Evaluation of an SSVEP based Brain-Computer Interface on the command and application levels

A brain-computer interface (BCI) provides the possibility to translate brain neural activity patterns into control commands for computers without user´s movement. The brain activity is most commonly measured non-invasively via standard electroencephalographic (EEG) electrodes placed on the surface of the scalp. We propose the evaluation of the Bremen-BCI system based on steady-state visual evoked potentials (SSVEPs), which was evaluated with 37 BCI-naive subjects, including eight handicapped persons, on the international rehabilitation fair RehaCare2008. In spite of the noisy environment during the fair, the spelling tasks were successfully completed. We propose two evaluation methods, one based on the main task to achieve and the second, on the commands that are needed to achieve the task. In the command level, the mean accuracy of the command detection is 92.84%, with an average information transfer rate of 22.6 bpm (bits per minute). In the speller level, the average information transfer rate is 17.4 bpm (equivalent to 3.5 letters per minute with 30 possible letters). These results highlight the differences between two evaluation methods. Differences can emerge between the raw BCI and its connection to an application.