Simple model for the prediction of seizure durations

Area of Science:

• Computational Neuroscience
• Complex Systems

Background:

• Neuronal systems exhibit complex dynamics, including seizure-like events.
• Understanding the mechanisms underlying seizure generation is crucial for neurological research.

Purpose of the Study:

• To develop a computational model simulating seizures in spiking excitatory neurons.
• To investigate the role of a firing threshold weakening mechanism in seizure dynamics.
• To explore the concept of a seizure-prone state and its characteristics.

Main Methods:

• A simple computational model of spiking excitatory neurons was employed.
• A novel mechanism weakening firing thresholds was introduced to incorporate memory effects.
• Simulations were conducted to observe system dynamics and phase transitions.
• Statistical analysis of avalanche sizes and seizure durations was performed.

Main Results:

• A 'mode-switching' phase was identified, characterized by a seizure-prone state.
• The system demonstrated transitions between a 'healthy' state (small, scale-free avalanches) and a 'seizure' state (large, periodic avalanches).
• Model predictions for seizure duration statistics aligned with experimental and theoretical findings in neuronal systems.

Conclusions:

• The introduced memory mechanism is key to observing seizure-like dynamics.
• The findings suggest a critical point controlling neuronal avalanches, distinct from previously considered types.
• This model provides a novel perspective on the relationship between neuronal avalanches, seizures, and potentially other complex phenomena like fracture.