xDAWN Spatial Filter Trainer


  • Plugin name : xDAWN Spatial Filter Trainer
  • Version : 1.0
  • Author : Yann Renard
  • Company : INRIA
  • Short description : Computes spatial filter coeffcients in order to get better evoked potential classification (typically used for P300 detection)
  • Documentation template generation date : Jul 24 2014


This box can be used in order to compute a spatial filter in order to enhance the detection of evoked response potentials. In order to compute such filter, this box needs to receive the whole contain of a session on the first hand, and a succession of evoked response potentials on the other hand. It then computes the averaged evoked response potential computes the spatial filter that makes this averaged potential appear in the whole signal. This can be used e.g. for better P300 signal detection.

It is important to consider the fact that this box will have best results for a reasonably big number of input channels, possibly all over the scalp (areas where the evoked response potential can not be seen will be naturally used as references to reduce noise). The spatial filter results in space reduction to only keep significant channels for later detection. Consider using at least 4 times more input channels than the number of output channels you want. For example, reducing 16 electrodes to 3 channels for P300 detection is OK.

For more details about xDAWN, see Rivet et al. 2009 or in case this links disappears, this website.


1. Stimulations

This input receives the experiment stimulations. As soon as the "train" stimulation is received, the spatial filter is computed.

  • Type identifier : Stimulations (0x6f752dd0, 0x082a321e)

2. Session signal

This input should receive the whole signal of the session.

  • Type identifier : Signal (0x5ba36127, 0x195feae1)

3. Evoked potential epochs

This input should receive the multiple evoked response potentials.

  • Type identifier : Signal (0x5ba36127, 0x195feae1)


1. Train-completed Flag

  • Type identifier : Stimulations (0x6f752dd0, 0x082a321e)


1. Train stimulation

This setting contains the stimulation to use to trigger the training process.

  • Type identifier : Stimulation (0x2c132d6e, 0x44ab0d97)
  • Default value : [ OVTK_StimulationId_Train ]

2. Spatial filter configuration

This setting tells the box what configuration file to generate. This configuration file can be used to set the correct values of a Spatial Filter box.

  • Type identifier : Filename (0x330306dd, 0x74a95f98)
  • Default value : [ ]

3. Filter dimension

This setting tells how many dimension should be kept out of the spatial filter.

  • Type identifier : Integer (0x007deef9, 0x2f3e95c6)
  • Default value : [ 4 ]