Sound imagery contributes to foot Mi-based BCI even through it does not influence on the sensorimotor rhythms

Recently, brain-computer interfaces (BCIs) attract much attention as novel human interfaces such as exoskeleton, powered-wheelchair, and P300-based speller. One of the EEG feature called event-related (de)synchronization (ERD/S) is used for the BCI, because it can be caused by motor execution (ME), motor imagery (MI), or motor observation (MO). ERD is a relative power decrease during ME/MI/MO, whereas ERS is a relative power increase after the termination of ME/MI/MO. Thus, not only healthy but also paralyzed patients can use BCIs. It is expected to improve quality of life and to be a neurore-habilitation system for paralyzed. Previous studies indicate that visual and auditory feedback corresponding to MI provide better training effect than non-feedback. From these previous studies, we hypothesized that imagining movement and sound at the same time could enhance ERD/S. We investigated this hypothesis by conducting an experiment. Furthermore, we considered that just motor imagery and motor imagery with sound imagery could be classified by support vector machine (SVM). As a result, we found that there are no significant differences between ERD/S during just MI and during MI with sound imagery, however the classification accuracy between just MI and MI with sound imagery is comparable to the classification accuracy between rest and just MI.