Neuron

Animals exhibit behavior in the absence of external stimuli or explicit tasks. Is the initiation of such spontaneous behavior shaped by internal brain states in a predictable manner? If so, does it engage specific brain circuits independent of behavioral form? Here, we studied the initiation of uninstructed behaviors of head-fixed mice in two contexts: a virtual burrow and a running wheel. Across both contexts, mice spent most of the time in quiet wakefulness and spontaneously initiated bouts of egress (exiting the burrow), running, or grooming. We employed functional ultrasound imaging (fUS) to record whole-brain activity and to identify whether the initiation of spontaneous behavior could be predicted from hemodynamic signals. We first identified distinct hemodynamic patterns associated with each behavior and subsequently performed time-resolved decoding to predict behavioral transitions from fUS data. We found that whole-brain hemodynamic signals could decode spontaneous egress and running around 10 seconds before their onset, a timescale that cannot be accounted for by preceding behavioral changes alone. Furthermore, we found a network of regions, including the medial septum (MS), that decreased their signal several seconds before the onset of egress and running. Mimicking this decrease by inhibiting neurons in the MS via optogenetics increased the probability of egress, running, and grooming. Through this unbiased approach, our work sheds light on a whole-brain transition-prone state that precedes uninstructed behavior transitions.

Microscale biophysical alterations in neuronal dynamics can have profound implications for macroscale pathological outcomes in the brain. Despite the critical need to link neuronal perturbations to large-scale disease manifestations, few studies successfully bridge these hierarchical scales. Here, we bridge microscale biophysical variability within neuronal dynamics to macroscale disease-related phenotypes. We find that Drosophila models expressing tauopathy- and epilepsy-associated molecular mutations exhibit increased dynamic instability in the timing of action potential initiation, and microscale biophysical changes are manifested at the level of the macroscale global brain state. We show that variability in voltage-gated sodium channel currents during non-stationary channel inactivation may act as a microscale biophysical contributor to the increased dynamic instability observed in action potential timing. We also find that treatment with antiepileptic drugs stabilizes neuronal dynamics by modulating this variability in voltage-gated sodium channel currents. Finally, we show that neurons derived from human induced pluripotent stem cells (iPSCs) from patients with Alzheimers disease and epilepsy exhibit analogous dynamic instability, which is reversible by administration of antiepileptic medications. Our results highlight how subtle microscale neuronal instabilities propagate and are amplified to produce macroscopic pathological phenotypes, providing new biophysical insights into neurological disorders and potential strategies for therapeutic intervention. Significance StatementLinking microscale neuronal changes to macroscale disease phenotypes remains a key challenge in neuroscience biophysics. Here, we show that subtle biophysical instability, such as variability in action potential timing and increased noise in voltage-gated sodium channel activity, can destabilize neuronal network integrity and cause systemic pathology. Stabilizing neuronal dynamics with antiepileptic drugs reverses tau-induced instabilities in a Drosophila disease model. Similar neuronal instabilities occur in fly neurons expressing epilepsy-linked sodium channel mutations and in human iPSC-derived neurons from Alzheimers and epilepsy patients, revealing a shared cellular mechanism. These findings highlight that targeting microscale instabilities may offer a unifying therapeutic approach for complex neurological disorders.

Neuromodulators powerfully shape circuit function via mechanisms that largely remain to be determined. As arguably the best-described cortical model system in neuroscience, the primary visual cortex (V1) of the rhesus monkey is an ideal model system to ask and answer these questions. There has been particular focus on acetylcholine actions in V1, where there is strong innervation from the basal forebrain and well-described local cholinergic anatomy in which {beta}2 subunit-containing (high affinity) nicotinic receptors are strongly expressed on thalamocortical axons arriving in layer 4C. Activating these receptors strongly enhances local gain, but it is not known to what extent these effects propagate to other V1 layers. To determine the magnitude and direction (enhancement vs suppression) of gain effects outside layer 4C following gain injection via the nicotinic receptors on thalamic axons, we recorded across the cortical depth in V1 while selectively delivering nicotine locally to layer 4C. We observed widespread and heterogeneous gain effects across layers that could not be explained by visual stimulus conditions, but were well predicted by an adaptation of a normalization model. These gain changes were not compensated by the circuit, and thus were evident during performance of a perceptual task.

Value-based decision making requires maintaining and comparing multiple option representations associated with different values. Then, the chosen option must be communicated to downstream regions to drive behavior. The neural circuits involved in value-based decisions are increasingly well understood, but less understood is how decisions shape option representations. We recorded from lateral prefrontal cortex (LPFC), a central hub for transforming options into actions, while non-human primates (NHPs) held two sequentially presented option-value pairs (spatial targets and abstract reward cues) in working memory, then chose the option with the higher value. This revealed how decisions dynamically reorganize option representations in LPFC. We found that, once decisions could be made, representations of chosen and unchosen options rotated into orthogonal subspaces and the chosen option representation was expanded. Before decisions, the first- and second-presented options were maintained separately. After decisions, the chosen option was rotated into a subspace with a consistent representation of the prescribed action, regardless of its presentation order. This suggests a mechanism for value-based decisions where the decision drives a neural subspace reorganization that facilitates selective and efficient readout of the chosen action.

The ventrolateral prefrontal cortex (vlPFC) is well known for its involvement in high-level functions such as cognitive control and language. However, vlPFCs role in visual processing is less clear. Here, we investigated how neuronal ensembles in the vlPFC dynamically encode different types of visual information. Using chronic recording of spiking activity, we investigated vlPFCs representational geometry in a macaque monkey passively viewing a large set of naturalistic images, and compared this to representations in deep neural networks (DNNs). We found that the vlPFC processes visual information in two stages. First, an "early" response from 50 to 90 ms after stimulus onset encodes the low spatial frequency component of an image. It contains sufficient information to form a coarse estimate of the position and category of a salient object. Then, from 100 ms on, the representational geometry changes and contains much richer information. This late period contains non-categorical information typically present in conscious experiences such as the orientation of a face and natural scenes in the background. The late window also enables sub-category identification, which is boosted by the low spatial category prior. These results suggest that the vlPFC has a dual role in natural vision: first forming fast low-spatial-frequency-based priors shaping feed-forward visual processing, and subsequently maintaining a detailed and rich representation of a visual scene. SignificanceWhat information does the prefrontal cortex represent during natural vision, and how does this relate to conscious experience? Theories of consciousness differ sharply on whether prefrontal activity reflects detailed perceptual content or only high-level, task-related information. Using dense neural recordings during passive viewing of thousands of naturalistic images, we show that the ventrolateral prefrontal cortex processes visual information in two distinct stages: an early, coarse estimate of the visual scene is followed by a richer, high-dimensional representation that includes sub-category identity and other perceptual details. These findings reveal that prefrontal circuits encode more of the content of visual experience than previously assumed, even in the absence of a task.

Cholinergic interneurons (CINs) in the nucleus accumbens (NAc) play a key role in regulating motivated behaviors. Here we examine the connectivity and functional impact of cortical and thalamic inputs onto CINs in the NAc core. We first use cell-type specific retrograde anatomy to identify the prefrontal cortex and thalamus as putative afferents. We then combine ex vivo slice physiology and optogenetics to characterize the properties of synapses onto CINs. We demonstrate that thalamic inputs strongly facilitate, whereas cortical inputs exhibit marked depression. We also show that a combination of AMPA and NMDA receptors contribute to both cortical and thalamic responses. Lastly, we establish how these inputs and receptors evoke action potentials and influence spontaneous firing. Our findings show how CINs in the NAc core process long-range inputs, highlighting differences from equivalent circuits in other parts of striatum. SIGNFICANCE STATEMENTCholinergic interneurons provide the primary source of acetylcholine in the striatum and are important for behavior and disease. The types of afferents that drive these interneurons have been examined in dorsal striatum but remain understudied in the nucleus accumbens. We found that inputs from prefrontal cortex and thalamus are the main drivers in the mouse nucleus accumbens core. We compare the sign, dynamics, and impact of these two excitatory inputs, showing how they engage multiple glutamate receptors to influence cholinergic interneuron firing.

Remembering our environment and its principal features is essential to our survival. The anterior cingulate cortex (ACC) has been implicated as a key region in remote memory (Frankland et al., 2004; Goshen et al., 2011; Almaguer-Melian et al., 2025). Whilst most studies investigating the ACCs contribution to memory have used fear conditioning, some have highlighted a role in the retention of object locations across various timescales, both recent and remote (Weible et al., 2009, 2012). However, how neurons in the ACC encode object locations across repeated exposure has not been investigated. Here, we investigated if object representations are supported by cell assemblies that are stable over days, or if the representation is volatile and dynamic across repeated experiences separated across time. Using calcium-imaging in freely moving mice, we recorded the activity of excitatory ACC neurons while mice explored objects placed in an environment over repeated days. We find that the ACC encodes object location in a mostly dynamic fashion: while the proportion of neurons allocated to object coding does not change across days, the specific neurons exhibiting object correlates fluctuate, featuring a dynamic turnover with a smaller set of stable cells. Interestingly, this was modulated by the animals behaviour, such as object cells from mice spending the most time exploring the objects showed a higher degree of stability. We next examined how dynamic single-cell coding relates to stability at the network level. Population analyses revealed stable representations emerging from collective dynamics, suggesting that downstream regions may rely on ensemble patterns rather than fixed cell identities. Decoding analyses supported this view: ensembles of 64-128 neurons were as accurate and more efficient than the full dataset, indicating that information about the animals location becomes more linearly separable when represented at a coarser population scale, making it more readily accessible to downstream regions that integrate population-level activity. Thus, we show that the ACC achieves stability through emergent organization across neurons, even as individual cells remain dynamic.

Segmenting mnemonic episodes from continuous experience is a key aspect of human episodic memory. The brain constantly forms predictions about what will happen next based on previous experience and knowledge, and prediction errors are thought to signal when a new event begins (cognitive boundaries). Dopamine has been closely linked to prediction error signals, yet it remains unknown how human midbrain neurons are modulated by cognitive boundaries and how their responses influence memory. To address these questions, we recorded activity of individual neurons in the human substantia nigra, a critical brain structure for dopamine production and regulation, while participants undergoing deep brain stimulation surgery watched a series of clips embedded with cognitive boundaries and performed a recognition memory task. We found that neural activity in the substantia nigra was robustly modulated by cognitive boundaries during clip viewing. Moreover, a subset of these boundary-responsive neurons also differentiated novel from familiar images during recognition, and their firing rates were indicative of participants memory success. These findings reveal that neurons in the human substantia nigra carry boundary- and novelty-related signals consistent with prediction error mechanisms that influence the encoding and retrieval of episodic memories.

Adaptive social behavior requires balancing self-interest with the welfare of others, a core axiom of social decision-making that determines whether actions are selfish or prosocial. Although the medial prefrontal cortex (mPFC) has been implicated in prosocial behavior, the broader cortical-subcortical networks that arbitrate between selfish and prosocial actions remain poorly understood. Moreover, most studies of social decision-making (at both the single-region and circuit levels) have focused on inbred C57BL/6 mice, leaving unclear whether similar neural mechanisms operate across genetically diverse populations. Here, we combined a social decision-making task with c-Fos mapping to examine activity across distributed cortical and subcortical regions in inbred C57BL/6 and outbred CD1 male mice during prosocial and selfish choices. We found that CD1 mice exhibited a stronger bias toward selfish behavior, whereas C57BL/6 mice were more prosocial. This behavioral divergence was associated with elevated c-Fos activity in the mPFC and nucleus accumbens core (NAcC) in CD1 mice compared with C57BL/6 mice, and mPFC activity positively correlated with selfish choice bias. At the network level, social decision-making selectively recruited coordinated activity among the distinct mPFC subregions, ventral tegmental area (VTA), and NAcC. Importantly, prosocial and selfish individuals recruited distinct prefrontal-subcortical network configurations during social decision-making. Together, these findings identify distributed cortical-subcortical network dynamics underlying social choice bias and reveal strain-dependent differences in the neural architecture supporting prosocial and selfish behavior. Significant statementSocial decisions require weighing personal benefit against the welfare of others, yet the neural circuits that bias individuals toward selfish versus prosocial choices remain poorly understood. Here, we show that two mouse strains with opposing social preferences recruit distinct cortical and subcortical network configurations during social choice, despite performing the same task. Rather than reflecting differences in single brain regions, social decision-making engaged coordinated activity across a prefrontal-striatal-midbrain circuit, with prosocial and selfish individuals recruiting different versions of this network. These findings reveal that social choice bias is encoded at the level of distributed circuit organization and that genetic background shapes how the brain implements social decisions.

Recognizing individuals or objects across different contexts is a hallmark of primate cognition. Beyond recognition, the brain organizes visual stimuli into meaningful categories, supporting efficient perception and adaptive behavior. Previous studies in the human temporal lobe have identified concept cells supporting invariant representations of specific stimuli. However, invariant coding alone cannot account for how categorical knowledge is structured across related visual domains. Here, we analyzed single-neuron recordings from the human amygdala during the presentation of familiar visual stimuli spanning multiple categories. Using a novel data-driven analytic framework that integrates supervised population decoding with single-neuron analyses, we identified a population of cross-category invariant neurons. These neurons preserved exemplar-invariant responses while generalizing across multiple categories linked by shared naturalistic or human/domestic contexts. Our findings demonstrate that the human amygdala supports both invariant and associative forms of categorical representations. By linking stable identity representations across contextually related categories, the amygdala provides a flexible neural substrate for high-level visual organization, supporting efficient perception and adaptive behavior in complex environments. HighlightsO_LIHuman amygdala neurons encode invariant category representations at the single-cell level C_LIO_LICategory representations are organized according to naturalistic and social contexts C_LIO_LICross-category invariant neurons bridge concept-cell and mixed-selectivity coding C_LIO_LICategory encoding relies on opponent patterns of enhancement and suppression across domain C_LI

Neuronal cell death is a hallmark of many neurodegenerative diseases. Effective detection and clearance of cell debris generated during cell death events is essential to prevent a degenerative cascade. Brain resident microglia are responsible for performing these functions through complex cell-cell signaling involving both "find-me" and "eat-me" cues. To examine microglial responses to neuronal cell death in vivo, we investigated neuron/microglia CX3CL1/CX3CR1 signaling using intravital optical imaging in mouse cortex and a single-cell ablation technique called 2Phatal. We find that CX3CL1 aggregates as puncta on microglia and that this pattern is maintained when microglia engulf dying neurons. Additionally, disruption of this signaling via Cx3cr1 deletion when both few and many neurons are dying leads to delayed cell corpse clearance, partly due to a delay in microglial engagement with the dying cells. Overall, our work uncovers a precise role for CX3CL1/CX3CR1 signaling in regulating the microglial response to dying neocortical neurons.

Stable and flexible neural representations of space in the hippocampus are crucial for navigating complex environments. However, how these distinct representations emerge from the underlying local circuit architecture remains unknown. Using two-photon imaging of CA3 subareas during active behavior, we reveal opposing coding strategies within specific CA3 subregions, with proximal neurons demonstrating stable and generalized representations and distal neurons showing dynamic and context-specific activity. To test the causal role of excitatory connectivity in neural computation, we employed a genetic manipulation approach in which local disruption of glutamatergic synaptic transmission impaired context-specific spatial coding in distal CA3. Aligned with these experimental results, we show in artificial neural network models that varying the recurrence level causes these differences in coding properties to emerge. We confirmed the contribution of recurrent connectivity to functional heterogeneity by characterizing the representational geometry of neural recordings and comparing it with theoretical predictions of neural manifold dimensionality. Our results indicate that local circuit organization, particularly recurrent connectivity among excitatory neurons, plays a key role in shaping complementary spatial representations within the hippocampus. HighlightsO_LIProximodistal CA3 subregions implement complementary coding strategies in relation to time and context C_LIO_LIDisrupting excitatory recurrence in distal CA3 impairs context-discriminative neural coding C_LIO_LISparse and dense neural networks capture the functional heterogeneity of CA3 subcircuits C_LIO_LIRecurrence level tunes the geometry of neural manifolds both in vivo and in silico C_LI

How does cortical connectivity support decision-making? Behavioral tasks often involve multiple sequential phases that implement different computations. For perceptual decision-making during navigation these can include evidence accumulation, decision commitment, and motor program read out. How are these different phases implemented in circuits with fixed anatomical synaptic connectivity? One potential contribution is that the connectivity of neurons is modulated in the different phases, but this has never been tested. Here we used an all-optical method to probe the causal connectivity of excitatory neurons in layer 2/3 of mouse retrosplenial cortex during different behavioral epochs of a navigation-based decision-making task, as well as in the absence of the task. In-task connectivity was different from no-task connectivity: furthermore, these differences were selective to the cue / decision phase, tapering off in later stages of the task. We propose that fast modulation of connectivity is a prevalent mechanism in neural circuit function.

Many stroke survivors with aphasia struggle to map their thoughts into words or motor plans. Neuroprostheses that decode concept representations could help these individuals communicate by predicting the words, phrases, or sentences that they are struggling to produce. Here we decoded concept representations measured using functional magnetic resonance imaging (fMRI) from participants with different aphasia profiles. The decoders generated continuous word sequences that could describe the concepts that the participants were hearing about, seeing, or imagining. To forecast how this approach would generalize across the heterogeneity of aphasia profiles, we characterized how stroke affects the anatomical organization and information capacity of conceptual processing. Mapping how concepts are organized across the brain, we found that conceptual tuning during non-linguistic processing was largely consistent between the participants with aphasia and neurologically healthy participants. Comparing information processing between the participants with aphasia and neurologically healthy participants, we found that both groups processed similar amounts of non-linguistic information. Our findings indicate that concept representations can be largely spared in individuals with aphasia and demonstrate how these representations can be decoded to support communication.

Transcranial magnetic stimulation (TMS) to the dorsolateral prefrontal cortex (dlPFC) is an FDA-cleared treatment for depression, yet how cortical stimulation influences single neurons in deep brain circuits remains unknown. Using intracranial microelectrode recordings in four neurosurgical patients, we resolved single-neuron spikes as early as 8 ms from 185 single neurons after single-pulse left dlPFC TMS. TMS elicited time-locked firing responses in 46% of neurons across deep cortical and subcortical structures bilaterally. TMS facilitated putative interneuron spiking in striato-thalamic regions from [~]8 ms, peaking at [~]80-100 ms, and lasting to [~]1000 ms, while suppressing putative pyramidal cell spiking with a delayed and slower time course. Trial-by-trial single-neuron modulations were positively correlated with cortico-striato-thalamic network activity and anti-correlated with limbic network activity. These findings reveal that dlPFC TMS facilitates inhibitory firing in executive control networks while suppressing limbic excitatory drive, providing a cellular mechanism for how cortical stimulation modulates distributed brain networks.

Hippocampal pyramidal neurons function as place cells, showing location-specific activity during navigation, to form an internal spatial map of the environment. They are hypothesized to be the neural substrate of episodic memory. However, place cell receptive fields tend to drift or have poor tuning in low demand tasks, lacking operant goals such as random foraging, or in sensory context-deprived environments. Through chronic two-photon calcium imaging of hippocampal area CA1, we directly compare stability in a low versus a high demand task within the same animals over the course of learning and recall in the same environment. We find that compared to random foraging, an odor-context based navigational task stabilizes place cell representations and increases place cell quality and quantity. To investigate the circuit mechanism that may support this stability, we manipulated the activity of lateral entorhinal cortex (LEC) excitatory neurons, which provide both indirect and direct multisensory inputs about context, odor, and time to CA1. We chemogenetically suppressed activity of excitatory neurons in LEC during recall of the odor-context based navigation task and found that context discrimination is impaired at both the behavioral and neural level. With LEC silencing, mice had lower behavioral performance, less stable population activity, and greater similarity between opposing trial types. Our study finds that increasing task demand increases CA1 stability and that this stability is partially supported by LEC.

The neural signature of rhythm and tempo remains difficult to study in both humans and non-human primates. Here we recorded from the motor cortex of human participants implanted with intracortical microelectrode arrays while they performed a series of rhythmic tapping tasks. We found that rhythmic tapping elicited low-dimensional rotational neural dynamics whose radii varied in a tempo-dependent manner and axes related to kinematic properties. Moreover, we observed a spectrum of kinematic and neural behavior as participants shifted from low tempo punctuated taps to high tempo smoother, continuous taps. Surprisingly, we observed that tactile feedback strengthened the rotational dynamics despite reduced kinematic range. Moreover, while tempo preparation did not produce dynamics of their own, motor cortex encoded it in an orthogonal dimension. Finally, we found that switching tempos was achieved with smooth neural transitions that could only be separated in higher dimensions. These results show that motor cortex directly encodes a multitude of rhythm related features.

Spatial attention is often partitioned into endogenous, exogenous, and social forms, yet it remains unclear whether a single neural circuit can support all three and how their population codes are organized. Here we trained recurrent artificial neural networks (ANNs) with convolutional sensory front-ends on three classic cueing paradigms (central, peripheral, and gaze cues) to reproduce human reaction time (RT) profiles across cue-target onset asynchronies. Despite differences in sensory architecture and visual experience, all ANNs captured the full facilitation-inhibition time course for all three attention types. Model-based targeted dimensionality reduction (mTDR) revealed that cue- and choice-related activity in the advanced cognitive module evolved as rotations within a shared low-dimensional manifold, with angular deflections that mirrored the distinct temporal dynamics of endogenous, exogenous, and social attention. Attentional signals were encoded by highly sparse, distributed population activity: a small subset of recurrent units explained most task-related variance, was sufficient to recover human-like RT patterns after virtual lesioning, and became progressively sparser as training improved performance. At the same time, single unit responses displayed pervasive mixed selectivity, dominated by nonlinear conjunctions of cue type, cue direction, and validity, whose strength and heterogeneity robustly predicted model performance. Together, these results identified low-dimensional geometric rotations, sparse coding, and nonlinear mixed selectivity as core computational principles through which a single recurrent circuit could generate human-like temporal dynamics across endogenous, exogenous, and social orienting, and provided testable predictions for population-level mechanisms of spatial attention in the brain.

The hippocampus and the prefrontal cortex interact across a range of cognitive functions, yet how these regions represent space and how these representations evolve across learning remains poorly understood. We recorded large-scale neural activity from the hippocampus (CA1) and the medial prefrontal cortex (mPFC) as rats acquired proficiency in radial maze tasks over weeks. We identified a form of trial-by-trial flickering in which one hippocampal place field gradually overtook the other across successive trials, whereas mPFC firing fields switched randomly. While hippocampal flickering decreased as the task was mastered, mPFC flickering remained random and persisted throughout learning. Population-level UMAP embeddings revealed that mPFC states transitioned smoothly across trials, with the within-session representational drift stabilizing only after a two-week period. These findings suggest that while the hippocampus stabilizes its spatial maps during learning, the mPFC maintains a flexible, flickering representation that facilitates the extraction of abstract task structure and the rapid adaptation required for behavioral flexibility.

Nicotinic acetylcholine receptors (nAChRs) facilitate striatal dopamine transmission but also suppress dopamine release during high-frequency stimulation, suggesting they act as as low pass filters of dopamine release. Because axonal excitability is key a determinant of transmission, we combined axonal recordings and calcium imaging to define the physiological conditions under which nAChRs bidirectionally control dopaminergic axons. Activation of cholinergic interneuron (CINs) recruited nAChRs to enhance dopaminergic axon signals under moderate activation but suppressed signals after strong high-frequency stimulation. Axonal recordings revealed that single-pulse striatal stimulation triggered a rapid ([~]125 Hz) burst of 2-3 nAChR-driven spikes in dopaminergic axons followed by a brief refractory period that inhibited further axon spiking. In sum, we show that nAChRs mainly enhance local excitability of striatal dopaminergic axons but also trigger axonal bursting that suppresses axonal excitability. This mechanism expands the computational power of dopaminergic axons and explains the apparent nAChR-mediated low-pass filtering of dopamine release.