EEG based zero-phase phase-locking value (PLV) and effects of spatial filtering during actual movement.

Area of Science:

• Neuroscience
• Biomedical Engineering
• Signal Processing

Background:

• Phase-locking value (PLV) is a key feature in sensorimotor rhythm (SMR) based brain-computer interfaces (BCIs).
• Zero-phase PLV is often disregarded due to concerns about volume conduction, a phenomenon where electrical signals spread across the scalp.
• Spatial filtering is commonly used in BCIs to enhance signal amplitude and reduce volume conduction, but its impact on PLV remains unclear.

Purpose of the Study:

• To investigate the potential significance of zero-phase PLV in BCI applications.
• To determine how spatial filtering techniques influence the interpretation of zero-phase PLV.
• To explore the relationship between zero-phase PLV, band power (BP), and volume conduction in EEG data.

Main Methods:

• Analysis of archival EEG data from 32 subjects performing left and right hand movement tasks.
• Comparison of band power (BP) and phase-locking value (PLV) features.
• Evaluation of data with and without pre-processing using a large Laplacian spatial filter.

Main Results:

• Ear-referenced data revealed that zero-phase PLV provided unique, non-volume-conducted information for task prediction, independent of BP.
• Application of a large Laplacian filter significantly reduced the predictive power of zero-phase PLV.
• The large Laplacian appeared to remove useful information from zero-phase PLV, indicating it captures both amplitude and phase effects.

Conclusions:

• Zero-phase PLV may possess functional significance beyond mere volume conduction in BCI.
• Spatial filtering methods can complicate the interpretation of phase-based features like PLV.
• Further research is needed to understand the complex interplay between spatial filtering, phase information, and BCI performance.