Attentional modulation of inter-areal coherence explained by frequency shifts

Area of Science:

• Computational neuroscience
• Systems neuroscience
• Neural oscillations

Background:

• Inter-areal coherence is hypothesized to mediate brain communication.
• Empirical studies link increased coherence to attention and stimulus salience.
• Shifts in gamma oscillation peak frequency in V1 suggest frequency plays a role in communication.

Purpose of the Study:

• To investigate how a sender's oscillation peak frequency influences inter-areal coherence using computational modeling.
• To determine the role of receiver properties (integration vs. resonance) in shaping coherence patterns.
• To develop a more accurate measure of inter-areal interactions.

Main Methods:

• Computational modeling of neural networks.
• Simulating sender-receiver interactions with varying oscillation frequencies.
• Analyzing coherence patterns based on receiver integration or resonance properties.
• Developing and validating the 'Explained Power' metric.

Main Results:

• Sender peak frequency significantly impacts coherence magnitude.
• Receiver properties (integrator vs. resonator) dictate the pattern of coherence changes.
• Resonant receivers produced coherence patterns inconsistent with empirical data.
• Integrator receivers produced coherence patterns aligning with empirical observations of frequency shifts.
• Coherence can be a misleading indicator of inter-areal interactions.

Conclusions:

• Inter-areal coherence patterns are critically dependent on receiver intrinsic properties.
• Coherence alone may not accurately reflect neural communication.
• The 'Explained Power' metric offers a more direct quantification of transmitted signals between brain regions.
• Findings provide a model for understanding changes in coherence and Granger-causality due to frequency shifts.