CoChat

Cortical reliability amid noise and chaos

Open access

Published version of a preprint

The DOI matches its Crossref record; 81 of the 81 references it cites were found in a registry or an index.

Abstract

Typical responses of cortical neurons to identical sensory stimuli appear highly variable. It has thus been proposed that the cortex primarily uses a rate code. However, other studies have argued for spike-time coding under certain conditions. The potential role of spike-time coding is directly limited by the internally generated variability of cortical circuits, which remains largely unexplored. Here, we quantify this internally generated variability using a biophysical model of rat neocortical microcircuitry with biologically realistic noise sources. We find that stochastic neurotransmitter release is a critical component of internally generated variability, causing rapidly diverging, chaotic recurrent network dynamics. Surprisingly, the same nonlinear recurrent network dynamics can transiently overcome the chaos in response to weak feed-forward thalamocortical inputs, and support reliable spike times with millisecond precision. Our model shows that the noisy and chaotic network dynamics of recurrent cortical microcircuitry are compatible with stimulus-evoked, millisecond spike-time reliability, resolving a long-standing debate.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/s41467-019-11633-8

Cites

81 of 81 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 Rapid divergence of spontaneous activity. a Morphologically detailed model of a neocortical microcircuit (NMC); depicted are 100 randomly selected neurons, out of 31,346 in total (~0.3%). Neurons are colored according to their layer. b Examples of simulated noise sources in the NMC-model: stochastic synaptic transmission, including (a) vesicle release failure and (b) spontaneous vesicle release (“miniature PSPs”) at all 36 million synapses; (c) probabilistic opening and closing of voltage-gated potassium channels in irregularly spiking inhibitory neurons (1137 out of 31,346 neurons); (d) a constant depolarizing current with a weak white noise component (σs²≪μs) injected into the somata of all neurons. c The membrane potential of four sample neurons (and population mean of all 31,346 neurons) during a network simulation of spontaneous activity. At t0, the state of the microcircuit is saved, and then resumed twice with identical initial conditions, but with different random seeds for all noise sources. d Root-mean square deviation (RMSDV(t)) and correlation (rV(t)) of the somatic membrane potentials between pairs of resumed simulations diverging from identical initial conditions (mean of all neurons and 40 saved base states ± 95% confidence interval). The dashed lines depict the steady-state RMSDV and rV between independent simulations (i.e., resumed from different base states). e The similarity of the system (sRMSD and sr) defined as the difference between the diverging and steady-state RMSDV and rV, normalized to lie between 1 (identical) and 0 (fully diverged) (mean ± 95% confidence interval). Exponential fit of sRMSD and sr for t − t0 < 40 ms (estimated time constant ± 68% confidence interval of fit). f1 Mean spike count and variance of spike count of 40 independent trials of 1000 ms duration for all neurons in the microcircuit, plotted separately for excitatory neurons (red) and inhibitory neurons (blue). The dashed lines indicate the expected values for a Poisson process. Black lines indicate minimum variance due to the fact that the spike count has to be an integer. f2 Distribution of Fano factors (variance divided by mean spike count) corresponding to f1
Fig. 1 Rapid divergence of spontaneous activity. a Morphologically detailed model of a neocortical microcircuit (NMC); depicted are 100 randomly selected neurons, out of 31,346 in total (~0.3%). Neurons are colored according to their layer. b Examples of simulated noise sources in the NMC-model: stochastic synaptic transmission, including (a) vesicle release failure and (b) spontaneous vesicle release (“miniature PSPs”) at all 36 million synapses; (c) probabilistic opening and closing of voltage-gated potassium channels in irregularly spiking inhibitory neurons (1137 out of 31,346 neurons); (d) a constant depolarizing current with a weak white noise component (σs²≪μs) injected into the somata of all neurons. c The membrane potential of four sample neurons (and population mean of all 31,346 neurons) during a network simulation of spontaneous activity. At t0, the state of the microcircuit is saved, and then resumed twice with identical initial conditions, but with different random seeds for all noise sources. d Root-mean square deviation (RMSDV(t)) and correlation (rV(t)) of the somatic membrane potentials between pairs of resumed simulations diverging from identical initial conditions (mean of all neurons and 40 saved base states ± 95% confidence interval). The dashed lines depict the steady-state RMSDV and rV between independent simulations (i.e., resumed from different base states). e The similarity of the system (sRMSD and sr) defined as the difference between the diverging and steady-state RMSDV and rV, normalized to lie between 1 (identical) and 0 (fully diverged) (mean ± 95% confidence interval). Exponential fit of sRMSD and sr for t − t0 < 40 ms (estimated time constant ± 68% confidence interval of fit). f1 Mean spike count and variance of spike count of 40 independent trials of 1000 ms duration for all neurons in the microcircuit, plotted separately for excitatory neurons (red) and inhibitory neurons (blue). The dashed lines indicate the expected values for a Poisson process. Black lines indicate minimum variance due to the fact that the spike count has to be an integer. f2 Distribution of Fano factors (variance divided by mean spike count) corresponding to f1CC BY · PubMed Central
Fig. 2 Robust divergence across dynamical states and circuit scale. a1 Population raster plot and population peristimulus time histogram (PSTH) for all 31,346 neurons in the microcircuit, during spontaneous activity. Neurons are ordered according to cortical depth, with deep layers at the bottom and upper layers at the top. Each row represents the spikes of one neuron. For visibility, raster lines extend over dozens of rows for each neuron. For t < t0, the top and bottom raster plots show the same simulation, whereas for t > t0, the raster plots depict two simulations resuming from identical initial conditions at t0, but using different random number seeds. a2 Same as a, but for supercritical activity. a3 rV across dynamical regimes (20 saved base states, mean ± 95% confidence interval; same as Fig. 1d for [Ca2+]o = 1.25 mM. b1 The microcircuit (center, blue), surrounded by six other microcircuits (gray), forming a continuous mesocircuit of ~220,000 neurons, with no boundary effects between the circuits. b2 rV for the center microcircuit when simulated without surrounding circuits (black), and of the center microcircuit when simulated as a mesocircuit (orange) (microcircuit: 40 saved base states; mesocircuit: 20 saved base states; mean ± 95% confidence interval). b3 Quantifying edge effects. Difference of rV between the same neurons in the microcircuit and the mesocircuit at 10–20 ms, plotted according to distance from horizontal center (mean ± 95% confidence interval). c Similarity sr at 10–20 ms for subsets of neurons grouped by in-degree (bin size: 50; mean ± 95% confidence interval)
Fig. 2 Robust divergence across dynamical states and circuit scale. a1 Population raster plot and population peristimulus time histogram (PSTH) for all 31,346 neurons in the microcircuit, during spontaneous activity. Neurons are ordered according to cortical depth, with deep layers at the bottom and upper layers at the top. Each row represents the spikes of one neuron. For visibility, raster lines extend over dozens of rows for each neuron. For t < t0, the top and bottom raster plots show the same simulation, whereas for t > t0, the raster plots depict two simulations resuming from identical initial conditions at t0, but using different random number seeds. a2 Same as a, but for supercritical activity. a3 rV across dynamical regimes (20 saved base states, mean ± 95% confidence interval; same as Fig. 1d for [Ca2+]o = 1.25 mM. b1 The microcircuit (center, blue), surrounded by six other microcircuits (gray), forming a continuous mesocircuit of ~220,000 neurons, with no boundary effects between the circuits. b2 rV for the center microcircuit when simulated without surrounding circuits (black), and of the center microcircuit when simulated as a mesocircuit (orange) (microcircuit: 40 saved base states; mesocircuit: 20 saved base states; mean ± 95% confidence interval). b3 Quantifying edge effects. Difference of rV between the same neurons in the microcircuit and the mesocircuit at 10–20 ms, plotted according to distance from horizontal center (mean ± 95% confidence interval). c Similarity sr at 10–20 ms for subsets of neurons grouped by in-degree (bin size: 50; mean ± 95% confidence interval)CC BY · PubMed Central
Fig. 3 Noise amplified by chaos determines divergence. a1 Time-course of correlation rV after resuming at t0 from identical conditions with different forms of perturbation. Full cellular noise as before, solid line (abcd); no cellular noise, but perturbing with a single extra spike in one neuron, dashed line (f); a miniscule step-pulse perturbation in all neurons, dotted line (e). (abcd: 40 saved base states; e, f: 20 saved base states; mean ± 95% confidence interval). a2 Steady-state root-mean square deviation RMSD∞ and correlation r∞ for stochastic (abcd) and deterministic simulations (e, f) as defined in a1 (mean ± 95% confidence interval in black; individual base states in purple dots). b1 As in a2, but for decoupled, replayed simulations. (20 saved base states). b2 Similarity sRMSD and sr at 10–20 ms with all noise sources enabled, for network and decoupled simulations (mean ± 95% confidence interval). c1 Decoupled replay paradigm. Presynaptic spike trains from a network simulation are saved and then replayed to the synapses of each neuron in a decoupled simulation, thereby removing variability due to feedback network dynamics. c2 Overview of sources of noise and perturbations. d Decoupled replay simulations (see c1) for a representative L4 PC neuron, with somatic membrane potential differences between the two trials only due to cellular noise sources (ab[c]d), a single extra presynaptic spike (f) or a miniscule step-pulse perturbation (e). [c] indicates that for some neuron types in the NMC-model, such as L4 PCs, no stochastic ion channels are present
Fig. 3 Noise amplified by chaos determines divergence. a1 Time-course of correlation rV after resuming at t0 from identical conditions with different forms of perturbation. Full cellular noise as before, solid line (abcd); no cellular noise, but perturbing with a single extra spike in one neuron, dashed line (f); a miniscule step-pulse perturbation in all neurons, dotted line (e). (abcd: 40 saved base states; e, f: 20 saved base states; mean ± 95% confidence interval). a2 Steady-state root-mean square deviation RMSD∞ and correlation r∞ for stochastic (abcd) and deterministic simulations (e, f) as defined in a1 (mean ± 95% confidence interval in black; individual base states in purple dots). b1 As in a2, but for decoupled, replayed simulations. (20 saved base states). b2 Similarity sRMSD and sr at 10–20 ms with all noise sources enabled, for network and decoupled simulations (mean ± 95% confidence interval). c1 Decoupled replay paradigm. Presynaptic spike trains from a network simulation are saved and then replayed to the synapses of each neuron in a decoupled simulation, thereby removing variability due to feedback network dynamics. c2 Overview of sources of noise and perturbations. d Decoupled replay simulations (see c1) for a representative L4 PC neuron, with somatic membrane potential differences between the two trials only due to cellular noise sources (ab[c]d), a single extra presynaptic spike (f) or a miniscule step-pulse perturbation (e). [c] indicates that for some neuron types in the NMC-model, such as L4 PCs, no stochastic ion channels are presentCC BY · PubMed Central
Fig. 4 Synaptic noise dominates divergence. a1 Time-course of correlation rV after resuming at t0 from identical conditions with different noise sources enabled (abcd: 40 bases states; a, ab, c, d: 20 base states; mean ± 95% confidence interval). a2 Steady-state root-mean square deviation RMSD∞ (cyan) and correlation r∞ (purple) with different noise sources enabled. Dots indicate individual trials; black error bars indicate 95% confidence interval. a3 Similarity sRMSD at 10–20 ms with different noise sources enabled, for all neurons (cyan) and irregular e-types (orange). b Steady-state root-mean square deviation for decoupled simulations, RMSD∞,dec, for all neurons (cyan) and irregular e-types (orange) (20 saved base states). Irregular e-types: 1137 out of 31,346 neurons. c Decoupled replay simulations for a representative L6 NBC neuron, with somatic membrane potential differences between the two trials only due to synaptic noise (ab), ion-channel noise (c), or a noisy current injection (d). d1 The effect of changing random seeds for the noisy depolarization only, for different noise strengths in a decoupled simulation. x: white noise variance as percentage of mean injected current. d2 The decoupled steady-state membrane potential fluctuations RMSD∞,dec^dx evoked by different magnitudes of white noise without network dynamics, vs. the similarity sRMSD at 10–20 ms during network simulations when either turning on only the white noise depolarization (d) or all noise sources (abcd). Similarly, in purple, RMSD∞,dec^ab for synaptic noise vs. the similarity at 10–20 ms when only turning on synaptic noise (ab). All error bars and shaded areas indicate 95% confidence intervals. Means for d2 are based on ten base states
Fig. 4 Synaptic noise dominates divergence. a1 Time-course of correlation rV after resuming at t0 from identical conditions with different noise sources enabled (abcd: 40 bases states; a, ab, c, d: 20 base states; mean ± 95% confidence interval). a2 Steady-state root-mean square deviation RMSD∞ (cyan) and correlation r∞ (purple) with different noise sources enabled. Dots indicate individual trials; black error bars indicate 95% confidence interval. a3 Similarity sRMSD at 10–20 ms with different noise sources enabled, for all neurons (cyan) and irregular e-types (orange). b Steady-state root-mean square deviation for decoupled simulations, RMSD∞,dec, for all neurons (cyan) and irregular e-types (orange) (20 saved base states). Irregular e-types: 1137 out of 31,346 neurons. c Decoupled replay simulations for a representative L6 NBC neuron, with somatic membrane potential differences between the two trials only due to synaptic noise (ab), ion-channel noise (c), or a noisy current injection (d). d1 The effect of changing random seeds for the noisy depolarization only, for different noise strengths in a decoupled simulation. x: white noise variance as percentage of mean injected current. d2 The decoupled steady-state membrane potential fluctuations RMSD∞,dec^dx evoked by different magnitudes of white noise without network dynamics, vs. the similarity sRMSD at 10–20 ms during network simulations when either turning on only the white noise depolarization (d) or all noise sources (abcd). Similarly, in purple, RMSD∞,dec^ab for synaptic noise vs. the similarity at 10–20 ms when only turning on synaptic noise (ab). All error bars and shaded areas indicate 95% confidence intervals. Means for d2 are based on ten base statesCC BY · PubMed Central
Fig. 5 Rapid divergence of evoked, reliable activity. a1 Population raster plot and population peristimulus time histogram (PSTH) for all 31,346 neurons in the microcircuit, during evoked activity with a thalamic (VPM) stimulus. Neurons are ordered according to cortical depth, with deep layers at the bottom and upper layers at the top, and each row representing the spikes of one neuron. For visibility, raster lines extend over dozens of rows for each neuron. a2 Mean somatic membrane potential correlation rV between independent simulations of the same VPM stimulus (mean ± 95% confidence interval). a3 Schematic of the VPM stimulus. Top: Raster plot spike times for the first 250 ms of the thalamic stimulus. Bottom: 310 VPM fiber centers are assigned 30 colors, and those with identical colors are provided with duplicate spike trains. The synapse density profile across layers for each fiber is shown to the right. b For t < 100, the top and bottom raster plots show the same simulation, whereas for t > 100, the raster plots depict two resumed simulations starting from the same saved state at t0 = 100, using different random number seeds. c1 Resuming from identical initial conditions at different times: during (top), at onset (middle), or before the stimulus (bottom). Mean rV between independent simulations (blue, as in a2), and mean rV between simulations starting from the same base state (red; mean ± 95% confidence interval). c2 The similarity, sr, defined as the difference between the rV of diverging and independent trials, normalized to lie between 1 (identical) and 0 (fully diverged) (mean ± 95% confidence interval). Means are based on 20 base states, no stimulus (spontaneous activity) on 40 as before
Fig. 5 Rapid divergence of evoked, reliable activity. a1 Population raster plot and population peristimulus time histogram (PSTH) for all 31,346 neurons in the microcircuit, during evoked activity with a thalamic (VPM) stimulus. Neurons are ordered according to cortical depth, with deep layers at the bottom and upper layers at the top, and each row representing the spikes of one neuron. For visibility, raster lines extend over dozens of rows for each neuron. a2 Mean somatic membrane potential correlation rV between independent simulations of the same VPM stimulus (mean ± 95% confidence interval). a3 Schematic of the VPM stimulus. Top: Raster plot spike times for the first 250 ms of the thalamic stimulus. Bottom: 310 VPM fiber centers are assigned 30 colors, and those with identical colors are provided with duplicate spike trains. The synapse density profile across layers for each fiber is shown to the right. b For t < 100, the top and bottom raster plots show the same simulation, whereas for t > 100, the raster plots depict two resumed simulations starting from the same saved state at t0 = 100, using different random number seeds. c1 Resuming from identical initial conditions at different times: during (top), at onset (middle), or before the stimulus (bottom). Mean rV between independent simulations (blue, as in a2), and mean rV between simulations starting from the same base state (red; mean ± 95% confidence interval). c2 The similarity, sr, defined as the difference between the rV of diverging and independent trials, normalized to lie between 1 (identical) and 0 (fully diverged) (mean ± 95% confidence interval). Means are based on 20 base states, no stimulus (spontaneous activity) on 40 as beforeCC BY · PubMed Central
Fig. 6 Spike-time reliability amid noise and chaos. a1 Mean somatic membrane potential correlation, rV, between independent simulations, and between decoupled replays of those simulations (network simulation identical to Fig. 5a2). a2 Difference in rV for decoupled and network simulations. a3 Schematic of network and decoupled replay simulation paradigms, including thalamic input. b Somatic membrane potentials (Vm) of three representative neurons for the time interval highlighted by the red box in a. Top: during six independent trials. Bottom: five decoupled replay trials (green) with the same presynaptic input as during the original network simulation trial (red), but with different random seeds. c Network and decoupled rV as in a, but only for the three sample neurons in b. d Top: Raster plot of spike times of the same three example neurons as in b, during 30 independent trials of evoked activity. Bottom: Decoupled replay trials (green) of the same input received during 5 of the 30 original trials (dark red). e1 Mean spike-time reliability rspike of 2024 pyramidal neurons from layers 4, 5, and 6 between independent network simulations, and between decoupled replay simulations with identical presynaptic inputs. e2 Difference between rspike of decoupled and replayed simulations. e3 Difference between rspike of decoupled and replayed simulations vs. position of somata across layers 4, 5, and 6 of microcircuit (1675 neurons)
Fig. 6 Spike-time reliability amid noise and chaos. a1 Mean somatic membrane potential correlation, rV, between independent simulations, and between decoupled replays of those simulations (network simulation identical to Fig. 5a2). a2 Difference in rV for decoupled and network simulations. a3 Schematic of network and decoupled replay simulation paradigms, including thalamic input. b Somatic membrane potentials (Vm) of three representative neurons for the time interval highlighted by the red box in a. Top: during six independent trials. Bottom: five decoupled replay trials (green) with the same presynaptic input as during the original network simulation trial (red), but with different random seeds. c Network and decoupled rV as in a, but only for the three sample neurons in b. d Top: Raster plot of spike times of the same three example neurons as in b, during 30 independent trials of evoked activity. Bottom: Decoupled replay trials (green) of the same input received during 5 of the 30 original trials (dark red). e1 Mean spike-time reliability rspike of 2024 pyramidal neurons from layers 4, 5, and 6 between independent network simulations, and between decoupled replay simulations with identical presynaptic inputs. e2 Difference between rspike of decoupled and replayed simulations. e3 Difference between rspike of decoupled and replayed simulations vs. position of somata across layers 4, 5, and 6 of microcircuit (1675 neurons)CC BY · PubMed Central
Fig. 7 High reliability requires recurrent cortical connectivity. a1 Overview of three simulation paradigms: spontaneous activity, network evoked activity (with network propagation intact and VPM input), and mixed replay (with network propagation replaced by replays of spontaneous activity spike trains, and VPM input). a2 Examples of population spiking activity during the three simulation paradigms. b1 Spike-time reliability, rspike, during spontaneous (blue) and evoked (purple) activity for 1675 excitatory neurons in the center of layers 4, 5, and 6. b2 Spike-time reliability, rspike, during a mixed replay with VPM input but with network propagation disabled for the same neurons as in b1. c Difference in rspike between evoked activity with and without network propagation for 1892 excitatory neurons in the center of layers 4, 5, and 6 (same for d1–3). d1 The number of presynaptic VPM fibers from which each neuron receives input vs. rspike in evoked simulations with (network) and without (mixed replay) network propagation. All reliabilities per neuron (points) are based on 30 trials. Mean of neurons per bin indicated by line; shaded area depicts standard error of mean of neurons in bin. Bins: 0–4, 5–9, 10–14, 15–19, 20–24, 25–29, 30+ VPM fibers
Fig. 7 High reliability requires recurrent cortical connectivity. a1 Overview of three simulation paradigms: spontaneous activity, network evoked activity (with network propagation intact and VPM input), and mixed replay (with network propagation replaced by replays of spontaneous activity spike trains, and VPM input). a2 Examples of population spiking activity during the three simulation paradigms. b1 Spike-time reliability, rspike, during spontaneous (blue) and evoked (purple) activity for 1675 excitatory neurons in the center of layers 4, 5, and 6. b2 Spike-time reliability, rspike, during a mixed replay with VPM input but with network propagation disabled for the same neurons as in b1. c Difference in rspike between evoked activity with and without network propagation for 1892 excitatory neurons in the center of layers 4, 5, and 6 (same for d1–3). d1 The number of presynaptic VPM fibers from which each neuron receives input vs. rspike in evoked simulations with (network) and without (mixed replay) network propagation. All reliabilities per neuron (points) are based on 30 trials. Mean of neurons per bin indicated by line; shaded area depicts standard error of mean of neurons in bin. Bins: 0–4, 5–9, 10–14, 15–19, 20–24, 25–29, 30+ VPM fibersCC BY · PubMed Central
Fig. 8 High reliability emerges near a critical EI-balance. a1–3 Raster plot of spike times of the same three example neurons as in Fig. 6d, during 30 independent trials of evoked activity, for three different extracellular calcium concentrations ([Ca2+]o = 1.15, 1.25, 1.35 mM). b Population firing rates for all 30 trials for eight different extracellular calcium concentrations. Loss of stimulus discriminatory power away from [Ca2+]o = 1.25 mM as observed by Markram et al.17. c Mean reliability of 2024 pyramidal neurons from layers 4, 5, and 6 during the evoked response vs. extracellular calcium concentration (mean of 30 trials ± 95% confidence interval). d Total population firing rates for excitatory and inhibitory subpopulations in the whole microcircuit during the evoked response
Fig. 8 High reliability emerges near a critical EI-balance. a1–3 Raster plot of spike times of the same three example neurons as in Fig. 6d, during 30 independent trials of evoked activity, for three different extracellular calcium concentrations ([Ca2+]o = 1.15, 1.25, 1.35 mM). b Population firing rates for all 30 trials for eight different extracellular calcium concentrations. Loss of stimulus discriminatory power away from [Ca2+]o = 1.25 mM as observed by Markram et al.17. c Mean reliability of 2024 pyramidal neurons from layers 4, 5, and 6 during the evoked response vs. extracellular calcium concentration (mean of 30 trials ± 95% confidence interval). d Total population firing rates for excitatory and inhibitory subpopulations in the whole microcircuit during the evoked responseCC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 81 references

Cited by

See all 56

Metadata

Authors
Nolte, M., Reimann, M. W., King, J. G., Markram, H., & Muller, E. B.
Affiliations
École Polytechnique Fédérale de Lausanne (CH)
Corresponding author
  • Nolte, Max
  • Muller, Eilif B.
Venue
Nature Communications
Published
2019-08-22
Volume
10
Issue
1
Pages
3792-3792
ISSN
2041-1723 (ISSN-L)
OpenAlex
W2969778699
Cited by
56
References
80
Citations by year
2026: 4 · 2025: 9 · 2024: 11 · 2023: 8 · 2022: 8 · 2021: 8 · 2020: 6 · 2019: 1
FWCI
3.34
Type
journal article
Language
en
Open access
Licence
Locations
  • Nature Communications (publishedVersion, cc-by)
  • bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • RePEc: Research Papers in Economics (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • Infoscience (Ecole Polytechnique Fédérale de Lausanne) (submittedVersion, cc-by-nc-nd)
Funders
  • École Polytechnique Fédérale de Lausanne
  • Board of the Swiss Federal Institutes of Technology
  • National Supercomputing Center, Korea Institute of Science and Technology Information
Topics
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Neural dynamics and brain function
  • Physical Sciences › Physics and Astronomy › Statistical and Nonlinear Physics › stochastic dynamics and bifurcation
  • Life Sciences › Neuroscience › Cellular and Molecular Neuroscience › Neuroscience and Neuropharmacology Research
Keywords
spike time coding, rate coding, cortical microcircuitry, chaos, thalamocortical input, biophysical neuron models
MeSH terms
Action Potentials, Animals, Cerebral Cortex, Models, Neurological, Nerve Net, Neurons, Thalamus, Time Factors, Reproducibility of Results, Nonlinear Dynamics, Neurotransmitter Agents, Rats, Synaptic Potentials

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Cortical reliability amid noise and chaos. https://app.cochat.ai/papers/10.1038/s41467-019-11633-8
The paper
Nolte, M., Reimann, M. W., King, J. G., Markram, H., & Muller, E. B. (2019). Cortical reliability amid noise and chaos. Nature Communications. https://doi.org/10.1038/s41467-019-11633-8

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.