Skip to content

Latest commit

 

History

History
33 lines (32 loc) · 1.38 KB

article-notes.md

File metadata and controls

33 lines (32 loc) · 1.38 KB

royer_eeg_2010

  • virtual helicopter control
  • successful BCI - similar to keyboard control
  • subject specific signals
  • subjects train in one dimension, then two, then three
  • 8 to 11 sessions of 45 minutes to master left and right
  • total session length was 2 hours: hair washing & stuff
  • controlled for EOG
  • 64 channel EEG cap
  • control at approx 50Hz (20-50ms via UDP)
  • the "3D" movement was really fixed in 1 dimension ("0.5 Blender units per second")
  • their "cone of guidance" sounds like a obscure way to note that 1/2 the time subjects did not have good control;
  • seems like "cone of guidance" creates an illusion of control because it was hidden from subjects, seems to have helped to reduce frustration
  • idea for mouse: slow down over buttons, similar to increasing their size or a "gravity well"
  • sessions were 7-13 five-minute runs

moyes_bci_2012

  • pong with EEG
  • one dimension of control
  • tried alpha and mu for control
  • seems they used general level of range (8-13Hz), not tightly focused on a specific frequency inside range
  • tried using SVM to detect p300; not successful
  • firmware and software on webpage
  • 75 USD budget
  • article states "There exists a Modular EEG serial packet data format that is typically used to transmit EEG data over serial"
  • references medical guidelines: ftp://ansuk.org/pub/clinical_governance/dig_eeg.pdf