Making totally paralysed patients "move" by translating their brainwaves into cursor action on a monitor

I.H. Iversen1, N. Birbaumer2, N. Ghanayim2, T. Hinterberger2, J. Kaiser2, B. Kotchoubey2, A. Kuebler2, N. Neumann2 and J. Perelmouter2

1Department of Psychology, University of North Florida, Jacksonville, FL, U.S.A.
2Institute of Medical Psychology and Behavioral Neurobiology, University of Tübingen, Germany

The purpose of the research is to enable totally paralyzed patients to communicate by means other than muscular activity. We used a computerized on-line task that translates signals from the encephalogram (EEG) into visual feedback [1,2]. Specifically, we use slow cortical potentials (SCP) from the EEG of patients with amyotrophic lateral sclerosis (ALS). The task is divided into individual 6-s trials. The first 2 s serve as a baseline period; during the next 4-s response period, the patient can make a response by changing the SCP compared to the baseline period. The patient faces a monitor that shows a ball-shaped cursor and a target. The vertical ball movement is determined by the EEG signal and gives visual feedback. SCP negativity (relative to baseline) moves the ball upwards and positivity moves it downwards. Thus, the patient has to self-regulate the polarity of the SCP to make the ball hit the target. We used operant conditioning methods to train the patients to control their EEG. Training progressed from moving the ball to a target to spelling of words. In spelling, the characters to select from are split into two letter banks, which are presented successively at the bottom of the monitor as the target. When a target is hit, it splits the presented characters into two halves, etc. Eventually the patient can choose between two characters. Training of spelling progressed from a reduced alphabet and predefined target words (copy spelling) to larger alphabet sets and free choice of what to write. With the full alphabet, the average number of trials to select a single character is 7. Thus, spelling is slow, usually two characters per minute. Two patients have learned to write using this method, and both currently write letters to friends and family. The same basic task has also been used for assessment of cognitive skills. The monitor presents two choices, and the patient is instructed to move the ball to one of the choices; for example, the screen presents an X and a Y with the instruction to move the cursor to the X (see Figure 1). The presentation will illustrate how brainwaves are translated into cursor movement, how the patients are trained, and presents results from the cognitive assessment trials.

Figure 1.

References

  1. Birbaumer, N.; Ghanayim, N.; Hinterberger, T.; Iversen, I.; Kotchoubey, B.; Kuebler, A.; Perelmouter, J.; Taub, E.; Flor, H. (1999). A spelling device for the paralyzed. Nature, 398, 297-298.
  2. Birbaumer, N.; Kuebler, A.; Ghanayim, N.; Hinterberger, T.; Perelmouter, J.; Kaiser, J.; Iversen, I.; Kotchoubey, B.; Neumann, N.; Flor, H. (2000). The thought-translation device (TTD) for completely paralyzed patients. IEEE Transactions on Rehabilitation and Engineering, in press.

Paper presented at Measuring Behavior 2000, 3rd International Conference on Methods and Techniques in Behavioral Research, 15-18 August 2000, Nijmegen, The Netherlands

© 2000 Noldus Information Technology b.v.