Home » Featured

Why Twittering with your Brain is Old Hat

6 May 2009 4,925 views No Comment

If you haven’t heard about this already than you may be living under a rock.

Adam Wilson, a graduate student at the University of Winsconsin-Madison, with support from the Wadsworth Center in Albany, NY, has successfully demonstrated that people can post twitter messages simply by thinking about them.  At least, that’s what the popular news media has been reporting.  Over.  And over.  In fact, this may be the most publicized account of a BCI application ever.  All thanks to the twitter buzz that has been flying around recently.  View the application in action here.

But in fact, the underlying technology behind this accomplishment has been around since at least 1988, when Farwell and Donchin demonstrated that the P300 event related potential can be used to select characters from a grid and eventually spell words.  Here’s how it works:

A subject views a grid of characters and makes a decision about what character he would like to select.  Entire rows or columns of the grid flash in a random order.  Each time the desired character is highlighted a P300 event related potential is elicited, which is a noticeable increase in EEG amplitude occurring about 300 milliseconds after the presentation of interesting or notable stimuli, often called the “oddball” response.  Through trial averaging, an algorithm can determine which character in the grid the subject is focusing on.  String together a bunch of these character selections and voila, you have a sentence.  Integrate this technology with twitter, and you have a tweet.

While this latest account may be nothing more than a repackaged old idea, at least it is presenting that idea in a way that is meaningful to the general population.  Greater public interest in brain-computer interface technology could lead to increased funding, which could lead to new discoveries.  I predict that the Facebook BCI is soon to follow.