Brain dynamics and spatiotemporal trajectories during threat processing.

Area of Science:

• Neuroscience
• Computational Neuroscience
• Systems Neuroscience

Background:

• Functional MRI (fMRI) research traditionally analyzes task processing statically.
• Understanding dynamic threat processing is limited by conventional experimental designs.

Purpose of the Study:

• To apply Switching Linear Dynamical Systems (SLDS) to uncover the temporal dynamics of threat processing.
• To characterize endogenous and exogenous factors influencing brain dynamics during threat exposure.

Main Methods:

• Utilized SLDS to analyze fMRI time series data from 85 regions of interest during a continuous threat-of-shock paradigm.
• Developed a 'region importance' measure to quantify individual brain region contributions to system dynamics.

Main Results:

• SLDS model successfully learned experimental regularities, reflecting threat proximity and approach/retreat dynamics.
• Identified dynamic multivariate patterns in brain activity shaped by intrinsic and extrinsic factors.
• Demonstrated generalizability of the SLDS approach across different threat-processing tasks.

Conclusions:

• Viewing threat processing through dynamical systems offers new insights beyond static fMRI analyses.
• The SLDS approach provides a powerful framework for understanding complex, time-varying brain processes.
• This method can reveal previously unveiled properties of dynamic threat processing in the brain.