A new model for dynamic mapping of effective connectivity in task fMRI

Area of Science:

• Neuroimaging
• Cognitive Neuroscience
• Computational Neuroscience

Background:

• Dynamic functional connectivity analysis is crucial for understanding brain network reorganization.
• Previous studies primarily focused on resting-state fMRI, limiting insights during task performance.
• Investigating dynamic effective connectivity during cognitive tasks remains an underexplored area.

Purpose of the Study:

• To adapt and apply dynamic functional connectivity methods to task-based fMRI data.
• To introduce and validate the Condition Activated Specific Trajectory (CAST) model for analyzing dynamic brain connectivity.
• To assess the efficacy of the CAST model in capturing task-related brain network dynamics and individual cognitive variations.

Main Methods:

• Combined psychophysiological interactions (PPI) with sliding-window analysis on N-back task fMRI data.
• Utilized the Human Connectome Project dataset for robust analysis.
• Developed the CAST model to represent spatiotemporal synchronous changes in activated connections over time windows.

Main Results:

• The CAST model demonstrated superior intra-group consistency in individual spatial patterns of psychophysiological interactions connectivity.
• CAST effectively represented temporal variability and hierarchy in individual task performance.
• The model showed strong correlations with cognitive traits, highlighting its representational ability.

Conclusions:

• The CAST model provides a novel and effective approach for analyzing dynamic effective connectivity during cognitive tasks.
• This dynamic perspective reveals the intrinsic nature of coherent brain activities during task engagement.
• The findings underscore the utility of CAST in understanding individual differences in cognitive function through brain network dynamics.