The DOI matches its Crossref record; 157 of the 161 references it cites were found in a registry or an index, and 4 are not checked yet.
If deep learning is the answer, what is the question?
Abstract
Neuroscience research is undergoing a minor revolution. Recent advances in machine learning and artificial intelligence research have opened up new ways of thinking about neural computation. Many researchers are excited by the possibility that deep neural networks may offer theories of perception, cognition and action for biological brains. This approach has the potential to radically reshape our approach to understanding neural systems, because the computations performed by deep networks are learned from experience, and not endowed by the researcher. If so, how can neuroscientists use deep networks to model and understand biological brains? What is the outlook for neuroscientists who seek to characterize computations or neural codes, or who wish to understand perception, attention, memory and executive functions? In this Perspective, our goal is to offer a road map for systems neuroscience research in the age of deep learning. We discuss the conceptual and methodological challenges of comparing behaviour, learning dynamics and neural representations in artificial and biological systems, and we highlight new research questions that have emerged for neuroscience as a direct consequence of recent advances in machine learning.
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.
Integrity signals
Registry
Cites
Statistics
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.
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
ImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155 · cited 2×
“…we highlight new research questions that have emerged for neuroscience as a direct consequence of recent advances in machine learning. pooling and normalization, the early layers acquire simple filters for orientation and spatial frequency 2 , just like neurons in the primary visual cortex (Fig.”
BackgroundLeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
Human-level control through deep reinforcement learning
Mnih, V., Kavukcuoglu, K., Silver, D., Rusu, A. A., Veness, J., Bellemare, M. G., Graves, A., Riedmiller, M., Fidjeland, A. K., Ostrovski, G., et al. (2015) · Nature · doi:10.1038/nature14236 · Cited by 30,357 · cited 4×
“By contrast, a successful AI model that has yet to impact neuroscience proposes instead that representation formation is driven by the need to accurately predict the motivational value of experience 5 .”
BackgroundInfluentialAuto-Encoding Variational Bayes
Kingma, D. P., & Welling, M. (2013) · arXiv · doi:10.48550/arxiv.1312.6114 · Cited by 20,067
Mastering the game of Go with deep neural networks and tree search
Silver, D., Huang, A., Maddison, C. J., Guez, A., Sifre, L., van den Driessche, G., Schrittwieser, J., Antonoglou, I., Panneershelvam, V., Lanctot, M., et al. (2016) · Nature · doi:10.1038/nature16961 · Cited by 15,612
Mastering the game of Go without human knowledge
Silver, D., Schrittwieser, J., Simonyan, K., Antonoglou, I., Huang, A., Guez, A., Hubert, T., Baker, L., Lai, M., Bolton, A., et al. (2017) · Nature · doi:10.1038/nature24270 · Cited by 9,084
The free-energy principle: a unified brain theory?
Friston, K. (2010) · Nature Reviews Neuroscience · doi:10.1038/nrn2787 · Cited by 8,120 · cited 1×
“However, a long tradition in neuroscience emphasizes unsupervised principles such as Hebbian learning, or relatedly that representations are formed by a pressure to accurately predict the spatially or temporally local environment under an efficiency constraint 66–68 .”
BackgroundOvercoming catastrophic forgetting in neural networks
Kirkpatrick, J., Pascanu, R., Rabinowitz, N., Veness, J., Desjardins, G., Rusu, A. A., Milan, K., Quan, J., Ramalho, T., Grabska-Barwinska, A., et al. (2017) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1611835114 · Cited by 7,899 · cited 1×
“…promising solutions to continual learning in the machine learning literature rely on the identification of weight subspaces in which new learning is least likely to cause retrospective interference, for example by ‘freezing’ synapses that are more likely to participate in extant tasks 129,130 .”
BackgroundThe hippocampus as a spatial map. Preliminary evidence from unit activity in the freely-moving rat
O'Keefe, J., & Dostrovsky, J. (1971) · Brain Research · doi:10.1016/0006-8993(71)90358-1 · Cited by 6,151 · cited 1×
“For example, in rodents, hippocampal place cells encode locations in a way that is invariant to viewpoint and heading direction 95 , and in primates, ‘schema’ cells activate across environments in a way that allows for generalization over common spatial configurations 96 .”
BackgroundEmergence of simple-cell receptive field properties by learning a sparse code for natural images
Olshausen, B. A., & Field, D. J. (1996) · Nature · doi:10.1038/381607a0 · Cited by 5,694 · cited 1×
“However, a long tradition in neuroscience emphasizes unsupervised principles such as Hebbian learning, or relatedly that representations are formed by a pressure to accurately predict the spatially or temporally local environment under an efficiency constraint 66–68 .”
Background
Research works printed but not linked: 6
- Golan, T., Raju, P. C. & Kriegeskorte, N. Controversial stimuli: pitting neural networks against each other as models of human recognition. Preprint at arXiv https://arxiv.org/abs/1911.09288 (2020). · arXiv
- Lillicrap, T. P. & Kording, K. P. What does it mean to understand a neural network? Preprint at arXiv https://arxiv.org/abs/1907.06374 (2019). · arXiv
- Saxe, A. Deep Linear Networks: A Theory of Learning in the Brain and Mind. Thesis, Stanford Univ. (2015).
- Tsividis, P. A., Pouncy, T., Xu, J., Tenenbaum, J. B. & Gershman, S. J. Human learning in Atari (AAAI, 2017).
- Cueva, C. J. & Wei, X.-X. Emergence of grid-like representations by training recurrent neural networks to perform spatial localization. Preprint at arXiv https://arxiv.org/abs/1803.07770 (2018). · arXiv
- Advani, M. S. & Saxe, A. M. High-dimensional dynamics of generalization error in neural networks. Preprint at arXiv https://arxiv.org/abs/1710.03667 (2017). · arXiv
Cited by
Ali, S., Abuhmed, T., El-Sappagh, S., Muhammad, K., Alonso-Moral, J. M., Confalonieri, R., Guidotti, R., Del Ser, J., Díaz-Rodríguez, N., & Herrera, F. (2023) · Information Fusion · doi:10.1016/j.inffus.2023.101805 · Cited by 1,762
Hulse, B. K., Haberkern, H., Franconville, R., Turner-Evans, D., Takemura, S.-Y., Wolff, T., Noorman, M., Dreher, M., Dan, C., Parekh, R., et al. (2021) · eLife · doi:10.7554/elife.66039 · Cited by 392
Brains and algorithms partially converge in natural language processing
Caucheteux, C., & King, J.-R. (2022) · Communications Biology · doi:10.1038/s42003-022-03036-1 · Cited by 318
“Following a recent methodology 33,42,44,46,46,50 – 56 , we address this issue by evaluating whether the activations of a large variety of deep language models linearly map onto those of 102 human brains.”
MethodA large and rich EEG dataset for modeling human visual object recognition
Gifford, A. T., Dwivedi, K., Roig, G., & Cichy, R. M. (2022) · NeuroImage · doi:10.1016/j.neuroimage.2022.119754 · Cited by 276
Neural population geometry: An approach for understanding biological and artificial neural networks
Chung, S., & Abbott, L. F. (2021) · Current Opinion in Neurobiology · doi:10.1016/j.conb.2021.10.010 · Cited by 270
“One often-heard objection to the use of ANNs in modeling neural circuits is that ANNs merely replace one complicated system with an equally complicated system [24].”
Method
Metadata
- Authors
- Saxe, A., Nelli, S., & Summerfield, C.
- Affiliations
- University of Oxford (GB)
- Corresponding author
- Saxe, Andrew
- Nelli, Stephanie
- Summerfield, Christopher
- Venue
- Nature Reviews Neuroscience
- Published
- 2020-11-16
- Volume
- 22
- Issue
- 1
- Pages
- 55-67
- ISSN
- 1471-003X (ISSN-L)
- 1471-0048
- OpenAlex
- W3100035092
- Cited by
- 397
- References
- 150
- Citations by year
- 2026: 27 · 2025: 72 · 2024: 89 · 2023: 88 · 2022: 66 · 2021: 47 · 2020: 7
- FWCI
- 6.99
- Type
- journal article
- Language
- en
- Open access
- closed
- Licence
- www.springer.com/tdm (tdm)
- www.springer.com/tdm (vor)
- Locations
- Nature reviews. Neuroscience (publishedVersion)
- PubMed (publishedVersion)
- Funders
- Wellcome Trust (216386/Z/19/Z)
- Topics
- Life Sciences › Neuroscience › Cognitive Neuroscience › Neural dynamics and brain function
- Physical Sciences › Computer Science › Artificial Intelligence › Neural Networks and Applications
- Life Sciences › Neuroscience › Cognitive Neuroscience › Memory and Neural Mechanisms
- Keywords
- deep learning, deep neural network, systems neuroscience, neural computation, neural representations, perception, cognition, learning dynamics, neural code
- MeSH terms
- Deep Learning, Brain, Humans, Neurosciences, Neural Networks, Computer
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: If deep learning is the answer, what is the question? https://app.cochat.ai/papers/10.1038/s41583-020-00395-8
- The paper
- Saxe, A., Nelli, S., & Summerfield, C. (2021). If deep learning is the answer, what is the question? Nature Reviews Neuroscience. https://doi.org/10.1038/s41583-020-00395-8
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.