Related Experiment Videos
Synchronous neural activity in scale-free network models versus random network models.
Geoffrey Grinstein1, Ralph Linsker
1IBM Thomas J. Watson Research Center, 1101 Kitchawan Road & Route 134, PO Box 218, Yorktown Heights, NY 10598, USA.
Summary
Neural network models with scale-free properties generate large synchronous firing peaks, unlike random networks. This suggests network topology influences neural activity patterns.
Area of Science:
- Computational Neuroscience
- Network Science
Background:
- Recent studies report synchronous firing peaks in neocortical tissue, exceeding background activity.
- A small group of neurons dominates these synchronous events.
Purpose of the Study:
- To investigate if a simple model can replicate the observed synchronous neural firing.
- To explore the role of network topology in generating these phenomena.
Main Methods:
- Construction and study of a model neural network.
- Utilized a modified Hopfield-type dynamical rule.
- Compared networks with power-law (scale-free) and random (Erdös-Rényi) node degree distributions.
Main Results:
- Scale-free networks readily generated extremely large synchronous firing peaks.
- These peaks were dominated by a small subset of nodes.
- Random networks did not exhibit this behavior.
Conclusions:
- Network topology, specifically scale-free properties, is crucial for generating large synchronous neural firing peaks.
- This finding highlights the importance of network structure in neural dynamics.