A Data-Centric Analysis of the Impact of Training Data Quality vs. Quantity on P300 Brain-Computer Interface Performance (Student Abstract)

The current standard for training brain-computer interface (BCI) models is user-specific. There is a high interest in developing generic models that are trained on data from other users to minimize BCI calibration time; however, this is limited by noisy, non-stationary brain signals and high inter-user variabilities. We investigate the trade-off between training data quality and quantity on P300 BCI performance in individuals with amyotrophic lateral sclerosis (ALS) with representative traditional and deep learning models. Results show that data quality and domain alignment are more critical than dataset size: user-specific models trained on significantly less data outperformed generic models; generic models trained on ALS data outperformed models trained on ALS data; dimensionality reduction with block averaging was detrimental to EEGNet; and ISI differences between ALS and non-ALS data had minimal effect. Our findings highlight the importance of individualized model tuning for reliable P300 BCIs.