Temporal-specific complexity of spiking patterns in spontaneous activity induced by a dual complex network structure

Area of Science:

• Neuroscience
• Computational Neuroscience
• Network Science

Background:

• Neural activity fluctuations are crucial for brain information processing.
• Spontaneous neural activity, including excitatory postsynaptic potentials (EPSPs), contributes to these fluctuations.
• Synaptic network structure (random vs. small-world) impacts neural network characteristics.

Purpose of the Study:

• To investigate the relationship between temporal complexity of spontaneous neural activity and the structural duality of synaptic connections.
• To explore how log-normal distributed synaptic weights and connectivity duality affect neural network dynamics.

Main Methods:

• Utilized a leaky integrate-and-fire spiking neural network model.
• Incorporated log-normal distribution for EPSP synaptic weights.
• Applied multiscale entropy analysis to temporal spiking activity.
• Performed surrogate data analysis to assess determinism in neural dynamics.

Main Results:

• Simulations revealed specific spiking patterns during irregular spatio-temporal activity when strong synaptic connections formed small-world networks.
• Enhanced complexity was observed at larger temporal scales (slower frequencies).
• Surrogate data analysis confirmed that slow temporal dynamics represent a deterministic process within the spiking neural networks.

Conclusions:

• The structural properties of neural networks, particularly small-world characteristics, significantly influence the temporal complexity of spontaneous activity.
• Slow temporal dynamics in neural networks are deterministic and linked to specific network structures.
• This modeling approach offers insights into the complex spatio-temporal neural activity observed in the brain.