifolds is also dictated by the connectivity structure of neural networks (Langdon et al, 2023). Here we further investigated the connectivity structure among NMS neurons that may underpin dimension expansion.…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

From Sensory to Perceptual Manifolds: The Twist of Neural Geometry

Ma,

Jiang,

Liu

et al. 2023

Preprint

View full text Add to dashboard Cite

To humans, everything is classifiable-big versus small, edible or poisonous, righteous or unjust. Similarly, at the heart of most machine learning tasks lies the fundamental goal of classification, yet the enduring challenge of linear inseparability has plagued artificial neural networks since their inception. Here we asked how biological neural networks tackle this issue by investigating the geometric embedding of neural manifolds in macaques’ V2 during orientation discrimination of motion-induced illusory contour. Specifically, we constructed a three-dimensional stimulus space that inherently rendered the orientation classification of these contours as a linear-inseparable problem. As expected, we identified a sensory manifold, formed by the V2 neuron population, that faithfully corresponded to this stimulus space. Crucially, this sensory manifold underwent a series of twist operations, resulting in new axes orthogonal to the original ones. Within this expanded, high-dimensional perceptual manifold, the problem of linear inseparability became linearly separable. Computational modeling further revealed that the geometric transformation from the sensory to perceptual manifold was achieved by neurons exhibiting nonlinear mixed selectivity and heterogeneous connectivity patterns among them. Taken together, our findings elucidate how perception arises from sensation from a neural geometry perspective, enriching our understanding of how cognitive functions derive from underlying anatomical structure.

show abstract

Section: Discussionsupporting

confidence: 58%

Section: Resultsmentioning

confidence: 99%

From Sensory to Perceptual Manifolds: The Twist of Neural Geometry

Ma,

Jiang,

Liu

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

Section: Discussionsupporting

confidence: 52%

“…Altogether, our results delineate a cortical network and a key area and PN types therein that articular the serial order and coordination of a skilled behavior; they begin to integrate the neural circuit and neural manifold explanation of cortical computation that control behavior 98 Besides wrist rotation movement, the hand rotation score is also a reflection of the palm-facing direction. Value 1 means the palm is facing upward and -1 indicates the palm facing downward.…”

Section: Cell Type Contribution To Neural Population Dynamics and Act...supporting

confidence: 52%

Cortical network and projection neuron types that articulate serial order in a skilled motor behavior

Li,

An,

Qian

et al. 2023

Preprint

View full text Add to dashboard Cite

Skilled motor behaviors require orderly coordination of multiple constituent movements with sensory cues towards achieving a goal, but the underlying brain circuit mechanisms remain unclear. Here we show that target-guided reach-grasp-to-drink (RGD) in mice involves the ordering and coordination of a set of forelimb and oral actions. Cortex-wide activity imaging of multiple glutamatergic projection neuron (PN) types uncovered a network, involving the secondary motor cortex (MOs), forelimb primary motor and somatosensory cortex, that tracked RGD movements. Photo-inhibition highlighted MOs in coordinating RGD movements. Within the MOs, population neural trajectories tracked RGD progression and single neuron activities integrated across constituent movements. Notably, MOs intratelencephalic, pyramidal tract, and corticothalamic PN activities correlated with action coordination, showed distinct neural dynamics trajectories, and differentially contributed to movement coordination. Our results delineate a cortical network and key areas, PN types, and neural dynamics therein that articulate the serial order and coordination of a skilled behavior.

show abstract

“…Despite the widespread recognition that behaviour is underpinned by the dynamics of neural populations 1,2 , interpreting the structure of these dynamics in the context of relevant computations remains a fundamental challenge 3 . Several works have focused on the geometry of neural manifolds-lowdimensional smooth subspaces of neural state space, over which the high-dimensional neural activities evolve [4][5][6][7][8][9][10][11][12] -while others have suggested that computations are encoded by the dynamical flows of neural population activity 3,11,13 . In the latter viewpoint, manifold geometry is merely a dynamical imprint that changes over time or across animals 4 .…”

Section: Introductionmentioning

confidence: 99%

Interpretable statistical representations of neural population dynamics and geometry

Peach

Arnaudon

Barahona

et al. 2023

Preprint

View full text Add to dashboard Cite

The dynamics of neuron populations during diverse behaviours evolve on low-dimensional manifolds. However, it remains challenging to disentangle the role of manifold geometry and dynamics in encoding task variables. Here, we introduce an unsupervised geometric deep learning framework for representing non-linear dynamical systems based on statistical distributions of local dynamical features. Our method provides geometry-aware or geometry-agnostic representations for robustly comparing dynamical systems based on sparse measurements. Our representations are generalisable to compare computations across systems, interpretable to discover a geometric correspondence between neural dynamics and kinematics in a primate reaching task, and intrinsically encode temporal information to give rise to a decoding algorithm with state-of-the-art accuracy. Our results suggest that using the manifold structure over temporal information is important to develop better decoding algorithms and assimilate data across experiments.

show abstract

A unifying perspective on neural manifolds and circuits for cognition

Cited by 75 publications

References 135 publications

From Sensory to Perceptual Manifolds: The Twist of Neural Geometry

From Sensory to Perceptual Manifolds: The Twist of Neural Geometry

Cortical network and projection neuron types that articulate serial order in a skilled motor behavior

Interpretable statistical representations of neural population dynamics and geometry

Contact Info

Product

Resources

About