A hybrid brain-computer interface system based on two different paradigms

Xiang Li,GAO Xiao-Rong,GAO Shang-Kai

DOI: https://doi.org/10.3969/j.issn.0258-8021.2012.03.002

2012-01-01

Abstract:The hybrid brain-computer interface is a new research direction in brain-computer interface which provides a possibility to further enhance the performance of brain-computer interface system. Many of existing hybrid brain-computer interfac uses paradigms with strong visual stimulation, which is likely lead to subjects' fatigue. In this study, a hybrid brain-computer interface system for character input was proposed. The system combined motor imagery with motion-onset visual evoked potential paradigms sequentially. In this system, motion-onset visual evoked potential was used for character input, while motor imagery was used to switch the interface and allow to input the next character. Both these two paradigms do not depend on strong visual stimulations. In order to verify the feasibility of the system, a total of 5 cases of experiments have been completed. In the experiments, subjects first performed training of the two paradigms, then accepted tests of switching the system interface and input characters. Results of the experiments show that subjects can operate the system well after certain training, with a shortest average time of 3.9 s in single-step motor imagery operation, and accuracy up to 93.3% in character input. The system provides less fatigue, more task types and more flexible control than the single sensory modality based brain-computer interfaces.