Biology subjects

Wang, X.-J.

Publications and source records attributed to Wang, X.-J..

At least 19 recordsLinked to original sources

bioRxiv2026

A neural circuit algorithm for action grammar

How does the brain generate sequences of discrete elements, such as actions or words, that follow an abstract rule (a grammar)? Motivated by an experiment in which monkeys learned to draw multi-object images according to grammatical rules, we trained a recurrent neural network to generate sequences Am Bn: producing shape A m times, then B n times, or vice versa. Our findings are threefold. First, the network plans the sequence by moving to an initial state encoding m and n along orthogonal directions for upcoming execution. Second, the network repeats each shape by progressing along a shape-specific manifold. Third, transitions from Am to Bn follow parallel trajectories between manifolds across values of n, thus preserving n. Each of these is an experimentally testable prediction. The network generalizes to longer novel sequences by starting further along the same orthogonal directions. These results reveal a neural circuit implementation of a symbolic algorithm for generating rule-based action sequences.

neuroscience↗

bioRxiv2026

Metabotropic Glutamatergic Signaling Adaptively Controls Trans-thalamic Communication

Trans-thalamic circuits link cortical regions through higher-order sensory (HOs) thalamic nuclei to support cognition. These circuits are rich with metabotropic glutamate receptors (mGluRs), whose seconds-long dynamics conflict with the thalamuss traditional relay role, leaving their function unknown. Here, we present a model of group I mGluR-mediated suppression of K+ leak currents in single neurons. Including mGluRs in a trans-thalamic network of spiking neurons revealed distinct modulatory roles for thalamocortical and corticothalamic mGluRs. In pyramidal neurons, thalamocortical mGluRs strengthen somato-dendritic coupling and dendritic plateau potentials, boosting feedforward propagation of sensory information to enhance decision-making in a detection task. In HOs thalamic neurons, corticothalamic mGluRs temporally integrate inputs to shift thalamic firing from burst to tonic, a prediction supported by pulvinar recordings in behaving animals. This effect increases detection of a novel stimulus ("wake-up call") while attenuating detection of repeated stimuli. Therefore, mGluRs enable the HOs thalamus to dynamically modulate cortical communication.

neuroscience↗

bioRxiv2026

Visual field map size constrains working memory precision

Individual differences in working memory are associated with cognitive functioning and real-world outcomes, including general intelligence, academic achievement, and psychiatric and neurological disorders. Despite its importance, the neurobiological mechanisms that drive individual differences in working memory are unknown. Here, we asked whether the precision of spatial working memory covaries with the cortical surface area of retinotopically organized maps that are thought to store the content of working memory. In particular, we hypothesized that individuals with larger visual field maps would have more precise working memory. We assessed human subjects (male and female) working memory precision using a memory-guided saccade task, and using fMRI we separately mapped retinotopically organized regions of visual, parietal, and frontal cortex. In support of our hypothesis, we found that people with larger maps in primary visual cortex and parietal cortex had better memory. To test hypotheses about the potential mechanisms driving our results, we simulated working memory in neural network models with varying numbers of synthetic neurons. Echoing our empirical results, larger networks exhibited less memory error. This reduction in error could be fully attributed to a decrease in the proportion of memory-perturbing noise along the attractor in larger networks, establishing a pure size effect on working memory precision.

neuroscience↗

bioRxiv2026

Interference-Free Propagation: Achieving Reliable Signal Propagation in Cortical Networks with Areal-Specific Local Dynamics

In cortical networks, how to simultaneously achieve both reliable inter-areal signal propagation and areal-specific local dynamics remains largely unclear. In general, strong connections between areas enhance signal propagation, but blend timescales of area-specific neuronal activity, whereas weak connections have the opposite effect. Here we identify a novel dynamical regime termed "interference-free propagation" (IFP) that reconciles the two contrasting demands in the cortex. In the IFP regime, mean signals from upstream areas can propagate reliably, but fluctuations indicative of upstream areas timescales are filtered out. This result provides new insights into the operational regime of the cortex, leading to the coexistence of reliable signal propagation and the distinct property of local temporal integration of information in the cortical network.

neuroscience↗

bioRxiv2026

CNS diseases cerebrospinal fluid single-cell atlas reveals immune characteristics of neuropsychiatric systemic lupus erythematosus

Neuropsychiatric systemic lupus erythematosus (NPSLE) is a potentially severe complication of systemic lupus erythematosus (SLE), yet its pathogenesis remains largely elusive. By jointly probing the immune dynamics of subjects cerebrospinal fluid (CSF) and peripheral blood, we showed that both innate and adaptive immune responses jointly contribute to the pathogenesis of NPSLE. In particular, we found the remarkable enrichment of BAM-CCL3, a subtype of border-associated macrophages with strong recruitment capacity, implicating its potential role in central nervous system (CNS) inflammation. We also observed pronounced activation of memory B cells and CD4+ regulatory T cells in NPSLE CSF, along with the preferential blood-to-CSF migration and subsequent within-CSF clonal expansion of CD8+ effector memory T cells in NPSLE patients, suggesting a persistent CNS-localized adaptive immune dysregulation. Finally, we developed the single-cell CNS disease CSF-Blood Atlas (scCDCB), a comprehensive collection for CSF and peripheral blood of multiple CNS diseases, which is publicly available at (https://sccdcb.gao-lab.org) to serve as a reference for future research on CNS diseases.

genomics↗

bioRxiv2026

Neural representation of action grammar structure in primate frontal cortex

Humans and other animals can solve new problems, even on the first attempt. This capacity to generate novel problem-solving behavior has been hypothesized to depend on brain mechanisms for recombining units of knowledge using systems of procedural rules, or grammars. Yet, whether and how the brain represents and implements grammars remains unclear, in large part due to the lack of neural recordings (typically performed in animal models) during grammar-based problem solving (typically studied in humans). Here, we address that gap, and in turn identify an underlying neural basis of grammatical behavior. We designed a visual-motor construction task in which macaque monkeys trace complex, often novel, geometric figures by generating a sequence of strokes, each a distinct shape, like a circle, dash, or chevron (action symbols1). Critically, sequencing was guided by a learned hierarchical action grammar AnBmCk, meaning "repeat shape A, then repeat shape B, then repeat shape C", where the number of repetitions (n, m, and k) varied across problems. Behavior was internally generated (no cues guiding stroke order), problem-directed (to construct a specific image), and exhibited zero-shot generalization (successful on the first trial for novel problems, including harder ones)--key hallmarks of grammar use. To identify underlying neural substrates, we recorded from multiple frontal cortical areas previously implicated in rule use or action sequencing. Activity during drawing encoded key structural properties of the action grammar: (i) shape index (A, B, C), (ii) abstract role (e.g., the role A independent of the specific shape), and (iii) ordinal position within a shape repeat. Moreover, all three properties were strongest in a single region, the pre-supplementary motor area (preSMA). We suggest the possibility that this conjunction of representations is a signature of an implementation of an iterative algorithm resembling a "for-loop" program: internally tracking how many repetitions remain for each shape index and switching to the next when complete. Our study establishes a paradigm for studying the neural basis of grammar-guided novel behavior and identifies in preSMA a dynamic representation of grammatical structure supporting such behavior.

neuroscience↗

bioRxiv2026

Smad7-based biologic targeting epidermis and stroma promotes healing of diabetic wounds in mice and pigs

This study aimed to identify novel mechanisms of diabetic wound healing defects and test a therapeutic intervention using diabetic mouse and pig models. We found Smad7 transgene expression in mouse epidermis promoted wound healing in diabetic mice. To isolate effects of Smad7 on wounds, we created a Smad7-based biologic (Tat-PYC-Smad7) that penetrated cells of the wound. Topical Tat-PYC-Smad7 treatment to diabetic pig and mouse wounds accelerated healing compared to vehicle controls. Tat-PYC-Smad7-treated wounds showed reduced TGF{beta}/NF{kappa}B signaling, faster re-epithelialization and better extracellular matrix remodeling. Tat-PYC-Smad7 also attenuated neutrophil NETosis, potentially acting through reductions in MPO enzymatic activity and MPO nuclear entry, consequently reducing chromatin decondensation and the release of NET components. Our study revealed that Tat-PYC-Smad7 promoted diabetic wound healing by targeting keratinocytes and neutrophils, providing insight into mechanisms of diabetic wound healing defects targetable by Smad7-based therapy.

immunology↗

bioRxiv2025

Duet model of predictive coding unifies diverse neuroscienceexperimental protocols

The brain continuously generates predictions about the external world. When stimulus X is presented repeatedly, the brain predicts that the next one is also X. A deviant stimulus Y elicits a stronger sensory response than the baseline, reflecting the amplification of an unexpected stimulus. Here, we introduce the duet predictive coding model, a minimal and biologically plausible framework in which neurons encode both positive and negative prediction errors. This model reproduces neural responses observed in vision and audition across diverse predictive coding paradigms, particularly omission. Our proposed circuit mechanism predicts (1) neurons tuned to negative prediction errors in the oddball paradigm, supported by experimental evidence in mice; (2) the magnitude of unexpected responses quantitatively depends on the dissimilarity between standard and deviant stimuli and diminishes with increasing interstimulus interval. Our findings suggest that the brains deviance detection relies on dual-error computation, offering a unifying explanation across seemingly disparate experimental protocols.

neuroscience↗

bioRxiv2025

Global error signal guides local optimization in mismatch calculation

1Corollary discharge denotes internal signals about the expected sensory consequences of ones own actions, leading to attenuation of sensory responses caused by self-produced stimulation. To investigate the underlying neural circuit mechanism, here we introduce a biologically plausible three-factor learning rule, where a global signal guides the updating of local inhibitory synapses to enable the computation of mismatch between a stimulus and its expectation or prediction. We show that our network model, endowed with positive and negative prediction error neurons, accounts for the salient physiological observations of motor-visual and motor-auditory mismatch responses in mice. Moreover, the model predicts that learning induces a bimodal distribution in activity correlation with stimulus and movement-induced prediction, supported by our analysis of neural data from a recent experiment. These results link global modulation to local learning for predictive error computation in the sensory areas, and shed insights into how disrupting inhibition impairs mismatch computation in specific ways.

neuroscience↗

bioRxiv2025

A hierarchy of time constants and reliable signal propagation in the marmoset cortex

A hierarchy of timescales in the cerebral cortex is functionally desirable for rapid information processing in sensory areas and slow time integration in association areas. Here, through an analysis of electrocorticography (ECoG) data, we identified a timescale hierarchy in the neocortex of marmoset, a primate species commonly used in neuroscience. Constrained by the anatomical and electrophysiological data, we developed a multi-regional model of the marmoset neocortex that captures the observed timescale hierarchy phenomenon. Furthermore, we used the model to investigate how the neocortex reconciles information integration on distinct timescales in local areas with reliable signal propagation globally across regions. We found that a near-criticality state is optimal for both localized signal integration within areas and reliable signal propagation across areas in the multi-regional neocortex. Our model also mechanically explains recent experimental observations that the structural and functional connectivities are less correlated in association areas than in sensory areas.

neuroscience↗

bioRxiv2025

Cross-species analysis of experimental PH at single cell resolution reveals prominent contributions Ednrb+ EC and Dhcr24+ macrophage populations

BackgroundAnimal models are used widely to study pulmonary hypertension (PH). The cell populations that respond to disease-inducing stimuli in these models and their relationship to human disease remain incompletely defined. Materials and methodThis study analyzed the relationship between several rodent models of PH and human disease at single-cell resolution. scRNA-seq was performed on lungs from mice exposed to hypoxia or Sugen/hypoxia, rats exposed to monocrotaline, and controls. A cross-species single-cell dataset was integrated with human lung cell atlas (HLCA) and single-cell dataset from idiopathic pulmonary arterial hypertension (IPAH) to identify overlapping cell subsets between experimental and human disease and species. ResultsHigh levels of overlap were found between species and models of PH, HLCA, and IPAH datasets. Cell subsets perturbed in rat and mouse PH were similar to those found in human disease, with macrophages and endothelial cells being most affected. A novel Dhcr24high macrophage subset harboring both tissue-remodeling and pro- inflammatory features was consistently increased across models. Several functionally diverse endothelial subtypes were found, including novel Ednrb+ and Nox2+ subpopulations, reflecting enhanced apoptosis, dysregulated angiogenesis and proliferation, and reactive oxygen species-mediated stress. These macrophage and endothelial subtypes expressed numerous PH drug target genes, and exhibited potential disease-specific intercellular interactions involving Angptl4, Cxcl12, and Sema3 signaling. Disease-associated changes in these populations were confirmed by immunofluorescence in lung tissues from animals and patients. ConclusionsWe established a comprehensive cross-species single-cell atlas of mainstream rodent PH models, highlighting several novel macrophage and endothelial subtypes and signaling motifs potentially contributing to human disease.

bioinformatics↗

bioRxiv2025

From comparative connectomics to large-scale working memory modeling in macaque and marmoset

Although macaques and marmosets are both primates of choice for studying the brain mechanisms of cognition, they differ in key aspects of anatomy and behavior. Interestingly, recent connectomic analysis revealed that strong top-down projections from the prefrontal cortex to the posterior parietal cortex, present in macaques and important for executive function, are absent in marmosets. Here, we propose a consensus mapping that bridges the two species cortical atlases and allows for direct area-to-area comparison of their connectomes, which are then used to build comparative computational large-scale modeling of the frontoparietal circuit for working memory. We found that the macaque model exhibits resilience against distractors, a prerequisite for normal working memory function. By contrast, the marmoset model predicts a sensitivity to distractibility commonly observed behaviorally in this species. Surprisingly, this contrasting trend can be swapped by scaling intrafrontal and frontoparietal, and offer a credible prediction to the marmosets behavior in this specific task.

neuroscience↗

bioRxiv2025

A Neural Circuit Framework for Economic Choice: From Building Blocks of Valuation to Compositionality in Multitasking

Value-guided decisions are a cornerstone of cognition, yet the underlying circuit-level mechanisms remain elusive. We used reinforcement learning to train recurrent neural network models endowed with Dales law on a battery of economic choice tasks, which revealed a two-stage computational framework. First, value estimation occurs at input level where learned weights store subjective preferences and approximate the nonlinear multiplication of reward magnitude and probability to yield expected values. This feedforward mechanism enables generalization to novel choice options. Second, option values are compared within the recurrent network, where specific connectivity patterns mediate robust winner-take-all decisions, with both excitatory and inhibitory neurons exhibiting value and choice selectivity. By training a single network on multiple tasks, we show compositional representations combining a shared computational schema with specialized neural modules. Reproducing key neurophysiological findings from the primate orbitofrontal cortex, our model unifies value computation, comparison, and generalization into a coherent framework with testable predictions. IN BRIEFBattista et al. use biologically plausible RNNs to uncover circuit mechanisms of economic choice. They propose a two-stage framework where feedforward inputs compute offer values and recurrent inhibition drives comparison. This architecture explains how the brain generalizes preferences and multitasks using compositional neural codes, offering a unified theory of decisionmaking. HIGHLIGHTSO_LIBiologically plausible RNN reveals a two-stage economic choice framework C_LIO_LIInput weights store preferences and multiply features to compute offer value C_LIO_LICompetitive recurrent inhibition mediates winner-take-all comparison C_LIO_LIMultitasking relies on compositional shared and specialized neural modules C_LI

neuroscience↗

bioRxiv2025

Neural representation of action symbols in primate frontal cortex

A hallmark of intelligence is proficiency in solving new problems, including those that differ dramatically from problems seen before. Problem-solving, in turn, depends on goal-directed generation of novel ideas and behaviors1, which has been proposed to rely on internal representations of discrete units, or symbols, and processes that recombine these units into a large set of possible composite representations1-9. Although this view has been influential in formulating cognitive-level explanations of behavior, definitive evidence for a neuronal substrate of symbols has remained elusive. Here, we identify a neural population encoding action symbols--recombinable representations of discrete units of motor behavior--localized to a specific area of frontal cortex. In macaque monkeys performing a drawing-like task, we found behavioral evidence that action elements (strokes) exhibit three critical features indicating an underlying symbolic representation: (i) invariance over low-level motor parameters; (ii) categorical structure, reflecting discrete types of action; and (iii) recombination into novel sequences. In simultaneous neural recordings across motor, premotor, and prefrontal cortex, we found that planning-related population activity in ventral premotor cortex (PMv) encodes actions in a manner that, like behavior, reflects motor invariance, categorical structure, and recombination. Activity in no other recorded area exhibited these three properties of symbols. These findings reveal a neural representation of action symbols localized to PMv, and therefore identify a putative neural substrate for symbolic cognitive operations.

neuroscience↗

bioRxiv2025

Inhibitory and disinhibitory VIP IN-mediated circuits in neocortex

Cortical GABAergic interneurons (INs) expressing the neuropeptide vasoactive-intestinal peptide (VIP) predominantly function by inhibiting dendritic-targeting somatostatin (SST) expressing INs, thereby disinhibiting pyramidal cells (PCs) and facilitating cortical circuit plasticity. VIP INs are a molecularly heterogeneous group, but the physiological significance of this diversity is unclear at present. Here, we have characterized the functional diversity of VIP INs in the primary somatosensory cortex (vS1) using intersectional genetic approaches. We found that VIP INs are comprised of four primary populations that exhibit different laminar distributions, axonal and dendritic arbors, intrinsic electrophysiological properties, and efferent connectivity. Furthermore, we observe that these populations are differentially activated by long-range inputs, and display distinct responses to neuromodulation by endocannabinoids, acetylcholine and noradrenaline. Stimulation of VIP IN subpopulations in vivo results in differential effects on the cortical network, thus providing evidence for specialized modes of VIP IN-mediated regulation of PC activity during cortical information processing.

neuroscience↗

bioRxiv2024

In Search of Transcriptomic Correlates of Neuronal Firing-Rate Adaptation across Subtypes, Regions and Species: A Patch-seq Analysis

Can the transcriptomic profile of a neuron predict its physiological properties? Using a Patch-seq dataset of the primary visual cortex, we addressed this question by focusing on spike rate adaptation (SRA), a well-known phenomenon that depends on small conductance calcium (Ca)-dependent potassium (SK) channels. We first show that in parvalbumin-expressing (PV) and somatostatin-expressing (SST) interneurons, expression levels of genes encoding the ion channels underlying action potential generation are correlated with the half-width (HW) of spikes. Surprisingly, the SK encoding genes are not correlated with the degree of SRA (dAdap). Instead, genes that encode proteins upstream from the SK current, as well as the low-voltage-activated Calcium channel, are correlated with dAdap. This general principle is supported by datasets from the mouse primary motor cortex and multiple macaque brain regions. Finally, we generate testable predictions by constructing a minimal model.

neuroscience↗

bioRxiv2024

Cognitive networks interactions through communication subspaces in large-scale models of the neocortex

Neocortex-wide neural activity is organized into distinct networks of areas engaged in different cognitive processes. To elucidate the underlying mechanism of flexible network reconfiguration, we developed connectivity-constrained macaque and human whole-cortex models. In our model, within-area connectivity consists of a mixture of symmetric, asymmetric, and random motifs that give rise to stable (attractor) or transient (sequential) heterogeneous dynamics. Assuming sparse low-rank plus random inter-areal connectivity constrained by cognitive networks activation maps, we show that our model captures key aspects of the cognitive networks dynamics and interactions observed experimentally. In particular, the anti-correlation between the default mode network and the dorsal attention network. Communication between networks is shaped by the alignment of long-range communication subspaces with local connectivity motifs and is switchable in a bottom-up salience-dependent routing mechanism. Furthermore, the frontoparietal multiple-demand network displays a coexistence of stable and dynamic coding, suitable for top-down cognitive control. Our work provides a theoretical framework for understanding the dynamic routing in the cortical networks during cognition.

neuroscience↗

bioRxiv2024

Topography of putative bidirectional interaction between hippocampal sharp wave ripples and neocortical slow oscillations

Systems consolidation relies on coordination between hippocampal sharp-wave ripples (SWRs) and neocortical UP/DOWN states during sleep. However, whether this coupling exists across neocortex and the mechanisms enabling it remain unknown. By combining electrophysiology in mouse hippocampus (HPC) and retrosplenial cortex (RSC) with widefield imaging of dorsal neocortex, we found spatially and temporally precise bidirectional hippocampo-neocortical interaction. HPC multi-unit activity and SWR probability was correlated with UP/DOWN states in mouse default mode network, with highest modulation by RSC in deep sleep. Further, some SWRs were preceded by the high rebound excitation accompanying DMN DOWN[->]UP transitions, while large-amplitude SWRs were often followed by DOWN states originating in RSC. We explain these electrophysiological results with a model in which HPC and RSC are weakly coupled excitable systems capable of bi-directional perturbation and suggest RSC may act as a gateway through which SWRs can perturb downstream cortical regions via cortico-cortical propagation of DOWN states.

neuroscience↗