license: apache-2.0
tags:
- pytorch
- computer-vision
- self-supervised-learning
- simclr
- resnet18
- egocentric-vision
- eccentricity
- visual-neuroscience
- vedb
- arxiv:2607.19316
VEDB SimCLR ResNet-18 — NSD Voxelwise Encoding Models
This repository contains subject-specific voxelwise encoding-model fits from:
Diaz, D. M., & Henderson, M. M. (2026). Eccentricity-Constrained CNN Training Reveals Adaptive Information Coding Around the Visual Field. Proceedings of the 9th Conference on Cognitive Computational Neuroscience.
DOI: 10.32470/0416gfsq
arXiv: 2607.19316
The encoding models were fit to fMRI responses from the Natural Scenes Dataset (NSD) using representations extracted from four VEDB-pretrained SimCLR ResNet-18 models:
- Baseline
- Fovea-Gaze
- Periph
- Periph-NF
Repository Structure
baseline/
fovea-gaze/
periph/
periph-nf/
Each folder contains voxelwise encoding-model fits for NSD subjects S1–S8.
Example:
fovea-gaze/
├── NSD_S1_resnet18-Fovea-Gaze_concat.npy
├── NSD_S2_resnet18-Fovea-Gaze_concat.npy
├── ...
└── NSD_S8_resnet18-Fovea-Gaze_concat.npy
Encoding Models
For each subject and visual-field condition, features were extracted from:
conv1
layer1.1
layer2.1
layer3.1
layer4.1
avgpool
Layer features were dimensionally reduced with PCA, concatenated, and used to fit voxelwise L2-regularized linear regression (ridge) encoding models.
Each .npy file contains a saved Python dictionary including:
- fitted voxelwise
weights - held-out
r2 - held-out
corr - candidate
lambdas best_lambda_inds- voxel mask and index information
- voxel noise ceilings
- subject and model metadata
Loading a Fit
import numpy as np
fit = np.load(
"baseline/NSD_S1_resnet18-Baseline_concat.npy",
allow_pickle=True
).item()
weights = fit["weights"]
r2 = fit["r2"]
corr = fit["corr"]
best_lambda_inds = fit["best_lambda_inds"]
Related Models
The pretrained SimCLR checkpoints used to generate these representations are available in the Eccentricity-Constrained SimCLR Models (VEDB) Hugging Face collection.
Release Status
Encoding-model fits are available now. Additional documentation and analysis code are forthcoming.
Citation
@inproceedings{diaz2026eccentricity,
author = {Diaz, Dylan M. and Henderson, Margaret M.},
title = {Eccentricity-Constrained CNN Training Reveals Adaptive Information Coding Around the Visual Field},
booktitle = {Proceedings of the 9th Conference on Cognitive Computational Neuroscience},
address = {New York, NY, USA},
year = {2026},
doi = {10.32470/0416gfsq}
}