Hierarchical Flows of Human Cortical Activity

Area of Science:

• Neuroscience
• Computational Neuroscience
• Brain Imaging

Background:

• Brain activity occurs as complex spatiotemporal patterns.
• Quantifying neural signal propagation on the folded cortex is difficult.
• Existing methods lack millisecond resolution and geometric accuracy.

Purpose of the Study:

• Introduce geodesic cortical flow, a novel surface-based optical-flow framework.
• Estimate millisecond-resolved, surface-tangent propagation fields from MEG data.
• Analyze frequency-specific cortical propagation and its relationship with aging and cognition.

Main Methods:

• Applied geodesic cortical flow to resting-state magnetoencephalography (MEG) data from 608 healthy adults.
• Estimated propagation direction and strength along the cortical surface.
• Analyzed frequency bands (slow: 1-13 Hz, beta: 13-30 Hz) and kinetic energy.

Main Results:

• Spontaneous cortical propagation is anisotropic and follows the unimodal-to-transmodal functional gradient.
• Slow activity propagates from sensory to association cortex (upstream); beta activity propagates oppositely (downstream).
• Aging shifts this balance towards weaker upstream slow and stronger downstream beta propagation.
• Propagation strength (kinetic energy) shows a posterior-to-anterior gradient.
• Higher kinetic energy in frontoparietal cortex correlates with better fluid intelligence.
• Kinetic energy dynamics reveal regional neuronal timescales.

Conclusions:

• Geodesic cortical flow provides a geometry-informed framework for quantifying cortical propagation.
• Reveals frequency-specific, age-dependent changes in brain signal flow.
• Links cortical propagation dynamics to cognitive abilities like fluid intelligence.