the future of mice

Vincent Ferrera vpf3 at columbia.edu
Thu Apr 11 15:50:07 EST 2002


Ryan wrote:
> 
> I read that researchers from Brown University have implanted some
> electrodes into monkeys´ brains so that they are able to control a
> computer cursor with plain thoughts (in real time; using neurones for
> hand movement). Has anyone got suggestions on how to achieve the same
> thing in humans ? And would it be possible to develop such a mouse
> using a noninvasive technique ?
> 
> P.S. any relevant resources (links etc) would be most welcome

I think there are three problems with this:

1. Calibration.  With a real mouse, you have proprioceptive and visual
feedback to keep the motor command in line with the desired movement. 
With a brain mouse, you only have visual feedback.  The question is
whether visual feedback alone, with its inherent processing delays, is
sufficient to keep the system in tune.  In subjects who have use of
their own limbs, this might not be a problem because normal usage of the
limb should keep the neuronal signals calibrated.  People who lack use
of their limbs might experience a gradual drift in the relationship
between neuronal activity and desired limb movement.  Of course, this
assumes that the appropriate hand movement neurons could be found in
such subjects to begin with.

2. When you control a real mouse, you generally don't think about your
hand movements.  You think about the goal and then subconsciously run
the motor program that brings the cursor to that goal.  It isn't clear
what strategy the monkeys were using.  But if you had to control the
brain mouse by constantly thinking about its location, that might become
very distracting.

3. Why not just use eye movements?  Eye movement can be monitored
non-invasively and eye position is more likey to be correlated with
where you want the cursor to go than is hand position.




More information about the Neur-sci mailing list