OSCR

Multidimensional Feature Tuning in Category Selective Areas of Human Visual Cortex.

Code ↔ Paper

11 matches between paragraphs of the paper and lines of its authors' code, computed by the harvester (lexical-v1). Click a colored paragraph or line to see its counterpart.

The 11 matches · 1 of them tie a paragraph to a whole file, not to given lines: a weak match, whose lines are not tinted
  1. [1] § Materials and Methods › Data-driven voxel decomposition › Estimating optimal dimensionality ↔ main_analyses.ipynb, lines 54–74 · score 0.83 · bi cross validation, pseudo inverse, 1–100, optimal dimensionality, matrix factorization, fMRI
  2. [2] § Materials and Methods › Data-driven dimension labeling › Representational sparseness ↔ main_analyses.ipynb, lines 139–147 · score 0.78 · L2 norm, tuned equally, dimension weights, single dimension, multidimensional tuning, L1
  3. [3] § Materials and Methods › Data-driven dimension labeling › Voxel-wise encoding models ↔ src/roidims/encoding.py, lines 163–178 · score 0.69 · fractional ridge regression, fold cross validation, intercept, model, tuned, training
  4. [4] § Materials and Methods › Data-driven dimension labeling › Voxel-wise encoding models ↔ main_analyses.ipynb, lines 157–165 · score 0.65 · fractional ridge regression, avoid overfitting, cross validation, fold, held, squares
  5. [5] § Materials and Methods › fMRI data › Natural Scenes Dataset ↔ src/roidims/prepro.py, lines 206–229 · score 0.64 · preprocessed single trial, single trial voxel, repetition, voxel responses, NSD, brain
  6. [6] § Materials and Methods › Data-driven dimension labeling › Deep learning-based label selection ↔ src/roidims/interpret.py, lines 42–91 · score 0.63 · prompt templates, photo, picture, candidate, CLIP, embedded
  7. [7] § Materials and Methods › Data-driven voxel decomposition › Bayesian non-negative matrix factorization ↔ src/roidims/prepro.py, lines 287–308 · score 0.63 · global training minimum, baseline shifted, subtracting, ROI, voxels
  8. [8] § Materials and Methods › Data-driven voxel decomposition › Consensus approach ↔ src/roidims/consensus.py, lines 62–76 · score 0.62 · medoids clustering, consensus, density, medians, outlier, aggregate
  9. [9] § Materials and Methods › Data-driven voxel decomposition › Bayesian non-negative matrix factorization ↔ main_analyses.ipynb, lines 54–74 · score 0.60 · baseline shifted, training minimum, burn, enforce, leakage, ran
  10. [10] § Materials and Methods › Data-driven voxel decomposition › Estimating optimal dimensionality ↔ src/roidims/bcv.py, the whole file · a weak match · score 0.59 · bi cross validation, shuffled, error, ranks, rows, matrix
  11. [11] § Results › Data-driven voxel decomposition reveals multiple interpretable dimensions in category-selective areas ↔ src/roidims/interpret.py, lines 93–153 · score 0.52 · text embedding, image embeddings, candidate, cosine, weighted, scoring

Paper

Loaded from Europe PMC by your browser, not stored by OSCR: doi.org · Europe PMC

The paper is loaded when this pane is shown.

The authors' code

Jupyter notebook · 185 lines · 8 KB · no license · 4 matches

  1. # %% [markdown]
  2. # # Multidimensional feature tuning in category‑selective areas of human visual cortex
  4. # %%
  5. from scripts.run_prepro import run_prepro
  6. from scripts.run_bcv import run_bcv
  7. from scripts.run_bnmf import run_bnmf
  8. from scripts.run_interpret import run_interpret
  9. from scripts.run_encoding import run_encoding
  11. from roidims.bnmf import compute_evar_consistent, compute_sim_dims_groups
  12. from roidims.encoding import compute_corr_H_beta, selectivity_vs_sparseness
  14. from roidims.plotting import (
  15. fig_consistency,
  16. fig_top_imgs,
  17. fig_dprime_vs_beta,
  18. fig_wordclouds,
  19. fig_flatmaps_encoding_r2,
  20. fig_r2_bar,
  21. fig_flatmaps_encoding_betas,
  22. suppfig_k_optim,
  23. suppfig_consistency,
  24. suppfig_sim_dims,
  25. suppfig_sparseness,
  26. suppfig_noiseceilings,
  27. )
  29. # %% [markdown]
  30. # ## Background
  31. # What functional organization underlies both category-selective areas and continuous feature maps in high-level visual cortex?
  32. #
  33. # We addressed this question by applying a data-driven decomposition to fMRI responses from face-, body-, and scene-selective areas, revealing that both views reflect a common principle: multidimensional tuning that is clustered within areas yet distributed across cortex.
  35. # %% [markdown]
  36. # ## fMRI data
  37. # We used a subset of the Natural Scenes Dataset (NSD; Allen et al., 2022) including fMRI responses (7T) from four participants, each viewing 10,000 natural images.
  38. #
  39. # We defined FFA, PPA, and EBA using an independent functional localizer (fLoc; Stigliani et al., 2015), selecting voxels with a strong category preference (t > 2) and sufficient reliability (SNR > 0.2).
  41. # %%
  42. # Specify subjetcs and ROIs
  43. subjects = ["subj01", "subj02", "subj05", "subj07"]
  44. rois = ["FFA", "EBA", "PPA"]
  46. # %% [markdown]
  47. # ## 0. Preprocessing
  48. # Single-trial responses were preprocessed with GLMdenoise and ridge regression, z-scored within sessions, averaged across repetitions, and split 70/30 into training and test sets.
  50. # %%
  51. # Preprocessing
  52. run_prepro()
  54. # %% [markdown]
  55. # ## 1. Data-driven voxel decomposition
  56. # First, we applied non-negative matrix factorization to reveal which underlying representational dimensions capture the fMRI activity patterns.
  58. # %% [markdown]
  59. # ### Background: Bayesian non-negative matrix factorization (BNMF)
  60. # We decomposed voxel responses from each ROI into representational dimensions using BNMF (Schmidt et al., 2009; following Khosla et al., 2022).
  61. #
  62. # BNMF factorizes the image × voxel response matrix into a response matrix W (image × dimension) and a weight matrix H (dimension × voxel). The non-negativity constraint encourages sparse, additive, part-based dimensions, allowing voxels to participate in multiple dimensions simultaneously. The Bayesian extension adds exponential priors on W and H and uses Gibbs sampling to obtain robust posterior estimates.
  63. #
  64. # We ran 3,000 iterations (1,000 burn-in, every 5th sample retained) and baseline-shifted responses to enforce non-negativity, using the training minimum for both sets to prevent data leakage.
  66. # %% [markdown]
  67. # ### 1.1. Estimating the optimal dimensionality
  68. # We determined the optimal number of dimensions k* for each participant and ROI using bi-cross-validation (Owen & Perry, 2009).
  69. #
  70. # Bi-cross-validation partitions the data into four blocks. BNMF is trained on one block, and pseudo-inverse operations on two others are used to predict the held-out block. This was repeated 5 times across ranks 1–100 (step size 3), and k* was chosen as the rank minimizing average test error.
  72. # %%
  73. # Bi-cross-validation
  74. run_bcv(subjects, rois, k_min=1, k_max=100, k_steps=1, n_perms=5)
  76. # %%
  77. # Supp Fig 1: Optimal dimensionality
  78. suppfig_k_optim(subjects, rois)
  80. # %% [markdown]
  81. # ### 1.2. Finding reliable dimensions
  82. # To obtain reliable and generalizable dimensions, we used a two-step consensus approach:
  83. #
  84. # First, we ran BNMF 100 times with k* but random initializations per ROI, removed outlier runs, and aggregated stable dimensions via k-medoids clustering (Kotliar et al., 2019).
  85. #
  86. # Second, we matched dimensions across participants using pairwise correlations and a greedy selection procedure, retaining only dimensions consistent across all four participants (r > 0.3; Khosla et al., 2022).
  87. #
  88. # This yielded 8–20 consensus dimensions per ROI. To test the generalizability to unseen images, we projected the test set into the learned embedding using non-negative least squares regression.
  90. # %%
  91. # Specify inter-subject consistency threshold
  92. r_thresh = 0.3
  94. # %%
  95. # Consensus approach
  96. run_bnmf(subjects, rois, n_runs=100, r_thresh=r_thresh)
  98. # %%
  99. # Compute explained variance
  100. compute_evar_consistent(subjects, rois)
  102. # %%
  103. # Fig 2a: Consistency
  104. fig_consistency(rois, r_thresh)
  106. # %%
  107. # Supp Fig 2: Mean inter-participant consistency of all dimensions
  108. suppfig_consistency(rois, r_thresh)
  110. # %%
  111. # Fig 2b: Top images combined across subjects
  112. fig_top_imgs(subjects, rois)
  114. # %%
  115. # Supp Fig 3: Mean similarity matrices
  116. suppfig_sim_dims(subjects, rois)
  118. # %%
  119. # Representational similarity of all dimensions
  120. compute_sim_dims_groups(subjects, rois)
  122. # %% [markdown]
  123. # ## 2. Data-driven dimension labeling
  124. # To interpret each dimension, we used a behavioral online experiment. We collected data from N=38 participants, who viewed collages of the 20 highest-scoring images per dimension and provided 3–5 short labels describing what the images had in common. Labels were translated from German, cleaned of generic entries, and standardized.
  125. #
  126. # We then used CLIP-ViT-L/14 (Radford et al., 2021) to select the most representative label per dimension from the participant-generated candidate set. For each label, we computed the cosine similarity between its text embedding and all image embeddings, weighted by each dimension's response profile, then z-scored across labels within each dimension to highlight selectively associated labels. Top-ranked labels were visualized as word clouds, yielding concise, semantically meaningful descriptors validated across the full image set.
  128. # %% [markdown]
  129. # ### 2.1. Finding the most representative labels
  131. # %%
  132. # CLIP-based interpretation
  133. run_interpret(subjects, rois)
  135. # %%
  136. # Fig 3: Top labels for all subjects
  137. fig_wordclouds(subjects, rois)
  139. # %% [markdown]
  140. # ## 3. Testing the relationship between multidimensional tuning and category selectivity
  141. # To quantify category selectivity for the ROI's preferred category, we used a category selectivity index (d'), comparing mean responses to a preferred category against all others, weighted by response variance.
  142. #
  143. # To measure the extent of multidimensional tuning, we computed a sparseness index (Hoyer, 2004) based on the normalized L1/L2-norm ratio of each voxel's dimension weight profile, ranging from 0 (tuned equally to all dimensions) to 1 (tuned to a single dimension).
  145. # %%
  146. # Fig 3: Correlation between dimension tuning (a.u.) and category selectivity (d')
  147. fig_dprime_vs_beta(subjects, rois)
  149. # %%
  150. # Plot sparseness for different selectivity bins
  151. suppfig_sparseness(subjects, rois)
  153. # %%
  154. # Print correlation
  155. selectivity_vs_sparseness(subjects, rois)
  157. # %% [markdown]
  158. # ## 4. Voxel-wise encoding models
  159. # Finally, we fit voxel-wise encoding models to predict fMRI responses to held-out images from the learned dimensions, revealing the cortical topography of this organization.
  160. #
  161. # We used fractional ridge regression with cross-validation to avoid overfitting, projected test images into the learned BNMF space with non-negative least squares, and evaluated prediction performance on held-out data. Significance was assessed with permutation tests and FDR correction, and noise ceilings were computed to benchmark model performance.
  163. # %%
  164. # Fractional ridge regression
  165. run_encoding(subjects, rois, n_folds=10, n_perms=3000)
  167. # %%
  168. # Print correlation between ROI coefficient weights
  169. compute_corr_H_beta(subjects, rois)
  171. # %%
  172. # Fig 4a: Prediction performance maps
  173. fig_flatmaps_encoding_r2
  175. # %%
  176. # Fig 4b: Prediction performance across ROIs
  177. fig_r2_bar(subjects, rois)
  179. # %%
  180. # Supp Fig 6: Voxel-wise prediction performance vs. noise ceiling estimate
  181. suppfig_noiseceilings(subjects, rois)
  183. # %%
  184. # Fig 5: Individual dimension tuning maps
  185. fig_flatmaps_encoding_betas(subjects, rois)

main_analyses.ipynb at commit 05bce4d, no license · at the source

Overview

Authors: Leonard E van Dyck1,2,3, Martin N Hebart2,4,3, Katharina Dobs1,3
ORCID iDs: Leonard E van Dyck, Martin N Hebart, Katharina Dobs
  1. Department of Computer Science, Justus Liebig University Giessen, Giessen 35390, Germany
  2. Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig 04103, Germany
  3. Center for Mind, Brain and Behavior (CMBB), Universities of Marburg, Giessen, and Darmstadt, Marburg 35032, Germany
  4. Department of Medicine, Justus Liebig University Giessen, Giessen 35390, Germany
Institutions: Justus-Liebig-Universität Gießen (Germany); Max Planck Institute for Human Cognitive and Brain Sciences (Germany)
Journal: The Journal of neuroscience : the official journal of the Society for Neuroscience, volume 46, issue 31, article e0038262026
Dates: received 7 January 2026; accepted 18 May 2026; published online 17 June 2026; in print 5 August 2026
Type: Research article · Language: English
License: CC BY
Identifiers: DOI 10.1523/jneurosci.0038-26.2026 · PMID 42309813 · PMCID PMC13446086 · OpenAlex W4411371039
Open access: hybrid, a free copy (OpenAlex)
Preprint: doi:10.1101/2025.06.17.659578
Status: code verified
Categories: fMRI (modality), human (organism), systems (subfield)
Methods: Statistics, Smoothing, state filtering, decompositions, Machine learning, Preprocessing, fMRI & imaging
Keywords: category selectivity, dimensions, distributed processing, fMRI, functional organization, visual cortex
MeSH: Pattern Recognition, Visual*, Visual Cortex*, Adult, Brain Mapping, Female, Humans, Image Processing, Computer-Assisted, Magnetic Resonance Imaging, Male, Photic Stimulation, Young Adult (* major topic)
Topic: Neural Networks and Applications (Artificial Intelligence, Computer Science), according to OpenAlex
Funding: European Research Council (101039712, 101117441, ERC-2021-StG-101039712, ERC-2023-STG-101117441); Hessisches Ministerium für Wissenschaft und Kunst (Excellence Cluster EXC3066 "The Adaptive Mind&quot, LOEWE Start Professorship); Studienstiftung des Deutschen Volkes (Doctoral Scholarship); Deutsche Forschungsgemeinschaft (222641018-SFB/TRR 135 TP)
Citations: cited by 1 paper (Europe PMC); 109 references in the paper

Abstract

Two prominent accounts describe the functional organization of human high-level visual cortex. A categorical view emphasizes category-selective areas, while a dimensional view highlights continuous feature maps spanning these areas. Here, we asked whether these two views reflect complementary expressions of the same underlying organization. Using a data-driven decomposition of fMRI responses from human participants (female and male) in face-, body-, and scene-selective areas, we identified spatially overlapping activity patterns that were shared across individuals. Each area encoded multiple interpretable dimensions capturing both finer within-category and coarser between-category distinctions, even in the most category-selective voxels. These dimensions formed distinct clusters within category-selective areas but extended as distributed maps across visual cortex. Together, these findings reveal an underlying organization that links category-selective areas to continuous feature maps, thereby reconciling categorical and dimensional accounts of high-level visual cortex.

Reproduced under the paper's license (CC BY), from the paper cited above.

Repository

Its files are read in the Code ↔ Paper reader above, with 11 matches between paragraphs and lines of code.

levandyck/roidims

License: none: the authors keep all their rights
State: the link answers, verified on 27 September 2026
Evidence: files inventoried
Commit: 05bce4dfcd26ec296da8cb5d914e6ff153d0ab3a, 21 July 2026
Languages: Python (17), Jupyter (1)
Size: 20 files, 18 scripts
Software Heritage: not archived
Found in: “Data Availability”
Holds: README, environment (requirements.txt, setup.py), 1 notebook
Not found: license file, CITATION.cff, tests, continuous integration, documentation
Tools: NumPy (11 files), scikit-learn (6 files), pandas (5 files), NiBabel (4 files), Matplotlib (3 files), SciPy (3 files), h5py (2 files), pycortex (2 files), Pillow (1 file), PyTorch (1 file), seaborn (1 file), statsmodels (1 file)
Availability: 1 check, the latest on 27 September 2026: the link answers
  • 27 September 2026: the link answers
19 files
At the source: github.com/levandyck/roidims

The paper's code and data availability statement is in the Data section.

Tracing map

Proposed by the machine: these links were found in the paper and verified at the source, without human review. The map will receive a Zenodo DOI once one of the paper's authors has validated it with their ORCID.

What the map holds:

  • 1 repository of the authors' code, each at its verified commit, with its license and how the link was found in the paper;
  • 18 scripts, each with its path and the digest of its content;
  • 11 matches between paragraphs of the paper and lines of the code (method lexical-v1);
  • neither the text of the paper nor the code itself.

Its JSON (tracing-map.json) is deposited on Zenodo with its DOI once the map is validated.

Data

No dataset and no data link were found in the paper.

Data Availability

The data supporting our analyses were obtained from the publicly available Natural Scenes Dataset (http://naturalscenesdataset.org/). The Python code (version 3.8.20) used for data analysis and visualization is publicly available on GitHub (https://github.com/levandyck/roidims).

Reproduced under the paper's license (CC BY), from the paper cited above.

Versions

The history of this record: each version stored by the harvester or made by a correction of its authors or of the maintainers of its code, and what changed in its facts. The texts of the paper (its abstract, its availability statements) are not part of it; versions that changed only those are not listed.

Version 1, 27 September 2026: the first record

Recorded: type, language, journal, volume, issue, pages, dates, 3 authors, 6 keywords, 11 MeSH terms, 4 funders, 103 references.

Cite

This paper

van Dyck, L. E., Hebart, M. N., & Dobs, K. (2026). Multidimensional Feature Tuning in Category Selective Areas of Human Visual Cortex. The Journal of neuroscience : the official journal of the Society for Neuroscience, 46(31), e0038262026. https://doi.org/10.1523/jneurosci.0038-26.2026

BibTeX

@article{vandyck2026multidimensional,
author = {van Dyck, Leonard E and Hebart, Martin N and Dobs, Katharina},
title = {{Multidimensional Feature Tuning in Category Selective Areas of Human Visual Cortex}},
journal = {The Journal of neuroscience : the official journal of the Society for Neuroscience},
year = {2026},
month = aug,
volume = {46},
number = {31},
pages = {e0038262026},
publisher = {Society for Neuroscience},
issn = {0270-6474},
doi = {10.1523/jneurosci.0038-26.2026},
url = {https://doi.org/10.1523/jneurosci.0038-26.2026},
pmid = {42309813},
pmcid = {PMC13446086}
}

RIS

TY - JOUR
AU - van Dyck, Leonard E
AU - Hebart, Martin N
AU - Dobs, Katharina
TI - Multidimensional Feature Tuning in Category Selective Areas of Human Visual Cortex
T2 - The Journal of neuroscience : the official journal of the Society for Neuroscience
J2 - J Neurosci
PY - 2026
DA - 2026/08/05
VL - 46
IS - 31
SP - e0038262026
SN - 0270-6474
PB - Society for Neuroscience
DO - 10.1523/jneurosci.0038-26.2026
UR - https://doi.org/10.1523/jneurosci.0038-26.2026
LA - en
ER -

CSL-JSON

{
"id": "10.1523/jneurosci.0038-26.2026",
"type": "article-journal",
"title": "Multidimensional Feature Tuning in Category Selective Areas of Human Visual Cortex",
"container-title": "The Journal of neuroscience : the official journal of the Society for Neuroscience",
"author": [
{
"family": "van Dyck",
"given": "Leonard E"
},
{
"family": "Hebart",
"given": "Martin N"
},
{
"family": "Dobs",
"given": "Katharina"
}
],
"container-title-short": "J Neurosci",
"volume": "46",
"issue": "31",
"page": "e0038262026",
"DOI": "10.1523/jneurosci.0038-26.2026",
"PMID": "42309813",
"PMCID": "PMC13446086",
"ISSN": "0270-6474",
"publisher": "Society for Neuroscience",
"URL": "https://doi.org/10.1523/jneurosci.0038-26.2026",
"language": "en",
"issued": {
"date-parts": [
[
2026,
8,
5
]
]
}
}

The tracing map gets a citation of its own once an author has validated it and it has a DOI.

Similar papers

The papers with a page that share the most with this one: the tools found in their code, their categories, datasets, cited references and authors, the rarest counting most.

[1] doi:10.1038/s42003-026-10843-3 [code]
Self-supervised learning yields representational signatures of category-selective cortex.
Journal: Communications biology
In common: 25 references
[2] doi:10.1162/imag.a.1309 [code]
Probing the content of semantic representations in body-selective regions.
Journal: Imaging neuroscience (Cambridge, Mass.)
In common: pycortex, h5py, Pillow, 9 other tools, 11 references
[3] doi:10.1038/s41467-026-73938-9 [code]
Higher visual areas act like domain-general filters with strong selectivity and functional specialization.
Journal: Nature communications
In common: Pillow, NiBabel, PyTorch, 4 other tools, 15 references
[4] doi:10.1038/s41467-026-76098-y [code]
A single computational objective can produce specialization of streams in visual cortex.
Journal: Nature communications
In common: h5py, Pillow, NiBabel, 8 other tools, 12 references
[5] doi:10.1038/s42003-026-10169-0 [code]
Shared representations in brains and models reveal a two-route cortical organization during scene perception.
Journal: Communications biology
In common: h5py, Pillow, NiBabel, 8 other tools, 8 references
[6] doi:10.1167/jov.26.5.7 [code]
Representations in vision and language converge in a shared, multidimensional space of perceived similarities.
Journal: Journal of vision
In common: pycortex, h5py, Pillow, 9 other tools, 5 references
[7] doi:10.1038/s41593-026-02285-1 [code]
Fixation duration on natural scenes is explained by memory encoding not processing demand.
Journal: Nature neuroscience
In common: h5py, Pillow, statsmodels, 7 other tools, 2 references, author Martin N Hebart
[8] doi:10.1038/s41467-026-73153-6 [code]
Latent neural architecture organising shared aesthetic evaluations of visual artworks.
Journal: Nature communications
In common: h5py, Pillow, PyTorch, 6 other tools, fMRI, 6 references
[9] doi:10.7554/elife.107933 [code]
Modality-agnostic decoding of vision and language from fMRI.
Journal: eLife
In common: h5py, Pillow, NiBabel, 8 other tools, fMRI, 4 references
[10] doi:10.1162/nol.a.271 [code]
Compositional Complexity in Text and Images.
Journal: Neurobiology of language (Cambridge, Mass.)
In common: Pillow, NiBabel, PyTorch, 6 other tools, fMRI, 6 references

Contribute

The authors of this paper can claim it, correct its record and validate its tracing map, and the maintainers of its code (its owner, or a public member of its organization) correct what it says of their repository; anyone signed in can ask for its removal. Every request goes to OSCR's own machine, which answers it; your account page follows them.

Sign in with ORCID to claim this paper as one of its authors, correct its record or validate its tracing map: when the paper's metadata lists your ORCID iD, you are recognized at once. Maintainers of its code: sign in with GitHub, then claim the repository on your account page.

Request its removal

To ask OSCR to remove this record, the copies of its authors' scripts or its tracing map, use the removal request page: signed in, you say who you are, what to remove and why, then review and confirm the request. Published rules decide every request (how).

Discussion, reproductions, activity

Discussion: questions and error reports about this paper and its code, from signed-in readers and its authors. It opens with sign-in.

Reproductions: reports from readers who ran the authors' code: what they reproduced, with which environment, commit and data. It opens with sign-in.

Activity: what happens around this paper: new versions of its record, its map's validation, discussions and reproductions. It opens with sign-in.