Neural Functional Alignment Space: Brain-Referenced Representation of Artificial Neural Networks

TL;DR

This paper introduces the Neural Functional Alignment Space (NFAS), a brain-referenced framework that models the evolution of neural network representations across layers using dynamical systems, revealing structured organization and modality-specific clustering.

Contribution

The paper presents NFAS, a novel dynamical systems-based approach for aligning artificial neural network representations with brain activity, moving beyond traditional layer-wise or task-specific methods.

Findings

NFAS reveals modality-specific clustering in neural network representations.

Cross-modal convergence is observed in integrative cortical systems.

Representation dynamics provide a principled basis for understanding neural network organization.

Abstract

We propose the Neural Functional Alignment Space (NFAS), a brain-referenced representational framework for characterizing artificial neural networks on equal functional grounds. NFAS departs from conventional alignment approaches that rely on layer-wise features or task-specific activations by modeling the intrinsic dynamical evolution of stimulus representations across network depth. Specifically, we model layer-wise embeddings as a depth-wise dynamical trajectory and apply Dynamic Mode Decomposition (DMD) to extract the stable mode. This representation is then projected into a biologically anchored coordinate system defined by distributed neural responses. We also introduce the Signal-to-Noise Consistency Index (SNCI) to quantify cross-model consistency at the modality level. Across 45 pretrained models spanning vision, audio, and language, NFAS reveals structured organization within this brain-referenced space, including modality-specific clustering and cross-modal convergence in integrative cortical systems. Our findings suggest that representation dynamics provide a principled basis for