Broad-Band Visually Evoked Potentials: Re(con)volution in Brain-Computer Interfacing

摘要: Brain-Computer Interfaces (BCIs) allow users to control devices and communicate by using brain activity only. BCIs based on broad-band visual stimulation can outperform other paradigms. Visual with pseudo-random bit-sequences evokes specific Broad-Band Visually Evoked Potentials (BBVEPs) that be reliably used in BCI for high-speed communication speller applications. In this study, we report a novel paradigm BBVEP-based utilizes generative framework predict responses sequences. study designed modulated Gold codes mark cells BCI. We defined linear model decomposes full into overlapping single-flash responses. These are sequences, which turn serve as templates classification. The explains average 50% up 66% of the variance both seen unseen an online experiment, 12 participants tested 6 × matrix On average, accuracy 86% was reached trial lengths 3.21 seconds. This corresponds Information Transfer Rate 48 bits per minute (approximately 9 symbols minute). indicates potential stimulation. predicted proven well-suited BCI, thereby enabling