Brain-computer interfaces: Has science fiction become reality?
When a team led by John Donoghue,
PhD, and Leigh Hochberg, MD, PhD,
published results from their pilot study of
the BrainGate system in Nature in 2006,
the headline in one Canadian newspaper
proclaimed, "Movement by Thought: Science
Fiction to Fact." A London newspaper
referred to the trial participant as "the first
bionic man." The editors at an Oakland
daily were equally impressed, running the
headline: "Paralyzed Man Moves Mountains
It was hard for even the most serious
science journalists to ignore the fascination
surrounding the researchers' stunning
achievement. The scientists—from Brown
University, VA and other institutions—had
enabled a 25-year-old man with quadriplegia
to operate a computer cursor and
perform other tasks solely through his
The technology, called BrainGate, uses a
tiny sensor implanted in the motor cortex, the
part of the brain that controls movement. The
sensor, about the size of Lincoln's head on a
penny, has 100 hair-thin electrodes that pick
up brain signals. The signals are sent to an external
decoder that turns them into commands
for electronic or robotic devices. For now, the
brain implant is wired to a computer, but the
researchers hope to go wireless in the future.
Development of the system is spearheaded
by Donoghue, a Brown neuroscientist
who became affiliated with VA when
the agency established its Providence-based
Center for Restorative and Regenerative
Medicine in 2004. Donoghue is also chief
scientific officer at Cyberkinetics Neurotechnology
Systems, a company formed by
Donoghue and colleagues in 2001 to bring
BrainGate to market.
A 'nascent science'
Hochberg, lead author on the landmark
Nature paper and the principal investigator
on current trials involving BrainGate,
admits there has been some hype in media
coverage of the technology, but says most
reports have been balanced and accurate.
"Overall, many people have been captivated
by the potential of the technology. But
thankfully, the media has generally been
responsible in describing these as early trials
[and making clear] that this is a nascent
science—that we're really at the beginning
of a tremendous period of learning and opportunity
in terms of restoring lost function
for people with paralysis or limb loss."
'The focus now, for people with spinal cord injury, brain stem stroke, ALS, and other diseases or injuries of the nervous system, is to be able to restore movement and communication.'
—Dr. Leigh Hochberg
Notwithstanding Hochberg's tempered
view, it may be fair to say that in the case
of brain-computer interfaces, yesterday's
science fiction—for example, the 1938
Andre Maurois novel The Thought-Reading Machine—has indeed become
Even so, what's been realized by
researchers to date has clear boundaries.
BrainGate and similar technologies have
little applicability with regard to "higher"
functions of the human mind: that which is
uniquely individual, such as memory, emotions,
creativity. "For the moment, that's
a theoretical discussion," notes Hochberg.
"The technology is not even close to being
able to read into memories or thoughts in
the general sense. The leading edge of the
field is the ability to extract a neural signal
that's related to the intention to move one's
limb—and thereby a computer cursor—in a
And even the notion of "reading
thoughts," while not wholly inaccurate, is
more a handy catchphrase for the media
than a precise description of what the technology
is designed to do.
"'Thoughts' is a useful word because
it's immediately meaningful to everyone,"
says Hochberg, "and the concept of being
able to 'read thoughts' has been around in
science fiction a long time. But that's not
what we're doing in our current research.
The focus now, for people with spinal cord
injury, brain stem stroke, ALS, and other
diseases or injuries of the nervous system, is
to be able to restore movement and communication."
Hochberg is principal investigator on
two BrainGate trials now underway: one
involving people with ALS and related
motor-neuron diseases, the other for people
with spinal cord injury, muscular dystrophy
Enabling those with paralysis to
The main way in which BrainGate could
restore communication for people who have
lost motor ability is to enable them to move
a computer cursor, which in turn could
allow them to use email, the Internet and
word processing, or operate a TV set.
Similar results have been achieved
through somewhat different means by Dr.
Jonathan Wolpaw and colleagues at the
Wadsworth Center, part of the New York
State Department of Health. Their method
relies on EEG technology—electrodes
placed on the scalp, not inside the brain.
Users wear a breathable cap on their head
that contains eight electrodes—down from
64, just a few years ago—wired to a laptop
loaded with software that translates the
brainwaves into commands for devices.
This approach avoids some of the risks
of brain implants, and may eventually prove
viable for many patients with disabilities.
One drawback with EEG, however, is that
the output is less precise than with implants,
and users need far more training than with
BrainGate to effectively control a cursor.
Yet another method for restoring communication—this one focused on patients
who've lost their speaking ability—is being
developed by a private Georgia-based company
called Neural Signals, Inc., the only
other neuroprosthetics group worldwide, to
Hochberg's knowledge, that is using recording
sensors inside the brain. Their product is
a computer-controlled prosthetic device that
would be controlled by brain signals and
reproduce the sounds of natural speech.
Brain waves may drive natural or
As for enabling movement, BrainGate
has already enabled research participants to
open and close a robotic arm. This aspect
of the work—using brain signals to activate
limbs—may benefit from a new $6.5
million grant from the National Institutes
of Health to Cyberkinetics, Brown and the
Cleveland Functional Electrical Stimulation
(FES) Center, which is sponsored jointly by
VA and Case Western Reserve University.
The partnership with the FES Center
represents an intriguing melding of approaches.
Most past FES work has involved
people with intact but non-functioning
limbs—such as those with spinal cord
injury or stroke. Electrodes are implanted
not in the brain but in the weakened or
paralyzed muscles that would normally
move the limb. Small electrical currents
from external or implanted devices activate
the muscles and restore movement and
function. Only recently, research there has
expanded to prosthetics applications. In
one project, electrodes would be implanted
onto intact arm and shoulder muscles near
the amputation and pick up brain signals to
drive an artificial hand.
The new FES-BrainGate collaboration is
"potentially very promising," says Hochberg,
in that two groups of patients might
benefit: those using prosthetics limbs, and
those whose natural limbs are intact but
disconnected from the brain and nervous
system. Either way, what BrainGate
inventor John Donoghue has described as
the ultimate goal of the technology—"to
reconnect brain to limb"—may eventually
be within reach.