Modeling Connectivity in Terms of Network Activity

TL;DR

This paper introduces a novel network growth model based on node activity, which better replicates real cortical networks compared to traditional models, using advanced analysis methods.

Contribution

A new activity-based network growth model is proposed, extending the Barabasi-Albert model and demonstrating superior fit to real cortical networks.

Findings

The model aligns closely with real cortical network data.

It outperforms other models like Watts-Strogatz in compatibility.

Advanced analysis confirms the model's effectiveness.

Abstract

A new complex network model is proposed which is founded on growth with new connections being established proportionally to the current dynamical activity of each node, which can be understood as a generalization of the Barabasi-Albert static model. By using several topological measurements, as well as optimal multivariate methods (canonical analysis and maximum likelihood decision), we show that this new model provides, among several other theoretical types of networks including Watts-Strogatz small-world networks, the greatest compatibility with three real-world cortical networks.