If a Monkey Can Do It...
From ArsTechnica:
[quote]Monkey uses thought to move cursors
Posted 3/14/2002 - 8:59PM, by Yaz
The latest issue of Nature reports that a group of researchers at Brown University have demonstrated that a monkey coupled with a neural implant can move a cursor with just its thoughts. One day this research may further enable paralyzed or physically disabled individuals to participate in activities previously restricted by their disability.
The researchers used "a Utah intracortical multi-electrode array" to measure the activity of neurons in the monkeys' motor cortex while they manipulated joysticks that controlled cursors. The monkeys were trained to point the cursor at a red dot and received positive-feedback in the form of orange juice. The researchers chose the best-performing monkey and refined their signal-movement algorithm. They randomly alternated between enabling the monkey to move the cursor with the joystick or its thoughts, and the monkey learned how to move the cursor without the joystick. If you have a subscription to Nature, you can check out the full paper here, and you can view a Quicktime movie of the monkey moving the cursor here.
Based on the movie's mediocre results--the cursor usually circles in on the target after a few corrections--either the algorithm has plenty of refining left, the monkey still is not very good at controlling the cursor with its mind, or the monkey was not great at moving the cursor to the red dot in the first place. Nonetheless, the results are encouraging since, compared with previous studies, this experiment utilized a smaller implant, required signals from fewer (7-30) neurons, and allowed for a greater movement repertoire. Similar studies include a 1998 demonstration by Emory scientists where a paralyzed man could move a cursor horizontally to have a computer speak phrases for him and a 2000 Duke report (Nature link) where monkeys controlled robotic arms through externally-wired implants.
Although an implant for humans may be available within 10 years, actually using the implant in disabled individuals presents an additional obstacle. We have to remember that there is an extended training period for the implants in which they are customized for each individual. In the case of the monkeys, the signal-movement algorithm could be refined based upon their use of the joysticks; however, if someone already cannot use a joystick, the algorithm will be significantly more difficult to refine. With no control signal, filtering out all irrelevant background neural activity and pinpointing the motor signals may take even more extensive training than the few months required for the monkeys.
Disabled people need not hold their breaths for the implant to become available though, since there are existing tools to help them use computers. Among them, my favorites are eye-tracking devices that enable people to type by focusing their eyes on an on-screen keyboard. Still, neural implants may afford individuals a degree of freedom not possible with other devices, and the related research may result in other exciting discoveries about humans' motor functions.
<hr></blockquote>
Very cool!! Thought you all would enjoy that.
[quote]Monkey uses thought to move cursors
Posted 3/14/2002 - 8:59PM, by Yaz
The latest issue of Nature reports that a group of researchers at Brown University have demonstrated that a monkey coupled with a neural implant can move a cursor with just its thoughts. One day this research may further enable paralyzed or physically disabled individuals to participate in activities previously restricted by their disability.
The researchers used "a Utah intracortical multi-electrode array" to measure the activity of neurons in the monkeys' motor cortex while they manipulated joysticks that controlled cursors. The monkeys were trained to point the cursor at a red dot and received positive-feedback in the form of orange juice. The researchers chose the best-performing monkey and refined their signal-movement algorithm. They randomly alternated between enabling the monkey to move the cursor with the joystick or its thoughts, and the monkey learned how to move the cursor without the joystick. If you have a subscription to Nature, you can check out the full paper here, and you can view a Quicktime movie of the monkey moving the cursor here.
Based on the movie's mediocre results--the cursor usually circles in on the target after a few corrections--either the algorithm has plenty of refining left, the monkey still is not very good at controlling the cursor with its mind, or the monkey was not great at moving the cursor to the red dot in the first place. Nonetheless, the results are encouraging since, compared with previous studies, this experiment utilized a smaller implant, required signals from fewer (7-30) neurons, and allowed for a greater movement repertoire. Similar studies include a 1998 demonstration by Emory scientists where a paralyzed man could move a cursor horizontally to have a computer speak phrases for him and a 2000 Duke report (Nature link) where monkeys controlled robotic arms through externally-wired implants.
Although an implant for humans may be available within 10 years, actually using the implant in disabled individuals presents an additional obstacle. We have to remember that there is an extended training period for the implants in which they are customized for each individual. In the case of the monkeys, the signal-movement algorithm could be refined based upon their use of the joysticks; however, if someone already cannot use a joystick, the algorithm will be significantly more difficult to refine. With no control signal, filtering out all irrelevant background neural activity and pinpointing the motor signals may take even more extensive training than the few months required for the monkeys.
Disabled people need not hold their breaths for the implant to become available though, since there are existing tools to help them use computers. Among them, my favorites are eye-tracking devices that enable people to type by focusing their eyes on an on-screen keyboard. Still, neural implants may afford individuals a degree of freedom not possible with other devices, and the related research may result in other exciting discoveries about humans' motor functions.
<hr></blockquote>
Very cool!! Thought you all would enjoy that.
Comments
However, several groups are currently working to develop methods of tracking brain signals without the use of surgically implanted electrodes
This isnt New news... I remeber seeeing stuff relating to this on TLC a few years ago using humans to think/move a ball on a comp. screen around by trying to think it to the left or right... they had some success without implants, just had a cloth helmet and a few jelled up detectors touching the temples...
Getting a monkey to move a cursor around a screen by thinking (something that has already been achieved by humans by use outside internal implants) using implanted detectors into brain tissue is extremely dangerous, this seems to me to be a step backwards...
------------------------------------
© FERRO 2001-2002