Thursday 10 March 2011

Speech: Tan Le - A headset that reads your brainwaves

Tan Le's astonishing new computer interface reads its user's brainwaves, making it possible to control virtual objects, and even physical electronics, with mere thoughts (and a little concentration). She demos the headset, and talks about its far-reaching applications.

Rated OMGWTF (in a good way) by Passive Obersvers.


4 comments:

  1. This is a very interesting video but the technology shown here, in its full potential is very far from reality today.

    There are several issues that arise when dealing with the brain and it all comes down to the complexity of the brain as a system. Brain cells fire in very complex patterns for every response of the person, whether cognitive or empirical. This complexity is even more evident when we see overlap of neural activity in unrelated areas for acts that are of a different kind. This makes it especially hard for a machine such as this to recognize brain activity. A much more precise machine will be needed and even still, the way my brain wire fire un when I think of "lights on" might not be the same as tomorrow, at least not 100%! In sum, at its current level this technology can only assign a very small number of commands to be read from brain activity. If you look at the guy shutting the curtains, what he is doing is essentially pushing the cube back, this shows you that there are only so many acts that you can do and each can be assigned to a different command.

    Secondly we must also consider brain plasticity. The brain is always changing and certain theories go as far to say that the brain is constantly changing because it is effected by every stimulation one experiences both consciously and unosciously, and that is a lot of stimulation! The machine will have to take the changes occurring in the brain in order for it to carry out its function. This demands that we know the pattern of plasticity and that is near impossible to map out. Then again the problem can be tackled by making the machine learn again and again each new brain pattern for every command, but this kind of beats the purpose.

    Now, I cant say that Tan Le has not thought of these problems, I am not the distinguished neuroscientist here, but the fact that she did not mention these problems led my to writting this comment. I probably have not made myself crystal clear and that is why I rarely write comments. I would prefer a conversation setting where things can be explained more effectively.

    ReplyDelete
  2. http://www.youtube.com/watch?v=jOkpn0BN2HE

    Old one, but showed the initial potential when monkeys managed to move prosthetic arms to feed themselves

    ReplyDelete