betaFeedback

WorksAgricultural and Biological Sciences

Regime-dependent reconstruction of structural connectivity: an in silico study of spatially constrained modular neuronal networks

Abstract

Inferring structural connectivity from neuronal activity is a major challenge in neuroscience, yet the conditions under which inferred interactions faithfully reflect anatomical connections remain unclear. Here, we investigate in silico how spatial constraints in a neuronal network and its dynamical regime jointly shape structural recoverability. We construct modular networks with varying average axonal lengths and simulate spontaneous activity using the Izhikevich model with synaptic depression across a broad range of excitatory strengths and background noise levels. Using Generalized Transfer Entropy, we infer effective connectivity and compare it with the underlying structural ground-truth at the levels of individual connections and global topology. We observe that the richness in the repertoire of activity patterns is highest in an intermediate regime between globally synchronized and noise-dominated activity, reflecting a balance between global excitation and stochasticity. Optimal structural reconstruction generally occurs under these same conditions, although this correspondence depends on network architecture and is weaker in highly modular networks. Together, these results show that structural recoverability depends jointly on network organization and dynamical regime. More broadly, our findings highlight the risks of treating inferred effective connectivity as a direct surrogate for structural connectivity.

Paper details

vlnc:W_zab576jq
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
arXiv nonexclusive license
Usage

views · 9

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Agricultural and Biological Sciences

Not recorded for this paper

View statement
Looking for related papers…