home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!cis.ohio-state.edu!rutgers!igor.rutgers.edu!planchet.rutgers.edu!nanotech
- From: mckendre@sal-sun70.usc.edu (Thomas Mc Kendree)
- Newsgroups: sci.nanotech
- Subject: Re: Multiple Experiences
- Message-ID: <Aug.29.01.49.26.1992.13201@planchet.rutgers.edu>
- Date: 29 Aug 92 05:49:27 GMT
- Sender: nanotech@planchet.rutgers.edu
- Lines: 35
- Approved: nanotech@aramis.rutgers.edu
-
- In article <Aug.21.18.26.44.1992.20235@planchet.rutgers.edu> hsr4@vax.oxford.ac.uk (Auld Sprurklie) writes:
- >Making a great many assumptions...
- >What was NMR and is now MRI (Magnetic Resonance Imaging) is a mechanism for
- >monitoring the chemical state of part of an organism (this necessarily very
- >vague). If the speed of MRI can be increased considerably to the point where
- >it can take a 'snapshot' of the chemical state of an entire brain, frozen in
- >action for a moment, and then process that chemical map to yield whatever
- >format would be necessary for an artificial brain, and then implement that new
- >map in the artificial unit....
-
- I believe that it might be necessary to take many 'snapshots' and actually
- make a movie, before you really understood how a brain was structured.
-
- From the standpoint of using MRI, however, there may be a much more important
- problem; I believe that the nature of MRI makes it impossible to get resolution
- down to the individual neuron, because of the fundamental resolution of the
- magnetic "rays." Even if that can be addressed (perhaps the wavelength can
- be reached, and one could have a giant battery of phased array SQUIBS, or some
- such), the way MRI works, it measures brain activity by looking at the
- distribution of an atomically tagged chemical (e.g. inject sugar with a special
- isotope. The cells that are "running hard" at the time draw in the sugar, and
- by using the MRI to look at the distribution of the isotope, infer the
- distribution of how neurons drew upon the sugar.)
- The upshot is that every neuron might not have the tagged isotope for any
- MRI scan to pick up, and even if they all do, there still is no obvious way
- to read out all the synapses. I would be very interested if anyone can show
- a non-invasive technique for reading off all the structure, and hence contents,
- of the brain. (Reading EOC gives one an idea of how this could be done
- non-destructively, albeit invasively.)
-
- [I wouldn't want to upload via such a (MRI) technique--much too vague,
- and too much chance of completely missing latent memories one isn't
- thinking about at the time. But by the same token, it might be a way
- to generate "partials".
- --JoSH]
-