The dynamic properties of a brain network during working memory based on the algorithm of cross-frequency coupling

Area of Science:

• Neuroscience
• Cognitive Science
• Computational Neuroscience

Background:

• Working memory (WM) is vital for cognitive functions, involving short-term information maintenance.
• Existing WM research often overlooks network-level neural signal coordination.
• Cross-frequency coupling (CFC) between neural oscillations, like theta-gamma, is a potential network-level mechanism.

Purpose of the Study:

• To map brain network changes during WM at the level of theta-gamma CFC.
• To analyze the dynamic properties of brain networks during WM using graph theory.
• To investigate the role of CFC in WM formation and information processing.

Main Methods:

• Constructed a 16-channel brain network model based on theta-gamma CFC of local field potentials (LFPs) during WM tasks.
• Analyzed network dynamic properties using graph theory metrics (average degree, shortest path length, global efficiency).
• Compared LFP power and CFC (specifically phase-amplitude coupling, PAC) across different states (rest, free choice, WM) and learning progression.

Main Results:

• LFP power increased in WM states but decreased with learning; theta-gamma CFC and PAC increased with learning days and were higher in WM states.
• Significant changes in average degree, shortest path length, and global efficiency were observed across learning days.
• Correct WM information storage enhanced network connectivity and information transmission efficiency without altering local transmission or small-world attributes.

Conclusions:

• Theta-gamma CFC within brain networks plays a significant role in the formation and consolidation of working memory.
• Learning and successful WM storage are associated with increased network integration and efficient information processing.
• Network-level analysis of CFC provides insights into the neural basis of working memory dynamics.