Exploring connectivity of resting-state EEG between BCI-literate and illiterate groups

Abstract

Although Motor Imagery-based Brain-Computer Interface (MI-BCI) holds significant potential, its practical adoption faces challenges due to the phenomenon known as BCI-illiteracy. BCI researchers have attempted to predict a BCI-illiteracy to mitigate this issue. As the significance of connectivity in neuroscience has grown, BCI researchers have applied connectivity to predict BCI-illiteracy with resting-state data. However, the use of connectivity metrics and interpretation of the results can be challenging due to several reasons. Firstly, there are various connectivity metrics, each with its own advantages and disadvantages based on their underlying hypotheses and perspectives. These pros and cons are shaped by several factors, increasing the complexity of their application and interpretation. Secondly, it is not evident which factor, such as frequency range or dataset, may influence the estimation of connectivity and which metric is suitable to predict BCI-illiteracy. To help address these issues, this study conducted an empirical analysis. Three large public datasets were analyzed using three functional connectivity (FC) and three effective connectivity (EC) metrics. Additionally, the structural difference in the resting-state network between BCI-literate and illiterate groups was examined. The analysis revealed that the appropriate frequency range to measure connectivity varies depending on the metric used. The alpha range was found to be suitable for FC while the alpha, alpha + theta, and beta ranges were found to be appropriate for EC. Furthermore, the results of connectivity estimation varied depending on the dataset and metric used. Although it was observed that BCI-literacy had stronger connections between nodes, no other significant structural differences were found between the two groups. However, the resting-state network of BCI-literacy displayed higher network efficiency compared to BCI-illiteracy, regardless of the used metrics. Therefore, it seems reasonable to use resting-state connectivity to predict BCI-illiteracy. In conclusion, each metric has a proper hypothesis and perspective for measuring connectivity under certain conditions.