One interesting aspect of BCI is the interaction between patient needs and scientific needs. There's always a tightrope to be walked when trying to enable a person to regain function - do you go for the surefire ability to restore apparently better solution, or the one that leads to greater scientific validity, which may give much greater returns, though further in the future? For casual BCI observers, the choice is usually pretty straightforward: do what will benefit the greatest number of people int he long run. But, sitting at a person's bedside and trying to express this importance is futile. No one wants to feel like they're being thrown under the train to save a nameless, faceless population. Martyrdom might work in a few moments of decision, but it rarely lasts long. Add to that the self-coercion that must occur should one decide to change their mind - if I back out now, I'm harming so many people - and the complications become staggering to any bioethicist.
Let's throw another monkey wrench into the whole setup. The best compromise would be to have experimental sessions in which alternative forms of communication are used - implants, EEG, etc. During 'normal business hours' patients are free to use whatever device they find most effective. Sounds great, huh? One word, and all the neuroscientists will cringe: plasticity. A) Is the new device being used long enough to induce long term changes in an optimal manner? That is, are they able to practice enough to really make the new device usable? B) Is the return to their 'comfortable device' washing out the effects of learning, or in some way altering it? This is particularly important as devices receive finer and finer signals. We'd expect less impact on EEG than individual action potentials. C) Is it better to use similar tasks, like spelling on an grid style keyboard, or radically different ones, like spelling versus drawing/tracing?
Anyhow, I'm sure someone can make a thesis out of that little bit, so you're welcome! This was sparked by a new artificial finger for amputees which is controlled by surrounding fingers, which provides 'life-like' articulation. He's a link to the MedGadget blog where more info, including videos can be found. Oh, and it's called the XFinger, because it is obviously 'X-treme'!