A Novel Asynchronous Brain–Computer Interface Using an Uncued SSVEP Paradigm and Early Electroencephalography Component

The brain-computer interface (BCI) research has achieved substantial progress in multitarget classification; however, understanding of the brain’s visual processing mechanisms remains limited. This study systematically investigates the effects of different frequency visual stimuli on steady-state visual evoked potentials (SSVEPs) and introduces a no-visual-cue asynchronous SSVEP-based BCI that incorporates early SSVEP components. By utilizing eye tracking to monitor the subject’s attention, we were able to assess the system’s accuracy. This approach was further applied to a 40-target online spelling task to evaluate the classification performance. The system successfully distinguished between intentional control (IC) and noncontrol (NC) states, achieving high accuracy and reliability in multitarget classification. These findings underscore the potential of this novel data augmentation approach to enhance the BCI performance in practical applications.