Functional imaging of Brain-Computer Interface control

Brainwaves recorded with EEG can be used to control equipment like a computer.
This so-called *Brain-computer Interface* (BCI) technique is currently being
developed to aid patients who can not move or communicate, due to motor-neuron
disease, spinal cord lesion or head trauma. If successful, patients can operate
a computer to write or control devices such as a TV or lights, by merely
thinking about making movements. For this purpose, electrodes are positioned on
the head, close to the motor cortex, and recorded signals are converted in
realtime to movements of a cursor on a computer screen. It requires a certain
period of training for subjects to achieve BCI control.

There are some unresolved issues that we intend to address in this project. We
will examine 1) whether with locating the motorcortex with functional MRI, the
best location of the electrodes can be predetermined, 2) how brain activity
changes in the course of learning to achieve BCI control, and whether that
causes a gradual decline of BCI control, and 3) whether other brain systems
than the motorsystem can be used to broaden BCI solutions. These issues are
important for further development of clinical BCI.

Four groups of healthy volunteers will be recruited. All groups will undergo an
fMRI and an EEG scan before and after multiple BCI training sessions, which
will be compared to determine whether brain activity changes. Group 1 will not
perform BCI training, serving as the control group. Group 2 will train to
control a cursor in 1 direction, by imagining movements of the right hand.
Group 3 will do the same but now in 2 directions, by also imagining movements
of the left hand. Group 4 will do the same as group 3, but instead of imagining
left hand movements, the subjects will make mental calculations with electrodes
recording from the prefrontal cortex. BCI training consists of learning to
control the cursor in a special computerprogram. Learning is measured by the
accuracy by which the subject can hit a target by moving this cursor.

The burden for participants is low: there are no invasive procedures and the
scan techniques (fMRI and EEG) are experienced as a light burden. Both are
often used for neuroscientific research in humans. The total duration of
participation is 20 hours on average. The exact duration depends on the groups
that the person is assigned to, and how quickly he/she learns to control the
BCI. Minimum duration is 5 hours (control group, no BCI training), and the
maximum duration is 29 hours (with 16 BCI training sessions of 90 minutes).
There are no risks involved in participation.