Cross-population amplitude coupling in high-dimensional oscillatory neural time series

Area of Science:

• Neuroscience
• Computational Neuroscience
• Signal Processing

Background:

• Neural oscillations are crucial for inter-regional brain communication.
• Analyzing coordinated oscillatory activity in high-dimensional neural recordings is challenging.

Purpose of the Study:

• To quantify time-varying covariation of oscillatory amplitudes between two brain regions during a memory task.
• To develop a robust statistical method for identifying dominant lead-lag effects in neural data.

Main Methods:

• Extended Canonical Correlation Analysis (CCA) to multiple time series by cross-correlating latent time series.
• Developed rigorous statistical procedures to manage high dimensionality and identify dominant lead-lag effects.
• Applied the method to local field potentials from prefrontal cortex and visual area V4.

Main Results:

• The novel method successfully identified ground truth structure in simulations.
• Analysis of prefrontal cortex and V4 local field potentials yielded highly plausible results.
• The methodology demonstrated effectiveness in quantifying coordinated oscillatory activity.

Conclusions:

• The developed statistical methodology provides a powerful tool for analyzing brain region interactions via neural oscillations.
• This approach can be applied to other high-dimensional, slowly varying time series data.
• Offers new insights into neural synchrony during cognitive tasks.