backtop


Print 75 comment(s) - last by MrJustin5.. on Jun 19 at 6:07 AM


The fast macaque monkeys quickly learned to grab treats with their new mechanical arm. What was surprising was the advanced natural motions they developed.  (Source: Andrew Schwartz/University of Pittsburgh)
An outlandish experiment is breaking exciting ground in the world of brain controlled prosthetics

While the jury is still out on whether monkeys deserve human rights, one thing's for sure -- they're good at controlling robotic arms.  In the past, humans have been able to control a computer mouse with brain signals.  In more recent DARPA grant research, prosthetic arms have been implanted in humans with basic control from electrodes on skin or electrodes implanted in muscle.  However, without directly interfacing with nerves, preferably near the brain, it is impossible to gain the fluid motion that human limbs have according to the current line of thought.

Researchers at Caltech recently made breakthroughs in repositionable neural probes, which will help such brain connections be made.  Now, researchers at the University of Pittsburgh and Carnegie Mellon University have taken the next step and for the first time ever have showcased the use of one of these nerve interfaced limbs with a nonhuman primate.

The researchers selected monkeys for the study due to their anatomical similarities to humans and their strong capacity to learn.  In the experiment, two Macaque monkeys were initially allowed to play with a joystick to get the feel for the basic capabilities of their new mechanical arm.  The arm featured shoulder joints, an elbow joint, and a two-fingered grasping claw.

Afterwards, a grid of electrodes the size of freckles were implanted just beneath the monkeys' skulls on their motor cortex.  The grid contained 100 electrodes and was placed on a section of the brain known to signal hand and arm movements.  Each electrode connected to a separate neuron, and the signals ran back out of the brain via wire to computers for processing.

The device collected the firings on the neurons and used it to generate a movement response which was sent to the arm.  The monkeys quickly learned from this biofeedback how to perform basic arm movements.  Within several days the monkey needed no assistance.  Sitting motionless they moved the arm like a normal limb, using it to delicate pick up small objects like grapes, marshmallows, and other chewy nuggets which were held in front of them.  Over two-thirds of the time the goodies found their way to the hungry little monkeys' mouths.

The monkeys learned to approach the morsel with an open "hand", to close the hand, and to slowly release as they bit into it.  They shocked researchers showcasing advanced movements, such as using one finger to pick up a sticky item by poking it, keeping the hand open.  They also would bring their arm by their mouth to lick clean, and would use it push morsels of food dangling from their mouth back in.  Researchers wrote that these were "displays of embodiment that would never be seen in a virtual environment".

Dr. Andrew Schwartz, a professor of neurobiology at the University of Pittsburgh and senior author of the paper on the research states, "In the real world, things don’t work as expected.  The marshmallow sticks to your hand or the food slips, and you can’t program a computer to anticipate all of that.  But the monkeys’ brains adjusted. They were licking the marshmallow off the prosthetic gripper, pushing food into their mouth, as if it were their own hand."

The new paper, released in the online journal Nature, is coauthored by Meel Velliste, Sagi Perel, M. Chance Spalding and Andrew Whitford.  The paper demonstrates that human brain-controlled prosthetics while not affordable in cost or difficulty, are technically feasible, or within reach.

Says expert Dr. William Heetderks, director of the extramural science program at the National Institute of Biomedical Imaging and Bioengineering, "This study really pulls together all the pieces from earlier work and provides a clear demonstration of what’s possible."

Another expert, Dr. John P. Donoghue, director of the Institute of Brain Science at Brown University commented that the paper is "important because it’s the most comprehensive study showing how an animal interacts with complex objects, using only brain activity."

One major problem that remains is that brain electrode grids currently fail within months for unknown reasons.  Furthermore, the system is cumbersome and needs calibration.  Also, so far a safe wireless interface has not been demonstrated, necessitating wires through the scalp.  However the researchers are striving ahead, looking for solution to each of these problems.

Dr. John F. Kalaska, a neuroscientist at the University of Montreal in an accompanying article in the Nature journal says that once the bugs have been resolved, researchers may be able to find other areas of the brain to give the limbs even more delicate response.  Kalaska says such possible future systems, "would allow patients with severe motor deficits to interact and communicate with the world not only by the moment-to-moment control of the motion of robotic devices, but also in a more natural and intuitive manner that reflects their overall goals, needs and preferences."



Comments     Threshold


This article is over a month old, voting and posting comments is disabled

Third arm?
By EODetroit on 5/29/2008 3:05:02 PM , Rating: 2
So could someone learn to control a third mechanical arm naturally in addition to their two natural ones... or would that take too much concentration?




RE: Third arm?
By Brandon Hill (blog) on 5/29/2008 3:10:45 PM , Rating: 2
Wouldn't you have to have the brain "pathways" to control a third arm first?

Otherwise, it's be like turning a pedaling on a bike without the chain attached to the rear wheels...

That's my reasoning, and I'm sticking to it ;)


RE: Third arm?
By AlvinCool on 5/29/2008 3:45:43 PM , Rating: 2
Actually, you might not have to have a different set of pathways for the third arm. The monkeys are not moving their original set of arms (lets hope the researchers didn't lop them off with cleavers). This gives rise to the thought that they know that simply thinking of moving the arm makes the mechanical arm move. So if they can do that it may be possible for you to move the real arm without thinking of the mechanical one. So the question is, can I use the set of pathways to move one or the other, but not both at once.


RE: Third arm?
By elpresidente2075 on 5/29/2008 4:10:53 PM , Rating: 2
The thing is, the brain does not work in the linear fashion that you seem to think it does. Since it is a self-healing, highly adaptive network, one could learn to use any number of any sort of peripherals, from arms to keyboards (sans keys) to PBX switches, much as you are able to learn to play golf or play cards.

Now, while my last example is a bit out there, the realm of potential for brain-machine interfacing is much greater than mere artificial limbs.

And yes, this research does open the door to the kind of hardware found in The Matrix, in much the same way the discovery (invention?) of the wheel opened the door to the Space Shuttle.


RE: Third arm?
By Baov on 5/29/2008 4:11:28 PM , Rating: 2
I'd think of it as software emulation of the hardware pathway for the third arm. Higher brain functions can imagine all kinds of crap, just like you can control control another person's movement by telling them, in words, to move left or right or stop. Not the most efficient way of doing it of course, just like it's more efficient for a GPU to decode your bluray movie vs CPU.


RE: Third arm?
By DeltaZero on 5/29/2008 4:42:25 PM , Rating: 2
I believe we are only talking about motion neurons here. Higher brain functions like "imagining" are much more complex.
For one thing, how can you learn to control an ability that you did not know were there? Not in two days you can't.


RE: Third arm?
By apollo7 on 5/30/2008 12:32:44 AM , Rating: 2
quote:
...it is not the spoon that bends, it is only yourself.


RE: Third arm?
By bobdeer1965 on 5/30/2008 2:36:48 AM , Rating: 1
What does "turning a pedaling" mean.

Look at your post below.

Wouldn't you have to have the brain "pathways" to control a third arm first?

Otherwise, it's be like turning a pedaling on a bike without the chain attached to the rear wheels...

That's my reasoning, and I'm sticking to it ;)


RE: Third arm?
By aBott on 5/30/2008 4:26:46 PM , Rating: 2
What is wrong with you?


RE: Third arm?
By Visual on 5/30/2008 5:42:00 AM , Rating: 2
In an earlier set of experiments, a monkey has been trained to control a cursor on the screen, and a robotic arm in a different room. This was initially done by monitoring its brain activity while it was using a joystick to move the cursor and eventually re-programming the computer to react to the brain activity instead of the joystick. For a while the monkey continued using the joystick even while it was actually controlling the cursor directly with its brain, but soon it learned that it doesn't need the joystick and was eventually able to control the cursor completely independently from its hand.

http://www.newscientist.com/article/dn4262-monkeys...

What I'm trying to say is, there are no hard-coded "pathways" in the brain. They can change and adapt and new ones can be created - that is called brain plasticity.
So it is completely likely that people might be able to learn to control a third or fourth arm, a tail, wings, wheelchairs, lights at home, TV channels and USB printers ;) OK maybe not the last one... there sure is a limit to this adaptability, but who knows where it is? Maybe we can even control a whole second body? Or a small army of them...


RE: Third arm?
By geddarkstorm on 5/30/2008 1:52:01 PM , Rating: 2
Yep, right on. The brain has the ability to change shape even in response to the simple thoughts and day dreams we entertain throughout the day. It is truly an incredible organ with potential limited mostly by only our imagination. As long as the interface works well enough, it wouldn't be too hard to learn to use a third, robotic arm, or four of them like Doc Oct.


“Then they pop up and say ‘Hello, surprise! Give us your money or we will shut you down!' Screw them. Seriously, screw them. You can quote me on that.” -- Newegg Chief Legal Officer Lee Cheng referencing patent trolls














botimage
Copyright 2014 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki