Co-activation Probability Estimation (CoPE): An approach for modeling functional co-activation architecture based on...

Area of Science:

• Neuroimaging
• Cognitive Neuroscience
• Network Neuroscience

Background:

• Functional neuroimaging studies brain activation during cognitive tasks.
• Coordinate-based meta-analysis identifies consistently activated brain regions across experiments.
• Within-experiment co-activation patterns, crucial for understanding functional brain relationships, remain understudied, especially at the voxel level.

Purpose of the Study:

• To propose and validate a novel method for modeling voxel-wise co-activation patterns within experiments.
• To deduce and characterize the co-activation network structure.
• To differentiate between local and long-range co-activations.

Main Methods:

• Development of the Co-activation Probability Estimation (CoPE) method to model within-experiment activations.
• Application of permutation testing for statistical significance assessment.
• Automatic classification of co-activations into local (convergent) and long-range (inter-regional) based on distance.

Main Results:

• The CoPE method successfully identified both local convergence and significant long-range co-activations in simulated and real working memory data.
• CoPE revealed a distinct 'core' co-activation network within the working memory dataset.
• The method effectively models voxel-wise co-activation patterns, providing detailed network configurations.

Conclusions:

• The CoPE method provides a robust approach for analyzing within-experiment co-activation patterns.
• It enables the discovery of detailed local and long-range co-activation networks.
• CoPE is a valuable data-driven tool for future research into inter-regional brain communication and cognitive function.