Spatiotemporal forward solution of the EEG and MEG using network modeling

Area of Science:

• Neuroscience
• Dynamical Systems Theory
• Computational Neuroscience

Background:

• Human coordination behavior, such as synchronization with auditory stimuli, can be modeled using dynamic systems.
• Electroencephalography (EEG) and magnetoencephalography (MEG) capture brain signal dynamics, revealing ordered patterns accessible by dynamic systems theory.
• Existing models of EEG/MEG dynamics often rely on phenomenological approaches like current dipoles or spatial patterns.

Purpose of the Study:

• To connect coordination behavior, observed EEG/MEG patterns, and underlying neuronal network dynamics.
• To develop a methodological framework for modeling neural field dynamics on a spherical representation of the brain.
• To bridge the gap between phenomenological models and the underlying neural network dynamics.

Main Methods:

• Developed a methodological framework defining spatiotemporal neural ensemble dynamics (neural field) on a 3D sphere.
• Utilized magnetic resonance imaging (MRI) to map neural field dynamics onto the cortical surface.
• Employed Volterra integrals to map finger movement profiles to EEG/MEG patterns.

Main Results:

• Demonstrated the propagation of neural activation within the developed framework.
• Successfully mapped neural field dynamics from a sphere to the folded cortical surface.
• Established a quantitative mapping between finger movement and EEG/MEG patterns.

Conclusions:

• The proposed framework successfully integrates neural network dynamics with observable EEG/MEG patterns and coordination behavior.
• This approach offers a more mechanistic understanding of brain signal generation related to motor control.
• The methodology provides a foundation for more sophisticated modeling of brain dynamics and behavior.