An alternative approach may be to use Visual Evoked Potential (VEP).
However, what I am writing below is completely 'hypothetical'.
I do not know whether dreaming about 'seeing' the light will
produce the same Visual Evoked Potential or not. Any comments?
Assume it does, then since supposedly in lucid dreaming your
consciousness is working and now suppose you get into an habit
of walking into a room where bright lights are turned on. If
the assumtion that VEP will occur even in the dream is true, then
you will receive a distinct response from the EEG pre-attached
to your scalp. This can then be used as onset. To make things
codable, you will alternatively open and close your eyes. Again,
since there is bright light, the opening of your eye will cause
VEP.
Alternatively, you can imagine a bright sun above you (or behind
you) which you normally won't see. Whenever you want to transmitt
information, you can look up to the sun. (If there is a big
contrast, VEP will occur on the recording.)
This approach seems much more doable. In fact, Dr. Pierre Cilliers
of Universit van Pretoria, who just visited me yesterday, has
successfully created a VEP control system allowing C4 quadroplegic
to control the computer using VEP. He uses four LED's each blinking
at different frequency (therefore, there's a phase shift in VEP).
He will be presenting his finding in IEEE-EBME conference coming up
next month in San Diego. Of course, this has nothing to do with
dreaming...
The problem, of course, lies in whether VEP occurs in dream and
whether the lucid dreamer can "create" scenarios (i.e. to have
the sun, or to have bright lights in a room.) I'll be very
interested to know about these.
Edwin Yeh
Case Western Reserve
Biomedical Engineering
& Applied Neural Control