Research on multimodal brain-computer interaction target identification technology based on eye movement and SSVEP

SSVEP (Steady-State Visual Evoked Potential) is characterized by its ease of implementation and high temporal resolution, and it can flexibly adjust experimental parameters based on different user needs. As a result, it has been widely applied in Brain-Computer Interface (BCI) system research in recent years. However, EEG signals are susceptible to interference from sensory input, and the simultaneous flashing of multiple visual stimulus targets can affect the quality of EEG signal acquisition, reducing accuracy. To improve the accuracy and speed of SSVEP, this study introduced eye-tracking technology and combined it with SSVEP, creating a multimodal BCI system. The system identifies targets by analyzing the participant's eye movement trajectory in conjunction with the SSVEP signals. The experiment recruited 6 healthy participants, with an average age of 24.2 years, to validate the feasibility of the multimodal BCI system. With the integration of eye-tracking technology, the multimodal BCI system reduced EEG data collection time from 2 seconds to 1.2 seconds under the designed experimental paradigm, achieving a data accuracy of 95.83%. This multimodal brain-computer interaction system, which combines eye movement, can be widely applied in various fields and scenarios, such as facilitating communication between paralyzed patients and the outside world in medical environments or enabling autonomous driving in complex traffic conditions.