Revealing Cross-Frequency Causal Interactions During a Mental Arithmetic Task Through Symbolic Transfer Entropy: A...

Area of Science:

• Neuroscience
• Cognitive Science
• Computational Neuroscience

Background:

• Working memory (WM) relies on communication between prefrontal and posterior brain regions.
• This communication involves cross-frequency coupling between theta (θ) and high-alpha (α2) brain waves.
• Understanding causal interactions between brain waves is crucial for cognitive process research.

Purpose of the Study:

• To propose a novel method for estimating transfer entropy (TE) using a neural-gas algorithm (NG) based symbolization scheme.
• To define delay symbolic transfer entropy (dSTE^NG) for quantifying causal interactions between different frequency brain waves.
• To assess the method's effectiveness in a working memory task.

Main Methods:

• Developed a novel transfer entropy estimation method using neural-gas (NG) algorithm for time series symbolization.
• Encoded bivariate time series into symbolic sequences to define delay symbolic transfer entropy (dSTE^NG).
• Applied the dSTE^NG method to multichannel EEG data from 16 subjects during a mental arithmetic task.

Main Results:

• Detected effective interactions between Frontal-theta (Fθ) and Parieto-occipital-high-alpha (POα2) brain waves.
• The dSTE^NG method showed improved computational efficiency and noise resilience compared to conventional methods.
• Identified a consistent efferent Fθ and afferent POα2 connectivity pattern across task difficulties.

Conclusions:

• The proposed dSTE^NG method effectively quantifies causal interactions between different frequency brain waves.
• The findings reveal specific directional connectivity patterns in working memory.
• dSTE^NG strength increases with task difficulty, offering insights into cognitive load.