Can we exploit event-related potentials for retraining language after stroke?

Connections are not prepared for this record yet, or its source metadata has changed. The original record remains available while background snapshots are rebuilt.