Yu Hu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Yu Hu has a total of 42 co-authors. Here are the top co-authors by number of shared publications:
Chen Shen
Hefei National Laboratory for Physical Sciences at the Microscale, Center for Integrative Imaging, University of Science and Technology of China
3 shared publicationsGuodong Tan
Division of Life Sciences and Medicine, University of Science and Technology of China
3 shared publicationsHongtai Ren
Department of Precision Machinery and Precision Instrumentation, University of Science and Technology of China
3 shared publicationsKexin Qi
Hefei National Laboratory for Physical Sciences at the Microscale, Center for Integrative Imaging, University of Science and Technology of China
3 shared publications- Quan Wen
Division of Life Sciences and Medicine, University of Science and Technology of China
3 shared publications Shiwu Zhang
Department of Precision Machinery and Precision Instrumentation, University of Science and Technology of China
3 shared publications
Latest Publications
- Open Access
- 8 January 2026
- Published by eLife Sciences Publications, Ltd in eLife preprints
Pallium-encoded valence-specific chemosensory amplification of eye-body coordination in larval zebrafish
Coordinated eye-body movements are essential for adaptive behavior, yet little is known about how multisensory input, particularly chemosensory cues, shapes this coordination. Using our enhanced Fish-On-Chips optofluidic platform, we uncovered complex dynamics in how larval zebrafish coordinate saccadic eye movements with tail flips. Under baseline conditions, spontaneous tail flips dynamically align with saccades in frequency and direction for coordinated turns. Chemosensory valence further modulates this coordination: death-associated cues intensify both the strength and frequency of coupled saccade-tail flips during turns, whereas food-related cues promote forward gliding without altering saccade coupling. Concurrent brain-wide neuronal imaging reveals that the zebrafish pallium represents the transformation of aversive valence into enhanced saccade-tail coordination, with stronger coupling associated with higher pallium activation. These findings uncover the neural basis by which chemosensory inputs of different valences distinctly regulate eye-body coordination to drive adaptive locomotion in a developing vertebrate, offering insights into principles of sensory-motor integration.
- 8 January 2026
- Published by eLife Sciences Publications, Ltd in eLife preprints
Open AccessPallium-encoded valence-specific chemosensory amplification of eye-body coordination in larval zebrafish
Coordinated eye-body movements are essential for adaptive behavior, yet little is known about how multisensory input, particularly chemosensory cues, shapes this coordination. Using our enhanced Fish-On-Chips optofluidic platform, we uncovered complex dynamics in how larval zebrafish coordinate saccadic eye movements with tail flips. Under baseline conditions, spontaneous tail flips dynamically align with saccades in frequency and direction for coordinated turns. Chemosensory valence further modulates this coordination: death-associated cues intensify both the strength and frequency of coupled saccade-tail flips during turns, whereas food-related cues promote forward gliding without altering saccade coupling. Concurrent brain-wide neuronal imaging reveals that the zebrafish pallium represents the transformation of aversive valence into enhanced saccade-tail coordination, with stronger coupling associated with higher pallium activation. These findings uncover the neural basis by which chemosensory inputs of different valences distinctly regulate eye-body coordination to drive adaptive locomotion in a developing vertebrate, offering insights into principles of sensory-motor integration.
- Open Access
Covariance spectrum in nonlinear recurrent neural networks
Advances in simultaneous recordings of large numbers of neurons have driven significant interest in the structure of neural population activity such as dimension. A key question is how these dynamic features arise mechanistically and their relationship to circuit connectivity. It was previously proposed to use the covariance eigenvalue distribution, or spectrum, which can be analytically derived in random recurrent networks, as a robust measure to describe the shape of neural population activity beyond the dimension (Hu and Sompolinsky 2022). Applications of the theoretical spectrum have broadly found accurate matches to experimental data across brain areas providing mechanistic insights into the observed low dimensional population dynamics (Morales et al. 2023). However, the empirical success highlights a gap in theory, as the neural network model used to derive the spectrum was minimal with linear neurons. In this work, we aim to close this gap by studying the covariance spectrum in networks with nonlinear neurons and under broader dynamical regimes including chaos. Surprisingly, we found that the spectrum can be precisely understood by equations analogous to the linear theory substituted with an effective recurrent connection strength parameter, that reflects both the connection weights and the nonlinearity of neurons. Across dynamical regimes, this effective connection strength provides a unified interpretation for the spectrum and dimension changes, and stays near the critical value in the chaotic regime without fine-tuning. These results further our understanding of nonlinear neural population dynamics and provide additional theoretical support for applying the covariance spectrum analysis in biological circuits.
Open AccessCovariance spectrum in nonlinear recurrent neural networks
Advances in simultaneous recordings of large numbers of neurons have driven significant interest in the structure of neural population activity such as dimension. A key question is how these dynamic features arise mechanistically and their relationship to circuit connectivity. It was previously proposed to use the covariance eigenvalue distribution, or spectrum, which can be analytically derived in random recurrent networks, as a robust measure to describe the shape of neural population activity beyond the dimension (Hu and Sompolinsky 2022). Applications of the theoretical spectrum have broadly found accurate matches to experimental data across brain areas providing mechanistic insights into the observed low dimensional population dynamics (Morales et al. 2023). However, the empirical success highlights a gap in theory, as the neural network model used to derive the spectrum was minimal with linear neurons. In this work, we aim to close this gap by studying the covariance spectrum in networks with nonlinear neurons and under broader dynamical regimes including chaos. Surprisingly, we found that the spectrum can be precisely understood by equations analogous to the linear theory substituted with an effective recurrent connection strength parameter, that reflects both the connection weights and the nonlinearity of neurons. Across dynamical regimes, this effective connection strength provides a unified interpretation for the spectrum and dimension changes, and stays near the critical value in the chaotic regime without fine-tuning. These results further our understanding of nonlinear neural population dynamics and provide additional theoretical support for applying the covariance spectrum analysis in biological circuits.
- Open Access
- 23 June 2025
- Published by eLife Sciences Publications, Ltd in eLife
- Vol. 14
The geometry and dimensionality of brain-wide activity
Understanding neural activity organization is vital for deciphering brain function. By recording whole-brain calcium activity in larval zebrafish during hunting and spontaneous behaviors, we find that the shape of the neural activity space, described by the neural covariance spectrum, is scale-invariant: a smaller, randomly sampled cell assembly resembles the entire brain. This phenomenon can be explained by Euclidean Random Matrix theory, where neurons are reorganized from anatomical to functional positions based on their correlations. Three factors contribute to the observed scale invariance: slow neural correlation decay, higher functional space dimension, and neural activity heterogeneity. In addition to matching data from zebrafish and mice, our theory and analysis demonstrate how the geometry of neural activity space evolves with population sizes and sampling methods, thus revealing an organizing principle of brain-wide activity.
- 23 June 2025
- Published by eLife Sciences Publications, Ltd in eLife
- Vol. 14
Open AccessThe geometry and dimensionality of brain-wide activity
Understanding neural activity organization is vital for deciphering brain function. By recording whole-brain calcium activity in larval zebrafish during hunting and spontaneous behaviors, we find that the shape of the neural activity space, described by the neural covariance spectrum, is scale-invariant: a smaller, randomly sampled cell assembly resembles the entire brain. This phenomenon can be explained by Euclidean Random Matrix theory, where neurons are reorganized from anatomical to functional positions based on their correlations. Three factors contribute to the observed scale invariance: slow neural correlation decay, higher functional space dimension, and neural activity heterogeneity. In addition to matching data from zebrafish and mice, our theory and analysis demonstrate how the geometry of neural activity space evolves with population sizes and sampling methods, thus revealing an organizing principle of brain-wide activity.
- Top Cited Papers
- 22 April 2025
- Published by Springer Nature in Nature Methods
- Vol. 22  (6) , 1376-1385
Mapping effective connectivity by virtually perturbing a surrogate brain
Effective connectivity (EC), which reflects the causal interactions between brain regions, is fundamental to understanding information processing in the brain; however, traditional methods for obtaining EC, which rely on neural responses to stimulation, are often invasive or limited in spatial coverage, making them unsuitable for whole-brain EC mapping in humans. Here, to address this gap, we introduce Neural Perturbational Inference (NPI), a data-driven framework for mapping whole-brain EC. NPI employs an artificial neural network trained to model large-scale neural dynamics, serving as a computational surrogate of the brain. By systematically perturbing all regions in the surrogate brain and analyzing the resulting responses in other regions, NPI maps the directionality, strength and excitatory/inhibitory properties of brain-wide EC. Validation of NPI on generative models with known ground-truth EC demonstrates its superiority over existing methods such as Granger causality and dynamic causal modeling. When applied to resting-state functional magnetic resonance imaging data across diverse datasets, NPI reveals consistent, structurally supported EC patterns. Furthermore, comparisons with cortico-cortical evoked potential data show a strong resemblance between NPI-inferred EC and real stimulation propagation patterns. By transitioning from correlational to causal understandings of brain functionality, NPI marks a stride in decoding the brain’s functional architecture and facilitating both neuroscience studies and clinical applications.
- 22 April 2025
- Published by Springer Nature in Nature Methods
- Vol. 22  (6) , 1376-1385
Top Cited PapersMapping effective connectivity by virtually perturbing a surrogate brain
Effective connectivity (EC), which reflects the causal interactions between brain regions, is fundamental to understanding information processing in the brain; however, traditional methods for obtaining EC, which rely on neural responses to stimulation, are often invasive or limited in spatial coverage, making them unsuitable for whole-brain EC mapping in humans. Here, to address this gap, we introduce Neural Perturbational Inference (NPI), a data-driven framework for mapping whole-brain EC. NPI employs an artificial neural network trained to model large-scale neural dynamics, serving as a computational surrogate of the brain. By systematically perturbing all regions in the surrogate brain and analyzing the resulting responses in other regions, NPI maps the directionality, strength and excitatory/inhibitory properties of brain-wide EC. Validation of NPI on generative models with known ground-truth EC demonstrates its superiority over existing methods such as Granger causality and dynamic causal modeling. When applied to resting-state functional magnetic resonance imaging data across diverse datasets, NPI reveals consistent, structurally supported EC patterns. Furthermore, comparisons with cortico-cortical evoked potential data show a strong resemblance between NPI-inferred EC and real stimulation propagation patterns. By transitioning from correlational to causal understandings of brain functionality, NPI marks a stride in decoding the brain’s functional architecture and facilitating both neuroscience studies and clinical applications.
- Open Access
- 24 March 2025
- Published by National Academy of Sciences in Proceedings of the National Academy of Sciences
- Vol. 122  (13)
The visuomotor transformations underlying target-directed behavior
The visual system can process diverse stimuli and make the decision to execute appropriate behaviors, but it remains unclear where and how this transformation takes place. Innate visually evoked behaviors such as hunting, freezing, and escape are thought to be deeply conserved, and have been described in a range of species from insects to humans. We found that zebrafish larvae would respond to predator-like visual stimuli with immobility and bradycardia, both hallmarks of freezing, in a head-fixed behavioral paradigm. We then imaged the zebrafish visual system while larvae responded to different visual stimuli with hunting, freezing, and escape behaviors and systematically identified visually driven neurons and behaviorally correlated sensorimotor neurons. Our analyses indicate that within the optic tectum, broadly tuned sensory neurons are functionally correlated with sensorimotor neurons which respond specifically during one behavior, indicating that it contains suitable information for sensorimotor transformation. We also identified sensorimotor neurons in four other areas downstream of the tectum, and these neurons are also specific for one behavior, indicating that the segregation of the pathways continues in other areas. While our findings shed light on how sensorimotor neurons may integrate visual inputs, further investigation will be required to determine how sensorimotor neurons in different regions interact and where the decision to behave is made.
- 24 March 2025
- Published by National Academy of Sciences in Proceedings of the National Academy of Sciences
- Vol. 122  (13)
Open AccessThe visuomotor transformations underlying target-directed behavior
The visual system can process diverse stimuli and make the decision to execute appropriate behaviors, but it remains unclear where and how this transformation takes place. Innate visually evoked behaviors such as hunting, freezing, and escape are thought to be deeply conserved, and have been described in a range of species from insects to humans. We found that zebrafish larvae would respond to predator-like visual stimuli with immobility and bradycardia, both hallmarks of freezing, in a head-fixed behavioral paradigm. We then imaged the zebrafish visual system while larvae responded to different visual stimuli with hunting, freezing, and escape behaviors and systematically identified visually driven neurons and behaviorally correlated sensorimotor neurons. Our analyses indicate that within the optic tectum, broadly tuned sensory neurons are functionally correlated with sensorimotor neurons which respond specifically during one behavior, indicating that it contains suitable information for sensorimotor transformation. We also identified sensorimotor neurons in four other areas downstream of the tectum, and these neurons are also specific for one behavior, indicating that the segregation of the pathways continues in other areas. While our findings shed light on how sensorimotor neurons may integrate visual inputs, further investigation will be required to determine how sensorimotor neurons in different regions interact and where the decision to behave is made.
- Open Access
- 21 March 2025
- Published by eLife Sciences Publications, Ltd in eLife preprints
- Vol. 14
The Geometry and Dimensionality of Brain-wide Activity
Understanding neural activity organization is vital for deciphering brain function. By recording whole-brain calcium activity in larval zebrafish during hunting and spontaneous behaviors, we find that the shape of the neural activity space, described by the neural covariance spectrum, is scale-invariant: a smaller, randomly sampled cell assembly resembles the entire brain. This phenomenon can be explained by Euclidean Random Matrix theory, where neurons are reorganized from anatomical to functional positions based on their correlations. Three factors contribute to the observed scale invariance: slow neural correlation decay, higher functional space dimension, and neural activity heterogeneity. In addition to matching data from zebrafish and mice, our theory and analysis demonstrate how the geometry of neural activity space evolves with population sizes and sampling methods, thus revealing an organizing principle of brain-wide activity.
- 21 March 2025
- Published by eLife Sciences Publications, Ltd in eLife preprints
- Vol. 14
Open AccessThe Geometry and Dimensionality of Brain-wide Activity
Understanding neural activity organization is vital for deciphering brain function. By recording whole-brain calcium activity in larval zebrafish during hunting and spontaneous behaviors, we find that the shape of the neural activity space, described by the neural covariance spectrum, is scale-invariant: a smaller, randomly sampled cell assembly resembles the entire brain. This phenomenon can be explained by Euclidean Random Matrix theory, where neurons are reorganized from anatomical to functional positions based on their correlations. Three factors contribute to the observed scale invariance: slow neural correlation decay, higher functional space dimension, and neural activity heterogeneity. In addition to matching data from zebrafish and mice, our theory and analysis demonstrate how the geometry of neural activity space evolves with population sizes and sampling methods, thus revealing an organizing principle of brain-wide activity.
- Open Access
Pallium-encoded valence-specific chemosensory amplification of eye-body coordination in larval zebrafish
Summary: Coordinated eye-body movements are essential for adaptive behavior, yet little is known about how multisensory input, particularly chemosensory cues, shapes this coordination. Using our enhanced Fish-On-Chips optofluidic platform, we uncovered complex dynamics in how larval zebrafish coordinate saccadic eye movements with tail flips. Under baseline conditions, spontaneous tail flips dynamically align with saccades in frequency and direction for coordinated turns. Chemosensory valence further modulates this coordination: death-associated cues intensify both the strength and frequency of coupled saccade-tail flips during turns, whereas food-related cues promote forward gliding without altering saccade coupling. Concurrent brain-wide neuronal imaging reveals that the zebrafish pallium represents the transformation of aversive valence into enhanced saccade-tail coordination, with stronger coupling associated with higher pallium activation. These findings uncover the neural basis by which chemosensory inputs of different valences distinctly regulate eye-body coordination to drive adaptive locomotion in a developing vertebrate, offering insights into principles of sensory-motor integration.
Open AccessPallium-encoded valence-specific chemosensory amplification of eye-body coordination in larval zebrafish
Summary: Coordinated eye-body movements are essential for adaptive behavior, yet little is known about how multisensory input, particularly chemosensory cues, shapes this coordination. Using our enhanced Fish-On-Chips optofluidic platform, we uncovered complex dynamics in how larval zebrafish coordinate saccadic eye movements with tail flips. Under baseline conditions, spontaneous tail flips dynamically align with saccades in frequency and direction for coordinated turns. Chemosensory valence further modulates this coordination: death-associated cues intensify both the strength and frequency of coupled saccade-tail flips during turns, whereas food-related cues promote forward gliding without altering saccade coupling. Concurrent brain-wide neuronal imaging reveals that the zebrafish pallium represents the transformation of aversive valence into enhanced saccade-tail coordination, with stronger coupling associated with higher pallium activation. These findings uncover the neural basis by which chemosensory inputs of different valences distinctly regulate eye-body coordination to drive adaptive locomotion in a developing vertebrate, offering insights into principles of sensory-motor integration.
- Open Access
The visuomotor transformations underlying target-directed behavior
Summary: The visual system can process diverse stimuli and make the decision to execute appropriate behaviors, but it remains unclear where and how this transformation takes place. We imaged the zebrafish visual system while larvae responded with hunting, freezing, and escape behaviors, and systematically identified visually driven neurons and behaviorally correlated sensorimotor neurons. Our analyses indicate within the optic tectum, broadly tuned sensory neurons are functionally connected to sensorimotor neurons that respond specifically during one behavior, transforming visual information into motor output. We also identified sensorimotor neurons in four other areas downstream of the tectum, and these neurons are also specific for one behavior, indicating that once the decision to behave has been made, the segregation of the pathways continues in later areas. Our findings suggest that the tectum receives visual sensory information and is responsible for selecting a single behavioral outcome, which is then relayed to downstream areas. Significance statement: Here, we developed a novel visually-evoked freezing paradigm in zebrafish, and combined this with escape and hunting behaviors to ask how visual stimuli are identified and converted into different behavioral outcomes. We found that the optic tectum contains neurons that detect all three stimuli, as well as sensorimotor neurons for the three behaviors, suggesting that it is a site of sensorimotor transformation, which was supported by our analysis of correlations between the populations. The sensorimotor neurons in the tectum are highly specific for one behavior, and this segregation is maintained in the three downstream areas where we also identified sensorimotor neurons, indicating that the tectum flexibly transforms visual information into a single behavioral output.
Open AccessThe visuomotor transformations underlying target-directed behavior
Summary: The visual system can process diverse stimuli and make the decision to execute appropriate behaviors, but it remains unclear where and how this transformation takes place. We imaged the zebrafish visual system while larvae responded with hunting, freezing, and escape behaviors, and systematically identified visually driven neurons and behaviorally correlated sensorimotor neurons. Our analyses indicate within the optic tectum, broadly tuned sensory neurons are functionally connected to sensorimotor neurons that respond specifically during one behavior, transforming visual information into motor output. We also identified sensorimotor neurons in four other areas downstream of the tectum, and these neurons are also specific for one behavior, indicating that once the decision to behave has been made, the segregation of the pathways continues in later areas. Our findings suggest that the tectum receives visual sensory information and is responsible for selecting a single behavioral outcome, which is then relayed to downstream areas. Significance statement: Here, we developed a novel visually-evoked freezing paradigm in zebrafish, and combined this with escape and hunting behaviors to ask how visual stimuli are identified and converted into different behavioral outcomes. We found that the optic tectum contains neurons that detect all three stimuli, as well as sensorimotor neurons for the three behaviors, suggesting that it is a site of sensorimotor transformation, which was supported by our analysis of correlations between the populations. The sensorimotor neurons in the tectum are highly specific for one behavior, and this segregation is maintained in the three downstream areas where we also identified sensorimotor neurons, indicating that the tectum flexibly transforms visual information into a single behavioral output.
- Open Access
The Geometry and Dimensionality of Brain-wide Activity
Understanding neural activity organization is vital for deciphering brain function. By recording whole-brain calcium activity in larval zebrafish during hunting and spontaneous behaviors, we find that the shape of the neural activity space, described by the neural covariance spectrum, is scale-invariant: a smaller, randomly sampled cell assembly resembles the entire brain. This phenomenon can be explained by Euclidean Random Matrix theory, where neurons are reorganized from anatomical to functional positions based on their correlations. Three factors contribute to the observed scale invariance: slow neural correlation decay, higher functional space dimension, and neural activity heterogeneity. In addition to matching data from zebrafish and mice, our theory and analysis demonstrate how the geometry of neural activity space evolves with population sizes and sampling methods, thus revealing an organizing principle of brain-wide activity.
Open AccessThe Geometry and Dimensionality of Brain-wide Activity
Understanding neural activity organization is vital for deciphering brain function. By recording whole-brain calcium activity in larval zebrafish during hunting and spontaneous behaviors, we find that the shape of the neural activity space, described by the neural covariance spectrum, is scale-invariant: a smaller, randomly sampled cell assembly resembles the entire brain. This phenomenon can be explained by Euclidean Random Matrix theory, where neurons are reorganized from anatomical to functional positions based on their correlations. Three factors contribute to the observed scale invariance: slow neural correlation decay, higher functional space dimension, and neural activity heterogeneity. In addition to matching data from zebrafish and mice, our theory and analysis demonstrate how the geometry of neural activity space evolves with population sizes and sampling methods, thus revealing an organizing principle of brain-wide activity.
- Top Cited PapersOpen Access
- 21 July 2022
- Published by Public Library of Science (PLoS) in PLOS Computational Biology
- Vol. 18  (7) , e1010327
The spectrum of covariance matrices of randomly connected recurrent neuronal networks with linear dynamics
A key question in theoretical neuroscience is the relation between the connectivity structure and the collective dynamics of a network of neurons. Here we study the connectivity-dynamics relation as reflected in the distribution of eigenvalues of the covariance matrix of the dynamic fluctuations of the neuronal activities, which is closely related to the network dynamics’ Principal Component Analysis (PCA) and the associated effective dimensionality. We consider the spontaneous fluctuations around a steady state in a randomly connected recurrent network of stochastic neurons. An exact analytical expression for the covariance eigenvalue distribution in the large-network limit can be obtained using results from random matrices. The distribution has a finitely supported smooth bulk spectrum and exhibits an approximate power-law tail for coupling matrices near the critical edge. We generalize the results to include second-order connectivity motifs and discuss extensions to excitatory-inhibitory networks. The theoretical results are compared with those from finite-size networks and the effects of temporal and spatial sampling are studied. Preliminary application to whole-brain imaging data is presented. Using simple connectivity models, our work provides theoretical predictions for the covariance spectrum, a fundamental property of recurrent neuronal dynamics, that can be compared with experimental data.
- 21 July 2022
- Published by Public Library of Science (PLoS) in PLOS Computational Biology
- Vol. 18  (7) , e1010327
Top Cited PapersOpen AccessThe spectrum of covariance matrices of randomly connected recurrent neuronal networks with linear dynamics
A key question in theoretical neuroscience is the relation between the connectivity structure and the collective dynamics of a network of neurons. Here we study the connectivity-dynamics relation as reflected in the distribution of eigenvalues of the covariance matrix of the dynamic fluctuations of the neuronal activities, which is closely related to the network dynamics’ Principal Component Analysis (PCA) and the associated effective dimensionality. We consider the spontaneous fluctuations around a steady state in a randomly connected recurrent network of stochastic neurons. An exact analytical expression for the covariance eigenvalue distribution in the large-network limit can be obtained using results from random matrices. The distribution has a finitely supported smooth bulk spectrum and exhibits an approximate power-law tail for coupling matrices near the critical edge. We generalize the results to include second-order connectivity motifs and discuss extensions to excitatory-inhibitory networks. The theoretical results are compared with those from finite-size networks and the effects of temporal and spatial sampling are studied. Preliminary application to whole-brain imaging data is presented. Using simple connectivity models, our work provides theoretical predictions for the covariance spectrum, a fundamental property of recurrent neuronal dynamics, that can be compared with experimental data.