Predictive coding yields a brain-like code for shape from motion
- 1Massachusetts Institute of Technology
Abstract
Interacting with the world relies on knowing its three-dimensional structure, yet vision receives a stream of two-dimensional images. Recovering 3D geometry from dynamic input is therefore foundational to vision, but lacks a computational account. Here we identify one, using artificial neural networks to predict human behavior and neuronal responses from the two major visual pathways in macaques. Using videos of camouflaged objects whose shape is revealed through motion, we found both pathways carry information for motion-defined surfaces, while neurons in parietal cortex most closely track what humans perceive. Only predictive coding video models--trained to fill in missing content in natural videos--reproduced both neuronal responses and human behavior. Predictive coding thus emerges as a principle that yields a brain-like geometric world model of the physical world.
Related
- Local learning is all you need: brain-aligned predictive coding networks
- A neuro-inspired computational framework for AGI: predictive coding, active inference and free energy minimisation.
- Dynamic Semantic–Prosodic Integration in Emotional Speech Perception: From Channel Dominance to Precision-Weighted Predictive Coding
- Predictive coding networks capture human neural representations missing in supervised DNNs
- Predictive coding unifies neural and behavioral signatures of contextual facilitation
BibTeX
@misc{bai2026predictive,
title = {{Predictive coding yields a brain-like code for shape from motion}},
author = {Y. H. Bai and T. P. O'Connell and Y. Friedman and A. Ayvazian-Hancock and H. Maver and J. B. Tenenbaum and J. DiCarlo},
howpublished = {bioRxiv},
year = {2026},
month = may,
doi = {10.64898/2026.05.13.724755},
url = {https://www.biorxiv.org/content/10.64898/2026.05.13.724755v2},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - Predictive coding yields a brain-like code for shape from motion
AU - Y. H. Bai
AU - T. P. O'Connell
AU - Y. Friedman
AU - A. Ayvazian-Hancock
AU - H. Maver
AU - J. B. Tenenbaum
AU - J. DiCarlo
PY - 2026
DA - 2026/05/13
PB - bioRxiv
DO - 10.64898/2026.05.13.724755
UR - https://www.biorxiv.org/content/10.64898/2026.05.13.724755v2
AB - Interacting with the world relies on knowing its three-dimensional structure, yet vision receives a stream of two-dimensional images. Recovering 3D geometry from dynamic input is therefore foundational to vision, but lacks a computational account. Here we identify one, using artificial neural networks to predict human behavior and neuronal responses from the two major visual pathways in macaques. Using videos of camouflaged objects whose shape is revealed through motion, we found both pathways carry information for motion-defined surfaces, while neurons in parietal cortex most closely track what humans perceive. Only predictive coding video models--trained to fill in missing content in natural videos--reproduced both neuronal responses and human behavior. Predictive coding thus emerges as a principle that yields a brain-like geometric world model of the physical world.
ER -