Real Time Signal Decoding in Closed Loop Brain Computer Interface for Cognitive Modulation

Abstract

This research presents a novel closed-loop Brain-Computer Interface (BCI) system designed to enhance cognitive performance through targeted neurofeedback. The study addresses the critical challenge of decoding and modulating higher-order cognitive states such as attention, memory, and decision-making, which are often hindered by inter-subject variability and limited datasets. By integrating EEG-based signal acquisition, advanced preprocessing, feature extraction using spatial and temporal analysis, and deep learning models such as CNNs, LSTMs, and Transformers, the system achieves robust and real-time classification of cognitive states. Neurofeedback mechanisms are adapted in real-time to align with user-specific neural profiles, promoting progressive cognitive improvement. Experiments involving participants aged 18 to 50 years demonstrated a classification accuracy exceeding 92% with significant task performance gains of 18% in attention and 22% in memory retention. The findings reveal the system's efficacy in decoding complex neural patterns while maintaining adaptability across diverse populations. This work contributes to the body of knowledge by providing a scalable framework for practical cognitive enhancement applications, bridging gaps between neuroscience, machine learning, and signal processing. Future research may extend the system's capabilities to multi-modal data integration and investigate long-term neuroplasticity effects, paving the way for broader applications in education, healthcare, and human-machine interaction.