[Neuroscience] Re: A Purely-Electronic Brain -- Possible?

r norman via neur-sci%40net.bio.net (by r_s_norman from _comcast.net)
Fri Apr 20 19:28:06 EST 2007


On Sat, 21 Apr 2007 02:06:40 +0200, Josip Almasi <joe from vrspace.org>
wrote:

>r norman wrote:
>> 
>> I recall Roger Sperry describing in class some 45 years ago his
>> experiments showing how general electric fields could not be
>> responsible for "gestalt" phenomena -- he shortcircuited brain regions
>> with tantalum wires and isolated brain regions with mica insulators
>> but neither intervention had any effect on cortical function.  
>
>But it's not about gestalt effect how you or Sperry imagine it.
>Quote from thread 'Molecules and Neurons - and photons' on 
>comp.ai.philosophy:
>(STM=Short Term Memory)
>
>'The STM within our brain as well as within networks of honey bee or 
>other animals is a product of a "3D field-like" effect that creates 
>memory trace.
>In fact the name STM is very unfortunate here and in my opinion the term 
>memory trace (MT) much better reflects underlying processes.
>The purpose of this "3D field" is to provide a very effective mechanism 
>of correlating multidimensional signals within the context of its past. 
>The "brain" didn't have a much choice here, careful consideration 
>(reverse engineering) of the temporal binding problem indicate that it 
>is the only possible solution available - neurons create "temporal-bind" 
>with each other within this field. Consequently the correlation outcome 
>is the superposition result of past memory traces ( "3D fields" ) where 
>new neurons are "hired" and incorporated/build into the network.
>
>You can't partition this process, you can't simplify this process, you 
>can't localize this process.
>Your model has to capture complete "temporal-bind" of the field with 
>many its distribution properties.
>
>- what a "memory trace" is?
>
>It is simply a change in a state of the physical 3D media between 
>neurons that represent there mutual activation histories.
>3D memory tracings have to occur within very short time slots ~3-10ms 
>and be registered as extremely low continuous media changes. At present 
>we don't have technology to register this type of process.
>
>- Where did you get that from?
> From my own research.
>
>/quote
>
>> The speculations of many physicists over the past century or more
>> about brain function have not yet led to breakthroughs in our
>> understanding.  Will one of these three prove to be the exception?
>
>Maybe will, maybe won't.
>One thing sure - ignoring them won't lead to any breakthrough.
>
>After all, sensory physiology of Helmholtz was the basis of Wilhelm 
>Wundt work, who is considered one of the founders of experimental 
>psychology. So maybe shrinks could learn from physicists again?;)
>
>Regards...

I will simply comment that neurophysiology, which looks specifically
at cellular mechanisms, and neurobiology in general, which looks
specifically at experimentally accessible mechanisms with actual
evidence to support them, is a rather different world from that of
artificial intelligence or behavioral psychology (I except
physiological psychologists, many of whom are indistinguishable  from
neurophysiologists) or speculations of  physicists or ... or ... ....


More information about the Neur-sci mailing list