[wordup] Monkey Moves Computer Cursor by Thoughts Alone

[Adam Shand]

Via: http://slashdot.org/article.pl?sid=02/01/31/198243&mode=flat
From: http://dailynews.yahoo.com/h/nm/20020130/hl/monkeys.html

Monkey Moves Computer Cursor by Thoughts Alone
Wednesday January 30 5:23 PM ET
By E.J. Mundell

NEW YORK (Reuters Health) - It may seem like science fiction, but
scientists say they have developed a technology that enables a monkey to
move a cursor on a computer screen simply by thinking about it.

The breakthrough could someday help totally paralyzed, ''locked-in''
patients ``operate external devices such as a robot arm, or a computer
to surf the Internet,'' explained researcher Daniella Meeker of the
California Institute of Technology in Pasadena.

Meeker described her team's findings at a recent meeting of the Society
for Neuroscience.

Their work focuses on a section of the primate brain known as the
posterior parietal cortex. Using high-tech brain scans, the researchers
determined that small clumps of cells in this region--as few as 16--were
active in the formation of the desire to carry out specific body
movements.

Armed with this knowledge, Meeker's group implanted sensitive electrodes
in the posterior parietal cortex of a rhesus monkey trained to play a
simple video game. The monkey's brain was first analyzed on high-tech
MRI as it used its hand to touch dots on a touch-sensitive computer
screen.

``After the monkey has done this task several times, we are able to
determine, for the particular neuron, the different patterns of
electrical activity when he is planning reaches in different
directions,'' Meeker said.

Going one step further, her team then trained the monkey to simply think
about a movement, without reaching out and touching the screen. A
computer program, hooked up to the implanted electrodes, interpreted the
monkey's thoughts by tracking flare-ups of brain cell activity. The
computer then moved a cursor on the computer screen in accordance with
the monkey's desires--left or right, up or down, wherever ``the
electrical (brain) pattern tells us the monkey is planning to reach,''
according to Meeker.

``In fact,'' she said, ``we found that he became quite reluctant to move
his arm to the reach command once the cursor was introduced into the
game. Apparently it was easier just to think about reaching.''

Speaking with Reuters Health, Meeker said her group's work differs from
previous research in that it seeks to replicate the brain-motor
connection ``at the level of the first inclination to make a movement.''

She added, ``This supposes that much of the downstream function--which
would be normally implemented by the rest of the brain, the spinal cord
and muscles--can be relegated to intelligent machines.''

Giving completely paralyzed patients full mental control of robotic
limbs or communication devices has long been a dream of those working to
free such individuals from their locked-in state.

``These patients are completely paralyzed and cannot speak, yet they are
fully conscious and aware,'' Meeker said. Brainstem strokes, injury to
the upper spinal cord or diseases such as ALS (Lou Gehrig's disease
(news - web sites)) condemn tens of thousands of patients to such an
existence.

``Restoring even the most rudimentary communication or motor function to
such patients will drastically improve their quality of life,'' she
said.

Of course, roadblocks remain and the technology remains crude. Still,
Meeker said her group ``can achieve over 90% accuracy'' in carrying out
the simple commands the monkey has been trained to perform so far. A
technology that could replicate more complex movements--such as
handwriting a letter or playing games with a robot arm--could still be
years or decades away.

Even more challenging are ``Terminator''-type applications that would
allow the minds of healthy individuals to meld with machines--allowing
drivers to ``think'' their way through traffic, for example, or granting
pilots the ability to navigate the skies with their mind.

``The largest difficulty that I see with this kind of sci-fi application
is the current risk that brain surgery presents,'' Meeker said. ``This
kind of technology could save the life of a locked-in patient, but has
the potential to kill a healthy pilot if something went wrong.''

However, several groups are currently working to develop methods of
tracking brain signals without the use of surgically implanted
electrodes. Such a breakthrough could lay these types of safety concerns
to rest and ``revolutionize'' the field, Meeker said.