The DOI matches its Crossref record; 81 of the 81 references it cites were found in a registry or an index; 18 of the 42 sources its claims cite were checked for support.
A virtual rodent predicts the structure of neural activity across behaviours
Open accessAbstract
Animals have exquisite control of their bodies, allowing them to perform a diverse range of behaviours. How such control is implemented by the brain, however, remains unclear. Advancing our understanding requires models that can relate principles of control to the structure of neural activity in behaving animals. Here, to facilitate this, we built a ‘virtual rodent’, in which an artificial neural network actuates a biomechanically realistic model of the rat1 in a physics simulator2. We used deep reinforcement learning3, 4–5 to train the virtual agent to imitate the behaviour of freely moving rats, thus allowing us to compare neural activity recorded in real rats to the network activity of a virtual rodent mimicking their behaviour. We found that neural activity in the sensorimotor striatum and motor cortex was better predicted by the virtual rodent’s network activity than by any features of the real rat’s movements, consistent with both regions implementing inverse dynamics6. Furthermore, the network’s latent variability predicted the structure of neural variability across behaviours and afforded robustness in a way consistent with the minimal intervention principle of optimal feedback control7. These results demonstrate how physical simulation of biomechanically realistic virtual animals can help interpret the structure of neural activity across behaviour and relate it to theoretical principles of motor control. We built an artificial neural network to control a biomechanically realistic virtual rodent, which, when trained to imitate real rats, predicts neural activity and variability across natural behaviours.
Overview
CoChat Summary
TL;DR
The authors built a “virtual rodent”: an artificial neural network implementing inverse dynamics that actuates a biomechanically realistic rat model in MuJoCo, trained with deep reinforcement learning to imitate freely moving rats. Its network activity predicted sensorimotor striatum (DLS) and motor cortex (MC) neural activity better than movement features, and its latent variability structured action variability in a way consistent with the minimal intervention principle.
Key findings
- The virtual rodent imitated held-out rat behavior across a 607-hour dataset; recurrent decoders outperformed other architectures, and most deviations from real kinematics were attributable to accumulated center-of-mass error.
- In DLS (3 animals, 353.5 hours, 1249 neurons) and MC (3 animals, 253.5 hours, 843 neurons), inverse dynamics model activations predicted single-unit activity better than kinematic or dynamic representational features, with best results from the first decoder layer (p < .001, one-sided permutation test; N = 732 DLS and 769 MC significantly predicted neurons).
- Representational similarity analysis showed that DLS and MC population activity was more similar to inverse dynamics model representations than to representational models, and this similarity correlated with imitation performance and robustness.
- Latent variability structured action variability across behaviors; latent variability alone discriminated behavior (p < .001), and uniform or inverted variability perturbations reduced robustness and increased termination (p < .001), consistent with the minimal intervention principle; latent variability also predicted the structure of neural variability across behaviors.
Methods
- Freely moving female Long Evans rats were recorded in a circular arena with six cameras and custom 128-channel tetrode drives targeting DLS or MC; the dataset spans 607 hours of simultaneous electrophysiology and videography.
- The MIMIC pipeline used DANNCE for 3D pose estimation of 23 keypoints, STAC to register a 74-degree-of-freedom skeletal model (38 controllable actuators) to those keypoints, and deep reinforcement learning (MO-VMPO) to train ANNs implementing inverse dynamics to control the model in MuJoCo at 50 Hz, using 842 five-second behavioral clips.
- Analyses included Poisson generalized linear models with cross-validation to predict spike counts from representational features versus ANN layers, representational similarity analysis using cross-validated Mahalanobis distance and whitened unbiased cosine similarity, and variability analyses that resampled the latent space 50 times per timestep and perturbed latent variability structure.
Limitations
- The paper states that the model uses point-torque actuation and omits muscular actuation, opting for the simplest abstraction sufficient for its objectives; it also notes that the models are most appropriate for short-timescale motor dynamics rather than long-term behavioral organization.
- The paper states that directly measuring instantaneous neural variability is not possible, so neural variability was approximated with a 1-second sliding-window variance; it also notes that alternative control models trained via supervised learning may differ from reinforcement-trained composite controllers.
- Evident from the text: neural comparisons are correlational and cover only DLS and MC with 3 animals each, and the virtual rodent’s behavior deviated slightly from real rats (e.g., center-of-mass drift), though the inverse dynamics model still outperformed representational models.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
9 Valid · 7 Not valid · 6 Not checked · 95 Own results and interpretations
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“This suggests that more performant models exhibit representations more similar to those of both DLS and MC, consistent with previous reports comparing neural activity with task-optimized neural networks 17,20.”
What the sources say
- Not checked — no full text of the sourcecites [17] Yamins, D. L. K. et al. Performance-optimized hierarchical models predict neural responses in higher visual cortex. Proc. Natl Acad. Sci. USA 111, 8619–8624 (2014). source
- Partly supportedcites [20] Kell, A. J. E., Yamins, D. L. K., Shook, E. N., Norman-Haignere, S. V. & McDermott, J. H. A task-optimized neural network replicates human auditory behavior, predicts brain responses, and reveals a cortical processing hierarchy. Neuron 98, 630–644 (2018). sourceExperimental — accuracy not yet measured
The ‘Relationship between cortical predictions and task performance’ section (p. 11) reports that better task performance strongly correlated with better prediction of auditory-cortex responses. This supports the general pattern invoked by the citation, but the paper does not establish that more performant models have representations more similar to those of DLS or MC specifically. (LLM, conf=0.88)
hedgedstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“suggesting a link between mechanisms for robustness and generalizability (see also 25,56).”
- Also stated in: § Abstract · plain
“Together, these results reveal a link between a computational mechanism for generalization (latent variable compression) and one for robustness in control (structured variability).”
What the sources say
- Partly supportedcites [25] Higgins, I. et al. Unsupervised deep learning identifies semantic disentanglement in single inferotemporal face patch neurons. Nat. Commun. 12, 6456 (2021). sourceExperimental — accuracy not yet measured
The Introduction (page not provided) describes a low-dimensional latent representation and says disentangled models allow strong generalisation. The Results section reports heterogeneous latent-unit responses, including for novel face identities. But the paper does not connect those findings to robustness in control or identify structured variability as a control mechanism, so it supports only the generalisation side of the proposed link. (LLM, conf=0.82)
- Not matchedcites [56] Lai, L. & Gershman, S. J. in Psychology of Learning and Motivation, Vol. 74 (ed. Federmeier, K. D.) Ch. 5 (Academic Press, 2021).
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Also stated in: § Abstract · plain
The sources with a verdict support this claim; the rest weren’t checked yet, have no full text we could read, or their verdicts aren’t rated yet.
In the paper
“However, with the notable exception of signal-dependent noise 54, how neural activity in biological control networks shapes variability in motor output remains largely unexplored (though see 55).”
What the sources say
- Not checked — no full text of the sourcecites [54] Harris, C. M. & Wolpert, D. M. Signal-dependent noise determines motor planning. Nature 394, 780–784 (1998). source
- Supportedcites [55] Wolpert, D. M. Probabilistic models in human sensorimotor control. Hum. Mov. Sci. 26, 511–524 (2007). source
Citation 54 is attached to the signal-dependent-noise exception, not the broader claim that the topic remains largely unexplored. Section 3.2 explains that motor-command noise increases with command size, producing greater motor-output variability at larger forces. (LLM, conf=0.92)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Thus, managing the trade-off between latent variable compression and motor performance 25,56 may structure neural variability in ways that are distinct from previously hypothesized mechanisms like signal-dependent noise 54 or energy constraints 7.”
What the sources say
- Partly supportedcites [25] Higgins, I. et al. Unsupervised deep learning identifies semantic disentanglement in single inferotemporal face patch neurons. Nat. Commun. 12, 6456 (2021). sourceExperimental — accuracy not yet measured
The Results section (‘Disentangled units are sufficient to decode novel faces’; page unavailable) describes a trade-off between the β-VAE’s latent representation and face-image reconstruction quality, and relates representational form to neural decoding performance. It does not study motor performance or show that this trade-off structures motor-related neural variability. The citations following signal-dependent noise and energy constraints are separate, so those comparisons are not assessed against this source. (LLM, conf=0.83)
- Not matchedcites [56] Lai, L. & Gershman, S. J. in Psychology of Learning and Motivation, Vol. 74 (ed. Federmeier, K. D.) Ch. 5 (Academic Press, 2021).
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not checked — no full text of the sourcecites [54] Harris, C. M. & Wolpert, D. M. Signal-dependent noise determines motor planning. Nature 394, 780–784 (1998). source
- Not checked — no full text of the sourcecites [7] Todorov, E. & Jordan, M. I. Optimal feedback control as a theory of motor coordination. Nat. Neurosci. 5, 1226–1235 (2002). source
hedgedstates a possible cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
The cited sources that could be checked couldn’t confirm or dispute this claim.
In the paper
“The distributions resemble previous reports in rats during spontaneous behavior 42.”
What the sources say
- Could not verifycites [42] Klibaite, U. et al. Deep phenotyping reveals movement phenotypes in mouse neurodevelopmental models. Mol. Autism 13, 12 (2022). sourceExperimental — accuracy not yet measured
The Results section describes behavioral classification of mice in an open field, while Methods > Statistics describes comparisons of behavioral classes. Neither those passages nor searches for GLM predictivity, pseudo r-squared, DLS, MC, and rat spontaneous behavior yielded a passage about neuronal predictivity distributions or their resemblance to reports in rats. No page numbers are provided in our copy. (LLM, conf=0.76)
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Could not verify
45 more claims
Create a free account to see all 117 claims and every source verdict.
- Probably not a substantial claim. 38
- It may state two separate claims as one. 35
- The quote only partly states this claim. 3
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated major: it would touch a result — a value a result rests on, reported and recomputed p-values on opposite sides of the significance level, two passages that conflict with no sign which is right, or a cited work retracted before this paper was published.
In the paper
§ Abstract
“Data includes neurons significantly predicted by each GLM (Benjamini-Hochberg corrected Wilcoxon signed-rank test, α = .05)”
§ Abstract
“Units with a confidence score greater than .05 with Bonferroni correction for multiple comparisons were labeled as unclassified.”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Abstract
“We next compiled a diverse catalog of behavioral motifs (847 5-second snippets) spanning the behavioral repertoire of the rat to provide training data for our ANN controllers.”
§ Abstract
“In the end we used a dataset of 842 clips.”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Abstract
“We trained GLMs using a five-fold cross validation scheme.”
§ Abstract
“We first divided the spiking, movement, and ANN data spanning the duration of a unit’s recording into 4-second chunks which were randomly distributed into ten folds. We trained GLMs using training data from nine of the folds and evaluated their performance on testing data from the remaining fold”
Citations
Cites
Statsmodels: Econometric and Statistical Modeling with Python
Seabold, S., & Perktold, J. (2010) · Proceedings of the Python in Science Conference · doi:10.25080/majora-92bf1922-011 · Cited by 6,465
MuJoCo: A physics engine for model-based control
Todorov, E., Erez, T., & Tassa, Y. (2012) · 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems · doi:10.1109/iros.2012.6386109 · Cited by 4,767
Self-citationThe coordination of arm movements: an experimentally confirmed mathematical model
Flash, T., & Hogan, N. (1985) · The Journal of Neuroscience · doi:10.1523/jneurosci.05-07-01688.1985 · Cited by 4,395
Representational similarity analysis-connecting the branches of systems neuroscience
Kriegeskorte, N. (2008) · Frontiers in Systems Neuroscience · doi:10.3389/neuro.06.004.2008 · Cited by 3,355
Optimal feedback control as a theory of motor coordination
Todorov, E., & Jordan, M. I. (2002) · Nature Neuroscience · doi:10.1038/nn963 · Cited by 3,081
Faisal, A. A., Selen, L. P. J., & Wolpert, D. M. (2008) · Nature Reviews Neuroscience · doi:10.1038/nrn2258 · Cited by 2,883
Signal-dependent noise determines motor planning
Harris, C. M., & Wolpert, D. M. (1998) · Nature · doi:10.1038/29528 · Cited by 2,498
Georgopoulos, A., Kalaska, J., Caminiti, R., & Massey, J. (1982) · The Journal of Neuroscience · doi:10.1523/jneurosci.02-11-01527.1982 · Cited by 2,130
Performance-optimized hierarchical models predict neural responses in higher visual cortex
Yamins, D. L. K., Hong, H., Cadieu, C. F., Solomon, E. A., Seibert, D., & DiCarlo, J. J. (2014) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1403112111 · Cited by 1,924
Context-dependent computation by recurrent dynamics in prefrontal cortex
Mante, V., Sussillo, D., Shenoy, K. V., & Newsome, W. T. (2013) · Nature · doi:10.1038/nature12742 · Cited by 1,847
Research works printed but not linked: 5
- Hasenclever, L., Pardo, F., Hadsell, R., Heess, N. & Merel, J. CoMic: complementary task learning & mimicry for reusable skills. In Proc. 37th International Conference on Machine Learning (eds Daumé, H. & Singh, A.) 4105–4115 (PMLR, 2020).
- Jordan, M. I. in Handbook of Perception and Action, Vol. 2 (ed. Heuer, H.) Ch. 2 (Academic Press, 1996).
- Lai, L. & Gershman, S. J. in Psychology of Learning and Motivation, Vol. 74 (ed. Federmeier, K. D.) Ch. 5 (Academic Press, 2021).
- Kingma, D. P. & Ba, J. Adam: a method for stochastic optimization. In Proc. 3rd International Conference on Learning Representations (eds Bengio, Y. & LeCun, Y.) (2015).
- Maas, A. L., Hannun, A. Y. & Ng, A. Y. Rectifier nonlinearities improve neural network acoustic models. In Proc. 30th International Conference on Machine Learning (ICML) (2013).
Cited by
The spatial and temporal structure of neural activity across the fly brain
Schaffer, E. S., Mishra, N., Whiteway, M. R., Li, W., Vancura, M. B., Freedman, J., Patel, K. B., Voleti, V., Paninski, L., Hillman, E. M. C., et al. (2023) · Nature Communications · doi:10.1038/s41467-023-41261-2 · Cited by 53
Artificial intelligence for life sciences: A comprehensive guide and future trends
Luo, M., Yang, W., Bai, L., Zhang, L., Huang, J., Cao, Y.-H., Xie, Y., Tong, L., Zhang, H., Lei, Y., et al. (2024) · The Innovation Life · doi:10.59717/j.xinn-life.2024.100105 · Cited by 61
Bikkasani, D. C. (2024) · AI and Ethics · doi:10.1007/s43681-024-00642-z · Cited by 65
NeuroMechFly v2: simulating embodied sensorimotor control in adult Drosophila
Wang-Chen, S., Stimpfling, V. A., Lam, T. K. C., Özdil, P. G., Genoud, L., Hurtak, F., & Ramdya, P. (2024) · Nature Methods · doi:10.1038/s41592-024-02497-y · Cited by 43
Flexible multichannel muscle impedance sensors for collaborative human-machine interfaces
Li, J., Wu, K., Xiao, J., Chen, T., Yang, X., Pan, J., Chen, Y., & Wang, Y. (2025) · Science Advances · doi:10.1126/sciadv.adv3359 · Cited by 42
“As human-machine interfaces (HMIs) become increasingly prevalent in fields such as domestic robots ( 1 , 2 ), assistive exoskeletons ( 3 , 4 ), and virtual reality (VR) ( 5 , 6 ), the need for sensing interfaces that seamlessly integrate with human physiology has become critical.”
Metadata
- Authors
- Aldarondo, D., Merel, J., Marshall, J. D., Hasenclever, L., Klibaite, U., Gellis, A., Tassa, Y., Wayne, G., Botvinick, M., & Ölveczky, B. P.
- Affiliations
- Harvard University (US)
- Google DeepMind (United Kingdom) (GB)
- Corresponding author
- Aldarondo, Diego
- ORCID
- Venue
- Nature
- Published
- 2024-06-11
- Volume
- 632
- Issue
- 8025
- Pages
- 594-602
- ISSN
- 0028-0836 (ISSN-L)
- 1476-4687
- PMC
- PMC12080270
- OpenAlex
- W4399534689
- Semantic Scholar
- b5ee2c42a8d6b6846b894d5ac04301f4c8a07ab6
- Cited by
- 70
- References
- 76
- Citations by year
- 2026: 28 · 2025: 29 · 2024: 12
- FWCI
- 16.71
- Type
- journal article
- Language
- en
- Update notices
- correction (2025-08-11)
- Open access
- green
- www.ncbi.nlm.nih.gov
- Licence
- www.springernature.com/gp/researchers/text-and-data-mining (tdm)
- www.springernature.com/gp/researchers/text-and-data-mining (vor)
- Locations
- Nature (publishedVersion)
- PubMed (publishedVersion)
- PubMed Central (acceptedVersion)
- Funders
- National Institutes of Health (5r01ns099323-08)
- National Institute of General Medical Sciences (R01GM136972)
- National Institute of Neurological Disorders and Stroke (R01-NS099323, F99NS125834)
- Topics
- Life Sciences › Neuroscience › Cognitive Neuroscience › Motor Control and Adaptation
- Life Sciences › Neuroscience › Cognitive Neuroscience › Neural dynamics and brain function
- Life Sciences › Neuroscience › Cognitive Neuroscience › Memory and Neural Mechanisms
- Keywords
- motor control, physical simulation, neural decoding
- MeSH terms
- Virtual Reality, Deep Learning, Animals, Behavior, Animal, Female, Models, Neurological, Motor Cortex, Movement, Neural Networks, Computer, Rats, Long-Evans, Rats, Sensorimotor Cortex
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.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-01, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-01 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
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: A virtual rodent predicts the structure of neural activity across behaviours. https://app.cochat.ai/papers/10.1038/s41586-024-07633-4
- The paper
- Aldarondo, D., Merel, J., Marshall, J. D., Hasenclever, L., Klibaite, U., Gellis, A., Tassa, Y., Wayne, G., Botvinick, M., & Ölveczky, B. P. (2024). A virtual rodent predicts the structure of neural activity across behaviours. Nature. https://doi.org/10.1038/s41586-024-07633-4
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.