I. Volosyak, H. Cecotti, and A. Gräser (Germany)
Brain-Computer Interface (BCI), Electroencephalogram(EEG), Steady State Visual Evoked Potential (SSVEP),Signal Detection
When a visual stimulus with a constant frequency is presented to a person, it is possible to observe a continuous
brain response in the visual cortical area. This response
is called a steady-state visual evoked potential (SSVEP).
The detection of this brain response can be analyzed as a
command for tailoring a Brain-Computer interface (BCI).
Its goal is to provide a new output channel for the brain that
requires voluntary control. The efﬁciency of a BCI depends
on the commands accuracy and their latency: the time between the order from the subject and the moment when
the command is detected. Given one particular SSVEP response classiﬁcation method, the BCI control is variable
among the users. This variance is translated into different performance between subjects. We propose a robust
classiﬁcation method of SSVEP responses, which does not
require any training. A performance comparison on 8 different time segment lengths over 10 subjects is proposed.
The average time segment length for obtaining a SSVEP
response recognition of 95% is determined to be 2.8s.