Integration-Segregation Trade-off

Research on how dynamic functional brain networks balance efficient information spreading with segregation under constrained connection costs.

Preprint · Poster

Static and dynamic network resources compared with guided and random-walk spreading in static and temporal networks

Research question

Determine whether time-varying functional connectivity offers an information-processing advantage over static networks built with the same total connection cost.

Data and comparison

The study analyzes dynamic functional connectivity reconstructed from Human Connectome Project fMRI data. Empirical temporal networks are compared with equal-cost static representations and network null models to separate effects of topology from effects of timing.

Analytical approach

Information-spreading measures quantify temporal reachability and transport efficiency, while complementary measures track local organization, temporal clustering, return behavior, and retention. Together they test integration and segregation within the same cost-constrained framework.

Main finding

The analysis indicates that empirical brain-network dynamics combine rapid information spreading with persistent local organization, resolving an apparent trade-off between integration and segregation.

The analysis led to the development of the tNet Analysis toolkit, which packages the temporal-network construction, null models, and measures into a documented and reusable workflow.

Tools and methods

  • Python
  • Human Connectome Project
  • Dynamic functional connectivity
  • Temporal networks
  • Network null models
  • Information diffusion