BCI Competitions

The BCI Competition website

BCI Competition IV:

• classification of continuous EEG without trial structure (data sets 1).
• classification of EEG signals affected by eye movement artifacts (data sets 2).
• classification of the direction of wrist movements from MEG (data sets 3).
• discrimination requiring fine grained spatial resolution in ECoG (data sets 4).

Data sets:

 data sets: 1    , 2a    , 2b    , 3    , 4   

number of classes: 2    , 4    , 2    , 4    , 5

signal type: EEG, EEG/EOG, EEG/EOG, MEG, ECoG

sampling rate: 1K,  /250, /250, 400, 1K

frequency:0.05-200Hz, 0.5-100Hz/ , 0.5-100Hz/  , 0.5-100Hz, 0.15-200Hz

number of channels: 64, 22/3, 3/3, 10, 48

number of subjects: 7, 9, 9, 2, 3

 Data set 3: ‹hand movement direction in MEG›

The data set contains directionally modulated low-frequency MEG activity that was recorded while subjects performed wrist movements in four different directions.

[10 MEG channels (filtered to 0.5-100Hz), 400Hz sampling rate, 4 classes, 2 subjects]

Detailed description:

The trials were cut to contain data from 0.4s before to 0.6s after movement onset and the signals were band pass filtered (0.5 to 100Hz) and resampled at 400Hz, whereas in Waldert et al. (JNeurosci 28(4), 2008) we showed that especially the low-frequency activity (<8Hz) contains information about movement direction. The data are composed of signals from ten MEG channels which were located above the motor areas.