Phase space analysis of networks based on biologically realistic parameters

Area of Science:

• Computational neuroscience
• Systems neuroscience
• Neural network modeling

Background:

• Cortical network dynamics are complex, influenced by connectivity and delays.
• Previous models often simplify spatial embedding and connectivity.
• Understanding these dynamics is crucial for brain function.

Purpose of the Study:

• To investigate cortical network dynamics in a spatially embedded model.
• To explore the impact of long-range connections and conduction delays.
• To identify factors governing activity states in neural networks.

Main Methods:

• Utilized a spatially embedded network model with conductance-based I&F neurons.
• Incorporated distance-dependent conduction delays and sparse connectivity.
• Modeled reduced neuron density with external Poissonian spike trains.

Main Results:

• Observed significant changes in the dynamical phase space compared to prior studies.
• Identified two distinct types of mixed states with coexisting phases.
• Found that the transition between high and low activity states is mainly driven by synaptic strength balance.

Conclusions:

• The balance of excitatory and inhibitory synaptic strength is a critical determinant of cortical network states.
• Standard regularity measures like the coefficient of variation may be insufficient for complex network dynamics.
• Spatially embedded models offer new insights into neural network behavior.