>>11957164New anon posting here.
Doing my grad studies at top 2 for neuroscience.
Brain cells can effectively be simulated by computers, because most of the biology of brain cells has to do with the limitations of it being a living system.
For example, it needs DNA to produce channel proteins, molecular motors, enzymes for metabolism, and so on.
This is the case for most cells (some don't have DNA such as red blood cells)
Brain cells, and cells in general, also have the structural lipid bilayer; another biological based problem.
The real intelligence part of the brain has (seemingly) almost entirely to do with ion flow in the cell, neurotransmitters at the synapse, and synaptogenesis.
A computer can easily model ion flow and neurotransmitter activity, since it contains enough molecules and ions to model statistically.
The real problem involves modelling synapse dynamics; ie receptor regulation, axon growth and guidance, cell survival by neurotrophins and so on.
I don't know if enough is known to emulate this statistically. My opinion is that enough is known that a computational model based on (for example) an electron microscopy map of a brain to get the specific architecture; or in neural nets, the connections between different nodes.
There would be problems because you wouldn't know which neurotransmitter is in which cell, and what type it is (inhibitor or excitatory), the dynamics of receptor regulation, the dynamics of synapse formation, the conductance and sheathing of axons, and so on.
But these things could be simulated by computers, and it should produce intelligence and consciousness (dependent on whether the architecture it's based on was intelligent and conscious).
The evidence that mental functions (almost) entirely have to do with ion flow, neurotransmitters, and synapse formation comes from studying the brain.
Simple examples include:
Caffeine affects neurotransmitter signalling
Benzos affect ion flow voltage mechanics
Synaptogenesis inhibition
