Block coherence: a method for measuring the interdependence between two blocks of neurobiological time series

Area of Science:

• Neuroscience
• Signal Processing
• Computational Biology

Background:

• Multisensor recordings are increasingly used to study brain functional connectivity.
• Current methods assess interdependence between brain regions by averaging pairwise coherence, which can be suboptimal.
• There is a need for methods that analyze functional connectivity between entire blocks of time series from different brain regions.

Purpose of the Study:

• To generalize the concept of coherence for analyzing interdependence between two blocks of non-overlapping time series.
• To introduce and mathematically validate a new measure called block coherence.
• To apply block coherence to analyze functional connectivity in local field potential (LFP) recordings during a visuomotor task.

Main Methods:

• Developed a mathematical framework for block coherence, generalizing ordinary coherence.
• Applied block coherence to analyze LFP data from a monkey performing a visuomotor task.
• Investigated the relationship between block coherence in specific frequency bands and behavioral performance.

Main Results:

• Block coherence was mathematically shown to possess properties analogous to ordinary coherence.
• An increase in block coherence between V4 and prefrontal cortex channels in the beta band was observed.
• This increase in beta-band block coherence was associated with a decrease in behavioral response time.

Conclusions:

• Block coherence provides a robust method for quantifying functional connectivity between multiple time series from distinct brain regions.
• The findings suggest a role for synchronized activity between V4 and prefrontal cortex in the beta band during visuomotor processing.
• Block coherence analysis offers a valuable tool for understanding complex neural dynamics and their relation to behavior.