Brain-like functional specialization emerges spontaneously in deep neural networks
Open accessPublished version of a preprint
The DOI matches its Crossref record; 58 of the 58 references it cites were found in a registry or an index.
Abstract
The human brain contains multiple regions with distinct, often highly specialized functions, from recognizing faces to understanding language to thinking about what others are thinking. However, it remains unclear why the cortex exhibits this high degree of functional specialization in the first place. Here, we consider the case of face perception using artificial neural networks to test the hypothesis that functional segregation of face recognition in the brain reflects a computational optimization for the broader problem of visual recognition of faces and other visual categories. We find that networks trained on object recognition perform poorly on face recognition and vice versa and that networks optimized for both tasks spontaneously segregate themselves into separate systems for faces and objects. We then show functional segregation to varying degrees for other visual categories, revealing a widespread tendency for optimization (without built-in task-specific inductive biases) to lead to functional specialization in machines and, we conjecture, also brains.
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
Figures
![Fig. 1. Distinct face and object representations in singly trained CNNs while a dual-task CNN performs well. (A) Three networks with VGG16 architecture (left) were optimized, one on face identity categorization (Face CNN in red), one on object categorization (Object CNN in orange), and one on both tasks simultaneously (dual-task CNN in gray). (B) Decoding accuracy of held-out face identities and held-out object categories using activation patterns extracted from the penultimate layer [i.e., FC2 in (A)] of the Face CNN and the Object CNN. The Face CNN outperforms the Object CNN in face decoding and vice versa for object decoding. Thus, the representations optimized for each task do not naturally support the other. The dashed gray line indicates chance level (1%). Error bars indicate SEM across classification folds. (C) A dual-task CNN optimized on both tasks performed and the separate networks (% top 1 accuracy on the test set). Error bars denote 95% confidence interval (CI) bootstrapped across classes and stimuli.](https://figures.cochat.ai/paper-figures/f7fd2e615ccf47089c9fb1193e6a0236/brain-like-functional-specialization-emerges-spontaneously-in-deep-neural-figure-1.jpg)





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.
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?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Using this approach, known as neuroconnectionism, numerous studies have identified shared representational structure across brains and ANN models, as well as factors influencing this alignment [5–11]”
From The Umwelt Representation Hypothesis: rethinking Universality (2026) · cites this paper as [7]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“Modular structure has been shown to spontaneously emerge in some end-to-end neural network systems in domains other than language [e.g., 203, 204]”
From Dissociating language and thought in large language models (2024) · cites this paper as [204]
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
Very Deep Convolutional Networks for Large-Scale Image Recognition
Simonyan, K., & Zisserman, A. (2014) · arXiv · doi:10.48550/arxiv.1409.1556 · Cited by 114,148 · cited 3×
“We used similar training parameters as suggested in the original VGG paper (20): stochastic gradient descent (SGD) with momentum with an initial learning rate of 10−3, a weight decay of 10−4, and momentum of 0.”
MethodImageNet: A large-scale hierarchical image database
Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., & Fei-Fei, L. (2009) · 2009 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2009.5206848 · Cited by 65,333 · cited 1×
“MATERIALS AND METHODS Task-optimizing CNNs on face and object recognition To measure object categorization performance in a network trained on object categorization only (Object CNN), we trained a randomly initialized VGG16 network (20) on 423 manually sampled categories of the ILSVRC-2012 database (53).”
MethodRich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation
Girshick, R., Donahue, J., Darrell, T., & Malik, J. (2014) · 2014 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2014.81 · Cited by 32,684
The fusiform face area: a module in human extrastriate cortex specialized for face perception
Kanwisher, N., McDermott, J., & Chun, M. M. (1997) · The Journal of Neuroscience · doi:10.1523/jneurosci.17-11-04302.1997 · Cited by 7,738 · cited 1×
“Here, we test this third hypothesis for one of the best-established cases of functional specialization in the brain: the visual recognition of faces (5, 7, 8).”
BackgroundSelf-citationVGGFace2: A Dataset for Recognising Faces across Pose and Age
Cao, Q., Shen, L., Xie, W., Parkhi, O. M., & Zisserman, A. (2018) · 2018 13th IEEE International Conference on Automatic Face & Gesture Recognition (FG 2018) · doi:10.1109/fg.2018.00020 · Cited by 3,032
Food-101 – Mining Discriminative Components with Random Forests
Bossard, L., Guillaumin, M., & Van Gool, L. (2014) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-10599-4_29 · Cited by 2,072 · cited 2×
“To test whether we would find segregation for other natural tasks, we additionally trained two dual-task networks: (i) We trained a network on object categorization and food discrimination using the Food101 ( 57 ) dataset.”
MethodPerformance-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
Learning Face Representation from Scratch
Yi, D., Lei, Z., Liao, S., & Li, S. Z. (2014) · arXiv (Cornell University) · doi:10.48550/arxiv.1411.7923 · Cited by 1,521
Understanding Neural Networks Through Deep Visualization
Yosinski, J., Clune, J., Nguyen, A. M., Fuchs, T. J., & Lipson, H. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1506.06579 · Cited by 1,358 · cited 2×
“In addition, periodic (every 10 iterations) blurring and random jittering were applied to the pixel values (50) to enhance visualization,”
BackgroundMethodDo Better ImageNet Models Transfer Better?
Kornblith, S., Shlens, J., & Le, Q. V. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.00277 · Cited by 1,325 · cited 2×
“Prior work suggests that this may not necessarily be the case: CNNs optimized for generic object recognition transfer well to many other tasks (12–15), including fine-grained discrimination within a category (13).”
Background
Research works printed but not linked: 6
- S. Finger Origins Of Neuroscience: A History Of Explorations Into Brain Function (Oxford Univ. Press 2001).
- C. R. Gallistel The replacement of general-purpose learning models with adaptively specialized learning modules in The Cognitive Neurosciences M. S. Gazzaniga Ed. (MIT Press ed. 2 2000) pp. 1179–1191.
- B. Long G. Kachergis K. Agrawal M. C. Frank Detecting social information in a dense dataset of infants’ natural visual experience (PsyArXiv 2020); https://psyarxiv.com/z7tdg/.
- 10.1037/0033-295x.98.2.164 · doi:10.1037/0033-295x.98.2.164 · Paper page
- A. Nayebi N. C. L. Kong C. Zhuang J. L. Gardner A. M. Norcia D. L. K. Yamins Unsupervised models of mouse visual cortex. bioRxiv 448730 [Preprint] (2021); www.biorxiv.org/content/10.1101/2021.06.16.448730v1. · doi:10.1101/2021.06.16.448730v1
- A. Krizhevsky I. Sutskever G. E. Hinton in Advances in Neural Information Processing Systems (MIT Press 2012) pp. 1097–1105.
Cited by
Items 101–137 of 137
Akbari, E., & Dobs, K. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2025.12.19.695398 · Cited by 1
“We chose a 20% lesion proportion because prior work (33) found that this level of ablation produces strong and interpretable task-specific effects while avoiding excessive overlap between critical filters across tasks.”
InfluentialSelf-citationSelf-supervised learning yields representational signatures of category-selective cortexPreprint
Janini, D., & Cichy, R. M. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.02.09.704031 · Cited by 1
van Dyck, L. E., & Dobs, K. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.02.16.706115 · Cited by 3
Self-citationTrognon, A. (2022) · theses.fr (ABES) · doi:10.70675/a61baf2dz0493z4668z80b0za8538881eb16 · Cited by 4
Cortical face processing systems
Yovel, G., Duchaine, B., & Freiwald, W. A. (2024) · Elsevier eBooks · doi:10.1016/b978-0-12-820480-1.00084-x
Abed, A. A. (2025) · Journal of Information Systems Engineering & Management · doi:10.52783/jisem.v10i46s.9175
The Representational Organization of Static and Dynamic Visual Features in the Human Cortex
Karimi, H., Wang, J., & Anzellotti, S. (2025) · Journal of Neuroscience · doi:10.1523/jneurosci.1164-24.2025
Sparse components distinguish visual pathways & their alignment to neural networksPreprint
Marvi, A. I., Kanwisher, N., & Khosla, M. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2510.08858
“…(DNNs) trained for object classification have been shown to exhibit similar internal activations (Yamins et al., 2014), functional selectivities (Blauch et al., 2022; Dobs et al., 2022), and behavioral capabilities (Dobs et al., 2023; Rajalingham et al., 2018) to the human ventral visual pathway.”
Self-citationFrom Networks to Circuits: A Structured Approach to AI Interpretability
Kowalska, B., & Kwaśnicka, H. (2025) · Communications in computer and information science · doi:10.1007/978-3-032-10209-6_26
From Tasks to Topology: Dorsal and Ventral Streams Emerge in Optimized Neural NetworksPreprint
Reza, T., Jordan, E., Luo, S. T. S., Patel, K., Tang, S. J., & Niemeier, M. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.11.16.688720
“Such modular organization with functional specialization has been shown to emerge in ventral substreams through task optimization even without architectural constraints 26 .”
CNNs Generalize Numerosity Across Naturalistic Stimuli Without Single-Unit Selectivity
Brandes, T. N. H. R., Groot, J., & Alhama, R. G. (2025) · UvA-DARE (University of Amsterdam)
Optimization of neural response in the primate dorsal visual pathway
Kreß, A. (2025) · open_UMR Marburg DSpace 10.0 (Philipps-Universität Marburg) · doi:10.17192/z2025.0657
Liu, M., & Wang, L. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2025.12.12.693899
“A growing body of research has demonstrated alignment between ANNs and human neural processing of visual objects 26,27,38 , as well as the reproduction of many human-like behaviors in face recognition tasks 6–8 , 39 .”
Cajal’s Neuron Doctrine: 1898–
(2026) · Royal Society of Chemistry eBooks · doi:10.1039/9781837679690-00540
Stolle, C. M., Yu, R., & Huang, Y. (2026) · Computers in Human Behavior Artificial Humans · doi:10.1016/j.chbah.2026.100250
Lu, X., & Jiang, Y. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.02.01.703162
Self-supervised learning yields representational signatures of category-selective cortex
Janini, D., & Cichy, R. M. (2026) · Communications Biology · doi:10.1038/s42003-026-10843-3
“Recently, analogous procedures have revealed category-selective representations in category-supervised 42,53 – 55 and self-supervised neural networks 49 – 51 .”
Neural Tuning for Ordinal Processing: Convergent Patterns in Human Brains and Artificial Networks
Hofstetter, S., Daghlian, M., & Dumoulin, S. O. (2026) · Journal of Neuroscience · doi:10.1523/jneurosci.1237-25.2026
Chen, C., Yang, Z., & Wang, X. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.02.11.705311
“This limitation applies to both audition and vision (Rajalingham et al., 2018; Zhang et al., 2018; Peters & Kriegeskorte, 2021; Dobs et al., 2022; Feather et al., 2023; Sun et al., 2025b; Schulze Buschoff et al., 2025).”
Spontaneous emergence of topographic organization in a multistream convolutional neural networkPreprint
Tamura, H. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.02.23.707577
“Recently, convolutional neural networks (CNNs) have been used to investigate the computational significance of functional organization in the brain 25 .”
Sasaki, K. S. (2026) · Journal of Neurophysiology · doi:10.1152/jn.00523.2025
Human-like sequential sound-to-meaning transfer drives artificial speech comprehensionPreprint
Zhang, S., Li, S., Yang, R., Chen, G., Tian, X., Wang, Q., & Fang, F. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.05.13.723203
Experience Alone Can Generate Human Face Specialization: Evidence From Deep Learning Models
Guy, N., Rosemblaum, M., & Yovel, G. (2026) · Open Mind · doi:10.1162/opmi.a.363
“Remarkably, the evidence suggests that they do: deep convolutional neural networks (DCNNs) trained predominantly on upright faces from a particular race exhibit superior recognition performance for faces of that race (Dobs et al., 2022, 2023; Tian et al., 2021).”
Associative Emotional Learning in Convolutional Neural Networks
Leem, S., Keil, A., Ding, M., & Fang, R. (2026) · Neural Computation · doi:10.1162/neco.a.1580
Systematic sampling of emotional experience is needed to model human brain function
Soderberg, K. M., Ma, Y., Riels, K. M., & Kragel, P. A. (2026) · Trends Open · doi:10.1016/j.treopn.2026.07.010
Chen, G., & Liu, X. (2026) · Electronics Letters · doi:10.1049/ell2.70694
Dong, L., Sun, L., & Jia, D. (2026) · Clinical Radiology · doi:10.1016/j.crad.2026.107468
Wang, L., Zhang, X., Xie, P., Zhao, S., & Hu, X. (2026) · NeuroImage · doi:10.1016/j.neuroimage.2026.122202
Abdelhedi, H., Bakhtiari, S., & Jerbi, K. (2026) · bioRxiv · doi:10.64898/2026.09.08.750266
“…optimization can impose specialized representational demands relative to object recognition (Abudarham et al., 2021), while task optimization more generally can shape functionally specialized representations without requiring completely segregated representational systems (Dobs et al., 2022, 2023).”
InfluentialSelf-Organized Context Dependent Processing in Neural Networks
GuangFu, H., Chen, Y., Qin, S., Alexandre, F., & Yu, S. (2026) · Cognitive Computation · doi:10.1007/s12559-026-10624-4
“And relative to Dobs et al. (2022) and Bakhtiari et al. (2021) [48, 49 ], who document emergent specialization for low-level sensory features, our work addresses the higher-level decoupling of contextual and sensory information underlying flexible stimulus-response mappings.”
Neural Networks for Fashion Image Classification and Visual Search
Rajak, K. K., & Sharma, D. P. K. (2026) · Journal of Recent Innovation in Science and Technology · doi:10.70454/jrist.020102
Lateral recurrence dynamically refines visual representations for task performancePreprint
Farzmahdi, A., Adeli, H., Boran, W., King, C., & Kriegeskorte, N. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.10.02.756219
Trait judgements of faces using a task without word labels
Xu, Y., Keleş, Ü., Lin, C., & Adolphs, R. (2026) · Nature Human Behaviour · doi:10.1038/s41562-026-02605-2
Learning Functional Specialisation for Bimanual Robotic ControlPreprint
Li, J., Kuhlmann, L., Friedman, J., & Kowadlo, G. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2610.33947
“Different training objectives encourage components to develop distinct functional priorities (Dobs et al., 2022; Rinaldo et al., 2025), while repeated practice reinforces hand-specific motor competencies (Marcori et al., 2019; Dexheimer et al., 2024; Arac et al., 2026).”
The transformer as a candidate canonical cortical microcircuitPreprint
Shah, Y. N., & Yamins, D. L. K. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.11.01.763457
Metadata
- Authors
- Dobs, K., Martinez, J., Kell, A. J. E., & Kanwisher, N.
- Affiliations
- McGovern Institute for Brain Research (US)
- Mortimer B. Zuckerman Mind Brain Behavior Institute (US)
- Corresponding author
- Dobs, Katharina
- Venue
- Science Advances
- Published
- 2022-03-18
- Volume
- 8
- Issue
- 11
- Pages
- eabl8913-eabl8913
- ISSN
- 2375-2548 (ISSN-L)
- PMC
- PMC8926347
- OpenAlex
- W4225899846
- Cited by
- 129
- References
- 52
- Citations by year
- 2026: 25 · 2025: 24 · 2024: 33 · 2023: 30 · 2022: 16
- FWCI
- 13.28
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- cc-by-nc
- Locations
- Science Advances (publishedVersion, cc-by-nc)
- PubMed Central (publishedVersion, cc-by-nc)
- PubMed (publishedVersion)
- Funders
- Eunice Kennedy Shriver National Institute of Child Health and Human Development (DP1HD091947)
- Topics
- Life Sciences › Neuroscience › Cognitive Neuroscience › Face Recognition and Perception
- Physical Sciences › Computer Science › Artificial Intelligence › Neural Networks and Applications
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Keywords
- functional specialization, functional segregation, face recognition (cognition), object recognition, deep neural network, computational optimization, inductive bias, brain organization, visual 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.
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
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.