Part of Advances in Neural Information Processing Systems 18 (NIPS 2005)
Le Song, Evian Gordon, Elly Gysels
Motor imagery attenuates EEG and rhythms over sensorimotor cortices. These amplitude changes are most successfully captured by the method of Common Spatial Patterns (CSP) and widely used in braincomputer interfaces (BCI). BCI methods based on amplitude information, however, have not incoporated the rich phase dynamics in the EEG rhythm. This study reports on a BCI method based on phase synchrony rate (SR). SR, computed from binarized phase locking value, describes the number of discrete synchronization events within a window. Statistical nonparametric tests show that SRs contain significant differences between 2 types of motor imageries. Classifiers trained on SRs consistently demonstrate satisfactory results for all 5 subjects. It is further observed that, for 3 subjects, phase is more discriminative than amplitude in the first 1.5-2.0 s, which suggests that phase has the potential to boost the information transfer rate in BCIs.