If one implants electrodes into a monkey's brain in order to allow the monkey to control a mechanical arm, then is the monkeys able to move both it's natural and it's mechanical arms at once?
What happens if, say, the same signal might tell the mechanical arm to move up, and the natural arm to move down?
The interface between the brain and the hand is not like a USB computer port, and we don't unplug the biological arm and plug in a mechanical one instead. We implant electrodes to just a tiny fraction of the neurons in M1 (about a hundred, out of millions), and use their signals to control the mechanical arm.
So the answer is yes. The monkeys don't lose any dexterity in their biological arm because of the Brain-machine interface (BMI). Actually, during the BMI experiments, they usually move their biological arms all over the place. There is no reason why, after enough training, they couldn't control the mechanical arm almost completely independently of the biological arm and at the same time, like you control your digits (almost) independently.
This is indeed possible, but I haven't seen it done experimentally for reasons other than feasibility. Namely, since the exact same signal controls both arms, the amount of interesting learning to be had is significantly less than if the signal controls only one arm (e.g. tries to learn to pick up a banana with the mech arm whilst his real arm whacks him in the face).
Also, the standard paradigm of having the monkey control a mechanical arm via neuronal impulses is scientifically valuable because it allows for insights into how we come to know our body in order to manipulate objects in our environment. This two arm procedure doesn't seem to test any conceivable hypothesis about the brain and behavior, so its probably closer to animal torture than science.
Now, if you added electrodes into a separate patch of brain to test if the monkey had enough neuroplasticity to adapt its brain towards controling the new arm and remain control over both its natural arms, then I can start to see some reasons for trying this out. Still haven't seen anyone write about actually doing this though.
Theoretically at least, yes, it is entirely doable.
In the central nervous system including the brain, sensation, decision and reaction form a feedback loop comprising of different types of nerves. Afferent neurons sense an input such as an external stimulus, interneurons are the connectors in the local circuit, while efferent neurons carry motor signals that activate an organ.
You are apparently referring to a motor signal when you say 'the same signal'. Assuming such a signal is generated in some way and then transmitted to the organ(s), it is thereafter a purely mechanical process where the organ responds with a movement according to the signal.
It does not matter if the signal is multiplexed over to more than one organ, or if the organ happens to be natural or artificial. All that matters is the design of the artificial organ.
That is as far as the control systems part of it is concerned. However, the overall functioning is a bit more complex -- e.g., generating the intended signal, and multiplexing it to heterogeneous external organs.
