In the realm of EEG decoding, enhancing the performance of artificial neural networks (ANNs) carries significant potential. This study introduces a novel motor imagery inspired approach, termed “Weight-Freezing”. The concept of Weight-Freezing revolves around the idea of reducing certain neurons’ influence on the decision-making process for a specific EEG task by freezing specific weights in the fully connected layer during the backpropagation process. This is actualized through the use of a mask matrix and a threshold to determine the proportion of weights to be frozen. Moreover, by setting the penalty factor to zero, Weight-Freezing can realize sparse connections in networks with a fully connected layer as the classifier. Through experiments involving three distinct ANN architectures and three widely recognized EEG datasets — BCI4-2A, BCI4-2B, and High-Gamma — we validated the effectiveness of Weight-Freezing. Our method significantly outperformed state-of-the-art classification accuracy across all examined datasets, achieving 84.9%, 86.5%, and 96.9% accuracy (averaged across all participants) for BCI4-2A, BCI4-2B, and High-Gamma, respectively. Supplementary control experiments offer insights into performance differences pre and post Weight-Freezing implementation and scrutinize the influence of the threshold in the Weight-Freezing process. Our study underscores the superior efficacy of Weight-Freezing compared to traditional fully connected layers for EEG classification tasks. With its proven effectiveness, this innovative approach holds substantial promise for contributing to future strides in EEG decoding research.

Top