The parcellation of cortical areas using replicator dynamics in fMRI

Area of Science:

• Neuroscience
• Computational Neuroscience
• Brain Imaging Analysis

Background:

• Functional magnetic resonance imaging (fMRI) time series are crucial for understanding brain function.
• Identifying distinct subregions within cortical areas is essential for detailed brain mapping.
• Traditional correlation methods may not fully capture complex relationships in fMRI data.

Purpose of the Study:

• To introduce replicator dynamics as an exploratory tool for analyzing fMRI data.
• To propose canonical correlation as a similarity measure for fMRI time series.
• To detect and delineate subregions within cortical areas, specifically the left lateral frontal cortex (LFC).

Main Methods:

• Application of replicator dynamics, a game theory concept, to fMRI data.
• Utilizing canonical correlation analysis (CCA) to measure similarity between fMRI time series.
• Testing the method on fMRI data from two experimental paradigms in the LFC.

Main Results:

• The replicator process successfully parcellated the LFC into subregions based on fMRI time series similarity.
• The identified parcellation aligns with recently proposed anterior-posterior subdivisions of the LFC.
• High consistency of results across different measurements within subjects and across a group of subjects.

Conclusions:

• Replicator dynamics, combined with canonical correlation, provide a robust method for exploratory brain region analysis.
• This approach offers a data-driven way to identify functionally distinct subregions within the cortex.
• The findings support the utility of these methods for reproducible brain mapping and subregion discovery.