Platonic Representations in the Human Brain: Unsupervised Recovery of Universal Geometry

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Marcos-Manchón, Pablo, Jha, Rishi, Fuentemilla, Lluís
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917514434314240
author Marcos-Manchón, Pablo
Jha, Rishi
Fuentemilla, Lluís
author_facet Marcos-Manchón, Pablo
Jha, Rishi
Fuentemilla, Lluís
contents The Strong Platonic Representation Hypothesis suggests that representational convergence in artificial neural networks can be harnessed constructively: embeddings can be translated across models through a universal latent space without paired data. We ask whether an analogous geometry can be recovered across human brains. Using fMRI data from the Natural Scenes Dataset, we propose a self-supervised encoder that learns subject-specific embeddings from brain data alone by exploiting repeated stimulus presentations. We show that these independently learned spaces can be translated across subjects using unsupervised orthogonal rotations, without paired cross-subject samples or intermediate model representations. Synchronizing pairwise rotations into a single shared latent space further improves cross-subject retrieval, indicating that subject-specific spaces are mutually compatible with a common coordinate system. These results provide evidence for a shared neural geometry in the human visual cortex: subject-specific fMRI representations are approximately isometric across individuals and can be translated through purely geometric transformations.
format Preprint
id arxiv_https___arxiv_org_abs_2605_20496
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Platonic Representations in the Human Brain: Unsupervised Recovery of Universal Geometry
Marcos-Manchón, Pablo
Jha, Rishi
Fuentemilla, Lluís
Neurons and Cognition
Computer Vision and Pattern Recognition
68T07
I.2.6; I.2.10; J.3
The Strong Platonic Representation Hypothesis suggests that representational convergence in artificial neural networks can be harnessed constructively: embeddings can be translated across models through a universal latent space without paired data. We ask whether an analogous geometry can be recovered across human brains. Using fMRI data from the Natural Scenes Dataset, we propose a self-supervised encoder that learns subject-specific embeddings from brain data alone by exploiting repeated stimulus presentations. We show that these independently learned spaces can be translated across subjects using unsupervised orthogonal rotations, without paired cross-subject samples or intermediate model representations. Synchronizing pairwise rotations into a single shared latent space further improves cross-subject retrieval, indicating that subject-specific spaces are mutually compatible with a common coordinate system. These results provide evidence for a shared neural geometry in the human visual cortex: subject-specific fMRI representations are approximately isometric across individuals and can be translated through purely geometric transformations.
title Platonic Representations in the Human Brain: Unsupervised Recovery of Universal Geometry
topic Neurons and Cognition
Computer Vision and Pattern Recognition
68T07
I.2.6; I.2.10; J.3
url https://arxiv.org/abs/2605.20496