Brain User Interface

My long-time buddy Paul points me to a recent announcement that Honda is demoing a way to control a robot through brainwaves.  At first you might think that's incredible, and really taking speech recognition to the next level.  I'm skeptical, so Paul replied with this:

Someone did an experiment, I think back in the K. Eric Drexler days of nascent nanotechnology, that showed that the brain had a remarkable ability to adapt and efficiently control a significantly different "body," in this case a virtual spider. The relevance to me is that the "ecotone" of the interaction of brain waves and external motor responses is quite fecund. Sure, it can be a mess, but that's what Spam filters et al are for.

For first applications, think of handicapped people, to me that's already more than enough to justify the whole business, at least until pluripotent stem cells are cultivated in sufficient numbers in 3D to allow you to give blood and get a whole bunch of cells back to help regrow your leg or fix your spinal column.

Yes, I agree that in situations like this it would be very nice to be connected directly via brain waves, but for the rest of us I bet we'll still find it's easier to use a keyboard.  The reason is that people can't talk and think at the same time.  Literally.  Take a look at some of the experiments done in this classic paper by Ben Schneiderman.

Your brain needs to focus, to concentrate on a single task, in order for you to effectively manipulate things. The limitations of the world make it easier for you to do that with your hands or vocal cords than to go directly from your brain.