Posted by Lurker on May 31, 2000, at 22:29:15
In reply to Esoterica, Part I, posted by Adam on May 31, 2000, at 21:39:25
Who needs to build models for experiments, I think we should use prisoners on death row.
> Bob brought up the interesting problem of models above. I often wonder how progress is
> advancing on modeling the human brain. It seems the reductionist approach that is
> becoming de rigeur in the study of the brain and its function has left us with so many
> different transmitters, receptors, foci and stimuli, that it's virtually impossible to
> keep it all straight. We may progress from specialists in neuroanatomy to specialists
> in the cingulate gyrus or the or the anterior thalamic nucleus, if we are not at or
> beyond that point already. It is conceivable that the separation between such specialists
> might produce an unwieldy and unrealistic conceptual separation between the actual
> structures and phenomena being studied.
>
> So, what is a scientist to do when he or she wants to know what tweaking neuron A in
> region X does to neuron B in region Y, if anything, without having to spend endless
> hours combing through the reams of information that might pertain to the actual question?
> Maybe all the information from the study of anatomy, functional imaging, signal
> transduction, genetics, psychology and pharmacology could be consistantly incorporated
> into a model of the brain that would simulate physiological processes based on what we
> know and thus give us some predictive information on which to base hypotheses. I'm sure
> someone out there in the artificial intelligence or psychobiology fields might be trying
> to do something similar. But are they, for instance, trying to make a model that would
> tell them what, if I have a drug that interacts in a certain way with some-or-other
> receptor(s), is the most accurate and comprehensive picture of what might happen if I
> administered said drug to a human being?
poster:Lurker
thread:35445
URL: http://www.dr-bob.org/babble/20000526/msgs/35450.html