neuron network computer

[quadrapod]

recently it has been discovered that neurons grow readily on electrodes and will not grow readily on silicon if there is an electrode present. this could enable the fusing of technology and the brain. i personally think that you could apply this to another experiment in which the section of the brain in charge of motor functions was tapped to allow the controller to move an electronic arm or cursor on a computer. what do you think

the chips are made by doping silicon so that it possesses an electrolyte-oxide-semiconductor and is capable of picking up signals sent by the brain.

LiveScience.com
Wikipedia
USC - University of Southern California

[Buffy]

would be interesting for analysis, but its really better to figure out how to model neural networks than to figure out how to support brain tissue in a machine:

• The infrastructure necessary to keep the tissue alive is non-trivial
• Neurons are actually incredibly slow: even vonNeumann architectures emulating them can keep up with simple operations and massively parallel neural networks will put them to shame (if we can figure out how to grow them)
• The moral issues are, uh, touchy. Stem-cell controversy on steroids (excuse the mixed metaphor!)

You could, but its not something I see people ramping up to commercial production...

Wetware,
Buffy

[InfiniteNow]

Originally Posted by quadrapod

recently it has been discovered that neurons grow readily on electrodes and will not grow readily on silicon if there is an electrode present.

This is an interesting concept, and I see it as being a step toward telekinesis (of the computer assisted variety)... but that is a bit more sci-fi than science.

Can you verify that said what you intended in the quote above? It seems to contradict itself.

"Neurons grow readily on electrodes."
"Neurons will not grow readily on silicon if there is an electrode present."


I'm kinda slow sometimes. Maybe you can clarify?


Cheers.

[quadrapod]

the silicon is covered with small electrodes the electrodes are not made of silicon but of a very fine layer of gold. the neuron grow on the gold electrode and uses the electrode to send signals to other neurons. after much testing with the output of brain matter when it is stimulated by electricity a program has been made that understands these signals. the only problem i seem to see with this research is it doesn’t seem to have any pay off. i don’t see how t can be used and that is why i propose the question. how can this be applied!

[akshat]

Originally Posted by quadrapod

how can this be applied!

Depends on the cost. Any application would have to justify the cost, not only morally but economically as well.

Prosthetics is one surefire application. Another could be robots (yes, imagine robots as intelligent as a few thousand neurons ) One could model veryyyyy simple organisms.