Biology subjects

Carandini, M.

Publications and source records attributed to Carandini, M..

At least 19 recordsLinked to original sources

bioRxiv2026

Brainwide representation of navigation

Tracking your position in the environment during navigation requires integrating multiple signals into a spatial representation. This process is thought to occur primarily in the hippocampal formation, but evidence for it has also been observed in other regions. To investigate it across the brain, we recorded from over 20,000 neurons in mice navigating a virtual corridor that decoupled position from correlated signals. Spatial position was encoded in every region, and was more common in neurons tuned to visual landmarks. Most neurons brainwide were also modulated by running speed, likely reflecting arousal, and many by rewards. The hippocampal formation represented spatial position slightly more uniformly than other regions, but less precisely than visual cortex. Spatial position and other navigational signals are thus represented widely across the brain.

neuroscience↗

bioRxiv2026

A causal spatial role for apical dendrites in the cortex

Cortical pyramidal neurons project a long apical dendrite to layer 1 (L1), to receive synaptic inputs that are thought to provide contextual and possibly suppressive modulation. In visual cortex, these inputs are thought to drive or suppress the responses to large stimuli. Here, we demonstrate a causal role for individual apical dendrites in the awake cortex. We performed two-photon assisted dendritic pruning in individual L5 neurons of the mouse primary visual cortex and discovered that apical dendrites do not suppress the responses to large stimuli; rather, they drive the responses of neurons that prefer such stimuli. To establish the synaptic basis of this effect, we imaged the excitatory synaptic inputs targeting the apical dendrite. We found that these inputs cover an extensive region, which is largest in neurons that prefer large stimuli. These results reveal a clear spatial role for apical dendrites in the cortex. HighlightsO_LIOptical dendritic pruning in V1 reveals a causal role for L5 apical dendrites C_LIO_LIApical dendrites drive responses in neurons that prefer large visual stimuli C_LIO_LIImaging of synaptic inputs to apical dendrites reveals extended apical receptive field C_LIO_LIThe apical receptive field is larger in L5 neurons that prefer large visual stimuli C_LI

neuroscience↗

bioRxiv2026

The mouse superior colliculus promotes competing actions independently of sensory inputs

In many circumstances the brain makes decisions according to logistic classification: summing weighted factors in favor of each option and choosing an option according to a softmax function. Here we find that the superior colliculus (SC) contributes additively to this computation, promoting competing options independently of stimuli. We trained mice to locate an audiovisual stimulus on their left or right, and found that SC encodes the auditory, visual, and action signals in largely separate neurons in different layers. Unilateral SC inactivation did not affect sensory sensitivity but reduced contralateral actions, mostly in favor of inaction. The effects of bilateral inactivation were additive, with balanced inactivations restoring left-right balance. A simple 3-option logistic classification model captured the results, with sensory and prefrontal cortex processing sensory signals and each side of SC additively promoting the contralateral action while slightly opposing the ipsilateral action. The SC thus simultaneously promotes all competing eligible actions.

neuroscience↗

bioRxiv2026

The basal ganglia transform visual identity into behavioral relevance

The basal ganglia are thought to be essential for transforming sensory signals into appropriate actions, yet it remains unclear how this transformation unfolds across their nuclei. We recorded neuronal activity across the striatum, globus pallidus external (GPe), and substantia nigra pars reticulata (SNr) in mice that were either naive or trained in visuomotor associations. In naive mice, visual stimuli drove responses in all three nuclei. These responses were feature-selective in the striatum but less so in GPe and SNr. Task training potentiated visual responses across the circuit, particularly for stimuli associated with movement. Training affected nuclei differently: while the striatum continued to distinguish between stimuli associated with the same behavioral action, GPe and SNr did not, responding similarly to all "Go" cues while distinguishing them from "No-Go" cues. These results demonstrate that the basal ganglia progressively transform high-dimensional sensory representations into low-dimensional signals of behavioral relevance.

neuroscience↗

bioRxiv2025

Invariant activity sequences across the mouse brain

Given the variety and plasticity of their circuits, different brain regions could potentially produce profoundly diverse patterns of population activity. Alternatively, the population activity could exhibit canonical patterns that are similar across the brain and stable over time. Here we show that a pattern observed in the cerebral cortex, where neurons fire in 10-100 ms sequences, is stable across weeks and ubiquitous across the brain. In mouse visual cortex, neurons respond to visual stimuli with fixed latency, forming a sequence. Similar sequences appeared during spontaneous activity: each neuron was coupled to the population with a characteristic strength and delay, which remained stable for weeks. Brainwide recordings revealed such sequences in every brain region. These results reveal a stable scaffold for neural activity across the brain.

neuroscience↗

bioRxiv2025

Characterizing neuronal population geometry with AI equation discovery

The activity of visual cortical neurons forms a population code representing image stimuli. There is, however, a discrepancy between our understanding of this code at the single-cell and population levels: direct measurements indicate the population code is high-dimensional, but established models of single-cell tuning give rise to low-dimensional codes. We reconciled this discrepancy by developing an AI science system to find a new parsimonious, interpretable equation for visual cortical orientation tuning. Candidate equations were expressed as short computer programs and evolved by Large Language Models (LLMs) using graphical diagnostics. The resulting equation not only improved single-cell fits, but also accurately modelled the population codes high-dimensional geometry. A novel parameter of the AI-discovered equation, which controls single-cell tuning smoothness, gives rise to high-dimensional population codes. The same parameter drives high-dimensional coding in head-direction cells, suggesting a common coding strategy across brain regions. We used this equation to hypothesize a circuit mechanism generating high-dimensional population codes, and to demonstrate the advantages of these codes in a simulated hyperacuity task. These results show that tuning smoothness has a key role in controlling population code geometry, and demonstrate how AI equation discovery can deliver interpretable models accelerating scientific understanding in neuroscience and beyond.

neuroscience↗

bioRxiv2025

Mapping the visual cortex with Zebra noise and wavelets

Studies of the early visual system often require characterizing the visual preferences of large populations of neurons. This task typically requires multiple stimuli such as sparse noise and drifting gratings, each of which probe only a limited set of visual features. Here we introduce a new dynamic stimulus with sharp-edged stripes called Zebra noise and a new analysis model based on wavelets, and show that in combination they are highly efficient for mapping multiple aspects of the visual preferences of thousands of neurons. We used two-photon calcium imaging to record the activity of neurons in the mouse visual cortex. Zebra noise elicited strong responses that were more repeatable than those evoked by traditional stimuli. The wavelet-based model captured the repeatable aspects of the resulting responses, providing measures of neuronal tuning for multiple stimulus features: position, orientation, size, spatial frequency, drift rate, and direction. The method proved efficient, requiring only 5 minutes of stimulus (repeated 3 times) to characterize the tuning of thousands of neurons across visual areas. In combination, the Zebra noise stimulus and the wavelet-based model provide a broadly applicable toolkit for the rapid characterization of visual representations, promising to accelerate future studies of visual function.

neuroscience↗

bioRxiv2025

Different representations in layer 2/3 and layer 5 excitatory neurons of the primary visual cortex

The cortex contains multiple types of excitatory neuron, differentiated primarily by their layer of residence. We recorded from neuronal populations in the mouse visual cortex using 2-photon calcium imaging or Neuropixels probes, and found that excitatory neurons in layer 2/3 (L2/3) and layer 5 (L5) differed in their encoding of visual vs. nonvisual signals, with L2/3 more strongly modulated by visual stimuli and L5 more strongly modulated by movement. Movement had opposite effects on population synchrony in the two layers: it desynchronizes L2/3, by abolishing spontaneous synchronous fluctuations that entrain that population, and synchronizes L5, where excitatory cells are less entrained by spontaneous synchronous fluctuations and more strongly correlated with movement itself. Spontaneous activity was lower-dimensional in L2/3 than L5, with L2/3 population activity dominated by a single dimension of overlap between spontaneous and stimulus-evoked subspaces. We conclude that excitatory neurons in different layers of the visual cortex carry different representations of visual and non-visual signals.

neuroscience↗

bioRxiv2025

Brainwide blood volume reflects opposing neural populations

The supply of blood to brain tissue is thought to depend on the overall neural activity in that tissue, and this dependence is thought to differ across brain regions and across brain states. Studies supporting these views, however, measured neural activity as a bulk quantity, and related it to blood supply following disparate events in different regions. Here we measure fluctuations in neuronal activity and blood volume across the mouse brain, and find that their relationship is consistent across brain states and brain regions but differs in two opposing brainwide neural populations. Functional Ultrasound Imaging (fUSI) revealed that whisking, a marker of arousal, is associated with brainwide fluctuations in blood volume. Simultaneous fUSI and Neuropixels recordings showed that neurons that increase vs. decrease activity with whisking have distinct hemodynamic response functions. Their summed contributions predicted blood volume across states. Brainwide Neuropixels recordings revealed that these two opposing populations coexist in the entire brain. Their differing contributions to blood volume largely explain the apparent differences in blood volume fluctuations across regions. The mouse brain thus contains two neural populations with opposite relation to brain state and distinct relationships to blood supply, which together account for brainwide fluctuations in blood volume.

neuroscience↗

bioRxiv2025

High-dimensional neuronal activity from low-dimensional latent dynamics: a solvable model

Computation in recurrent networks of neurons has been hypothesized to occur at the level of low-dimensional latent dynamics, both in artificial systems and in the brain. This hypothesis seems at odds with evidence from large-scale neuronal recordings in mice showing that neuronal population activity is high-dimensional. To demonstrate that low-dimensional latent dynamics and high-dimensional activity can be two sides of the same coin, we present an analytically solvable recurrent neural network (RNN) model whose dynamics can be exactly reduced to a low-dimensional dynamical system, but generates an activity manifold that has a high linear embedding dimension. This raises the question: Do low-dimensional latents explain the high-dimensional activity observed in mouse visual cortex? Spectral theory tells us that the covariance eigenspectrum alone does not allow us to recover the dimensionality of the latents, which can be low or high, when neurons are nonlinear. To address this indeterminacy, we develop Neural Cross-Encoder (NCE), an interpretable, nonlinear latent variable modeling method for neuronal recordings, and find that high-dimensional neuronal responses to drifting gratings and spontaneous activity in visual cortex can be reduced to low-dimensional latents, while the responses to natural images cannot. We conclude that the high-dimensional activity measured in certain conditions, such as in the absence of a stimulus, is explained by low-dimensional latents that are nonlinearly processed by individual neurons.

neuroscience↗

bioRxiv2025

Suite3D: Volumetric cell detection for two-photon microscopy

In two-photon imaging of neuronal activity it is increasingly common to acquire 3-dimensional volumes. However, these volumes are typically processed plane by plane, leading to uncorrected axial movement, duplicated cells across planes, reduced signal-to-noise ratio per cell, and missed cells. To overcome these limitations, we introduce Suite3D, a volumetric cell detection pipeline. Suite3D corrects for rigid and non-rigid 3D brain motion, both lateral and axial. It detects neurons using 3D correlation, improving detectability. Finally, it performs 3D segmentation, identifying cells across imaging planes. We validated Suite3D with data from conventional multi-plane microscopes and advanced volumetric microscopes, at multiple resolutions and in multiple brain regions, and with ground truth anatomical labelling. Suite3D successfully detected cells appearing on multiple imaging planes, improving cell detectability and signal quality, avoiding duplications, and running faster than a prior volumetric pipeline. Suite3D offers a powerful solution for analyzing volumetric two-photon data.

neuroscience↗

bioRxiv2025

Glutamate indicators with increased sensitivity and tailored deactivation rates

Identifying the input-output operations of neurons requires measurements of synaptic transmission simultaneously at many of a neurons thousands of inputs in the intact brain. To facilitate this goal, we engineered and screened 3365 variants of the fluorescent protein glutamate indicator iGluSnFR3 in neuron culture, and selected variants in the mouse visual cortex. Two variants have high sensitivity, fast activation (< 2 ms) and deactivation times tailored for recording large populations of synapses (iGluSnFR4s, 153 ms) or rapid dynamics (iGluSnFR4f, 26 ms). By imaging action-potential evoked signals on axons and visually-evoked signals on dendritic spines, we show that iGluSnFR4s/4f primarily detect local synaptic glutamate with single-vesicle sensitivity. The indicators detect a wide range of naturalistic synaptic transmission, including in the vibrissal cortex layer 4 and in hippocampal CA1 dendrites. iGluSnFR4 increases the sensitivity and scale (4s) or speed (4f) of tracking information flow in neural networks in vivo.

neuroscience↗

bioRxiv2025

Neuropixels Opto: Combining high-resolution electrophysiology and optogenetics

High-resolution extracellular electrophysiology is the gold standard for recording spikes from distributed neural populations, and is especially powerful when combined with optogenetics for manipulation of specific cell types with high temporal resolution. We integrated these approaches into prototype Neuropixels Opto probes, which combine electronic and photonic circuits. These devices pack 960 electrical recording sites and two sets of 14 light emitters onto a 70 m wide, 1 cm long shank, allowing spatially addressable optogenetic stimulation with blue and red light. In mouse cortex, Neuropixels Opto probes delivered high-quality recordings together with spatially addressable optogenetics, differentially activating or silencing neurons at distinct cortical depths. In mouse striatum and other deep structures, Neuropixels Opto probes delivered efficient optotagging, facilitating the identification of two cell types in parallel. Neuropixels Opto probes represent an unprecedented tool for recording, identifying, and manipulating neuronal populations.

neuroscience↗

bioRxiv2024

Adaptation shapes the representational geometry in mouse V1 to efficiently encode the environment

Sensory adaptation dynamically changes neural responses as a function of previous stimuli, profoundly impacting perception. The response changes induced by adaptation have been characterized in detail in individual neurons and at the population level after averaging across trials. However, it is not clear how adaptation modifies the aspects of the representations that relate more directly to the ability to perceive stimuli, such as their geometry and the noise structure in individual trials. To address this question, we recorded from a population of neurons in the mouse visual cortex and presented one stimulus (an oriented grating) more frequently than the others. We then analyzed these data in terms of representational geometry and studied the ability of a linear decoder to discriminate between similar visual stimuli based on the single-trial population responses. Surprisingly, the discriminability of stimuli near the adaptor increased, even though the responses of individual neurons to these stimuli decreased. This improvement arose mainly from how adaptation reshaped the geometry of the mean population responses, with changes in the noise structure playing a minor role. Locomotion modulated the population responses along a direction largely independent of adaptation, allowing the adaptation-induced increase in discriminability to persist during locomotion. Finally, artificial neural networks trained to reconstruct the visual stimulus under metabolic constraints reproduced the increase in discriminability and the decrease in responses. The paradoxical effects of adaptation are thus consistent with the efficient coding framework, allowing the brain to improve the representation of frequent stimuli while limiting the associated metabolic cost.

neuroscience↗

bioRxiv2024

Sensory choices as logistic classification

Logistic classification is a simple way to make choices based on a set of factors: give each factor a weight, sum the results, and use the sum to set the log odds of a random draw. This operation is known to describe human and animal choices based on value (economic decisions). There is increasing evidence that it also describes choices based on sensory inputs (perceptual decisions), presented across sensory modalities (multisensory integration) and combined with non-sensory factors such as prior probability, expected value, overall motivation, and recent actions. Logistic classification can also capture the effects of brain manipulations such as local inactivations. The brain may implement by thresholding stochastic inputs (as in signal detection theory) acquired over time (as in the drift diffusion model). It is the optimal strategy under certain conditions, and the brain appears to use it as a heuristic in a wider set of conditions.

neuroscience↗

bioRxiv2023

Tracking neurons across days with high-density probes

Neural activity spans multiple scales from milliseconds to months. Its evolution can be recorded with chronic electrodes, and especially with high-density arrays such as Neuropixels probes, which measure each spike at tens of sites and record hundreds of neurons. These arrays often record units with consistent spike waveforms over time, but produce vast amounts of data that require new approaches for tracking neurons across recordings. To meet this need, we developed UnitMatch, an open-access pipeline that operates after spike sorting, based only on each units average spike waveform. We tested UnitMatch in Neuropixels recordings from the mouse brain, where it tracked neurons across weeks. In visual cortex, a neurons selectivity for visual stimuli and correlation with other neurons remained stable over days. In striatum, neuronal responses changed across days during learning of a task. UnitMatch is thus a promising tool to reveal invariance or plasticity in neural activity across days.

neuroscience↗

bioRxiv2023

Reusable, flexible, and lightweight chronic implants for Neuropixels probes

Electrophysiology has proven invaluable to record neural activity, and the development of Neuropixels probes dramatically increased the number of recorded neurons. These probes are often implanted acutely, but acute recordings cannot be performed in freely moving animals and the recorded neurons cannot be tracked across days. To study key behaviors such as navigation, learning, and memory formation, the probes must be implanted chronically. An ideal chronic implant should (1) allow stable recordings of neurons for weeks; (2) allow reuse of the probes after explantation; (3) be light enough for use in mice. Here, we present the "Apollo Implant", an open-source and editable device that meets these criteria and accommodates up to two Neuropixels 1.0 or 2.0 probes. The implant comprises a "payload" module which is attached to the probe and is recoverable, and a "docking" module which is cemented to the skull. The design is adjustable, making it easy to change the distance between probes, the angle of insertion, and the depth of insertion. We tested the implant across eight labs in head-fixed mice, freely moving mice, and freely moving rats. The number of neurons recorded across days was stable, even after repeated implantations of the same probe. The Apollo implant provides an inexpensive, lightweight, and flexible solution for reusable chronic Neuropixels recordings.

neuroscience↗

bioRxiv2023

A Brain-Wide Map of Neural Activity during Complex Behaviour

A key challenge in neuroscience is understanding how neurons in hundreds of interconnected brain regions integrate sensory inputs with prior expectations to initiate movements and thereby make decisions. It is difficult to meet this challenge if different laboratories apply different analyses to different recordings in different regions during different behaviours. Here, we report a comprehensive set of recordings from 621733 neurons across 139 mice in 12 labs performing a decision-making task with sensory, motor, and cognitive components, obtained with 699 Neuropixels probe insertions covering 279 brain areas in the left forebrain and midbrain and the right hindbrain and cerebellum. We provide an initial appraisal of this brain-wide map, assessing how neural activity encoded key task variables. Representations of visual stimuli appeared transiently in classical visual areas after stimulus onset and then spread to ramp-like activity in a collection of mid- and hindbrain regions that also encoded choices. Neural responses correlated with motor action almost everywhere in the brain. Responses to reward delivery and consumption versus negative feedback were also widespread. This publicly available dataset represents an unprecedented resource for understanding how computations distributed across and within brain areas drive behaviour.

neuroscience↗