Lagged covariance structure models for studying functional connectivity in the brain.

Area of Science:

• Neuroscience
• Cognitive Neuroscience
• Brain Imaging

Background:

• Cognitive processes rely on complex brain networks.
• Understanding functional connectivity between brain regions is crucial for network analysis.

Purpose of the Study:

• To establish a statistical framework for studying effective and functional brain connectivity.
• To utilize the timing information from event-related optical signal (EROS) neuroimaging.
• To model dynamic cognitive processes over time.

Main Methods:

• Utilized event-related optical signal (EROS) neuroimaging.
• Applied lagged cross-correlations to analyze timing information between brain areas.
• Employed dynamic factor analysis to test structural models on lagged covariance matrices.
• Developed structural equation models with latent variables.

Main Results:

• Demonstrated a statistical framework for analyzing brain connectivity using EROS data.
• Successfully modeled neural activity propagation from V1 to V3.
• Interpreted inter-hemispheric interactions based on anatomical connection integrity.
• Showcased the ability to fit complex models to dynamic cognitive processes.

Conclusions:

• The developed framework effectively analyzes brain connectivity using EROS.
• Timing information from EROS enhances the interpretation of functional connections.
• Anatomical connection integrity plays a key role in inter-hemispheric interactions.
• This approach enables modeling of dynamic cognitive processes.