Research on steady-state visual evoked brain–computer interface based on moving stimuli

In the past decades, brain–computer interface(BCI) based on steady-state visual evoked potentials(SSVEP) has been widely investigated because of its strong adaptability to different subjects and high information transmission rate(ITR). In most SSVEP-based BCI studies, EEG is used to select static targets such as selecting the desired letters from the static alphabet in BCI speller based on SSVEP. However, we think that moving targets can help to improve the attention of subjects while in the SSVEP-based BCI and contribute to the selection effect using SSVEP. In this paper, the effect of 4 stimuli with different speeds on the system performance was investigated, and the phase interval of the movement between different stimuli was also taken into account. The results showed that subjects’ visual adaptation and attention to stimuli differed significantly for stimuli with different speeds and phase intervals, with the system performing optimally when the speed was 200 pixels/s and the phase interval was 0.5π. Then, a 3 × 3 moving paradigm was designed, the data of stationary and moving paradigms were collected respectively for analysis. The canonical correlation analysis(CCA) method was used for target recognition to evaluate the performance of both paradigms in terms of recognition accuracy and ITR. The results show that the moving paradigm has the equivalent system performance as the stationary paradigm, providing an alternative for SSVEP-based BCI.