Robust Resting-State Dynamics in a Large-Scale Spiking Neural Network Model of Area CA3 in the Mouse Hippocampus

Area of Science:

• Neuroscience
• Computational Neuroscience
• Systems Neuroscience

Background:

• The CA3 region of the hippocampus is crucial for episodic memory via pattern completion.
• Resting-state activity in CA3 involves complex interactions between glutamatergic and GABAergic neurons.
• The precise network mechanisms governing these spontaneous firing patterns remain largely unknown.

Purpose of the Study:

• To develop a data-driven, large-scale spiking neural network (SNN) model of the mouse CA3 hippocampal circuit.
• To investigate the network mechanisms underlying resting-state activity and computational functions of CA3.
• To explore the impact of neuron-type specific circuitry on network stability and function.

Main Methods:

• Utilized the Hippocampome.org knowledge base to construct a detailed SNN model.
• Employed the CARLsim4 multi-GPU simulation environment with Izhikevich and Tsodyks-Markram formalisms.
• Analyzed population activity following transient activation and simulated external inputs.

Main Results:

• The SNN model exhibited stable, biologically plausible oscillations with robust firing frequencies.
• Neuron and connection-type specificity were essential for network stability and robustness.
• External inputs mimicking dentate mossy fibers induced gamma-band rhythms.

Conclusions:

• The study highlights the critical role of specific neuronal and connection types in CA3 circuit function.
• The developed SNN framework provides a valuable tool for investigating hippocampal hypotheses and circuit mechanisms.
• The approach is scalable to the entire hippocampal formation, aiding in understanding its cognitive roles.