Topological Sensitivity in Connectome-Constrained Neural Networks

TL;DR

This study critically reevaluates claims that biological connectome topology enhances neural network learning, demonstrating that apparent advantages often result from confounds like initialization and null-model biases.

Contribution

It provides a controlled analysis showing that topology benefits diminish under stricter controls, challenging previous assertions of biological graph superiority.

Findings

Connectome topology advantages often stem from confounds.

Stricter controls eliminate early loss and activity advantages.

Degree-preserving null models remove apparent topology benefits.

Abstract

Connectome-constrained neural networks are often evaluated against sparse random controls and then interpreted as evidence that biological graph topology improves learning efficiency. We revisit that claim in a controlled flyvis-based study using a Drosophila connectome, a naive self-loop-matched random graph, and a degree-preserving rewired null. Under weak controls, in which both models were recovered from a connectome-trained checkpoint and the null matched only global graph counts, the connectome appeared substantially better in early loss, mean activity, and runtime. That picture changed under stricter controls. Training both graphs from a shared random initialization removed the early loss advantage, and replacing the naive null by a degree-preserving null removed the apparent activity advantage. A five-sample degree-preserving ensemble and a pre-training activity-scale diagnostic further strengthened this revised interpretation. We also report a descriptive mechanism analysis of the earlier weak-control comparison, but we treat it as behavioral characterization rather than proof of causal superiority. We show that previously reported topology advantages in connectome-constrained neural networks can arise from initialization and null-model confounds, and largely disappear under fair from-scratch initialization and degree-preserving controls.