RogerBlack
Senior Member
- Messages
- 902
The best thing that could happen is that neurology develops into a more mature science that uses technologies that produce validated evidence leading to appropriate treatments based on sound evidence.
This is almost a 'you can't get there from here' problem.
In order to produce 'proper' results from neurology, you damn near need to get to complete brain/body simulations.
A hundred neurons (or even a million) is not going to get a meaningful result. It can say things like 'response to X artificial stimuli was 4% down under the measured conditions'.
It can't say things like 'X causes depression and anxiety'.
There are 1500000000000000 or so junctions between neurons in the human brain, and to go 'bottom up' - you nearly need to simulate the whole number.
FMRi is attempting to go 'top-down' - but has problems in that the signals are small, the resolution (both in time and volume) is terrible, with millions of neurons and many billion synapses in the sampled volume, and a thousand firings in the time resolution.
Electrodes are impractical in humans, and almost meaningless in other species, and only get at most a few hundred samples.
MEG is a magnetic technique to pick up large scale correlated activity in the brain, but it suffers from similar, though different problems as MRI, with the notable one that it only works at the brain surface.
And even if you got to brain/body simulations, then fun ethical problems arise.