[MURG] Cortex activity

[digfarenough]

digfarenough:
> but as joseph said, you can't take the analogy too
> far because while computers have a simple way to
> read out data, the brain doesn't--the computer has
> direct access to every bit of memory, something our
> brains don't have..
Thomas: Our brains are an old neantherthal design.
This example proves superiority of computers over
brains. Now imagine the brain paired with the
computer. All important as well as new memories are
stored in the computer part and easily accessible.


I know what you mean by saying computers are superior, but I don't think it's accurate.. my brain doesn't crash and lose all its information when a neuron dies and my brain doesn't corrupt information when I wear my magentic helmet

if you store memories in a computer the same way they're stored in a brain, they won't be easily accessible at all.. neuronal activity will be directly accessible, but memories are stored as subtle changes in synapses throughout many brain areas.. you can't just read it out and make sense of it like you can open a jpg on a computer


digfarenough:
we would have to use an external
> probe to access each neuron, but unlike a computer
> in which execution can be paused so the data doesn't
> change while it's being read, the brain goes on in
> time, so the entire state would have to be read
> nearly simultaneously to get all the data

Thomas: In a paired system it would be possible to
stop the computer part while allowing the brain to
think what to do next.


not the way you're proposing it.. you're proposing the computer become a fundamental part of the substrate supporting the mind, if you just stop the computer who knows what would happen.. definitely not something good..

 have you ever read "true names" by vernor vinge? you ought to.. it's a short story that you should be able to find online easily enough..


digfarenough:
> 
> as for your idea of a cortical interface between the
> brain and computer (which someone else had mentioned
> last week as well), the problem is that at *best*,
> the brain would use the computer as an addition to
> its neural network and would become inexorably tied
> to your identity...

Thomas: That would bring us very close to my vision of
mind uploading.


that's good, cause I was describing how I interpret the results of what you want to do.. which means I must at least partly understand it..


digfarenough:
I suppose if the number of
> neurons in the computer were much much larger than
> those in the brain (like orders of magnitude
> larger), and if you set up one of these huge
> networks for each function area of the cortex,
> eventually, if the brain was plastic enough, you'd
> be able to get enough function into the network that
> removing the brain wouldn't much effect the new
> cortex
> Thomas: In my view when that happens the mind is....
UPLOADED. Do we have a computer system big enough for
this? Yes we do! Imagine that the computer brain
extention is connected to the internet. Interactive
web sites would take care of various functions. And of
course you would have a direct intimate brain
connection to all the internet information.


so you're saying that the mind is the result of the cortical operation, not of the brain's operation.. you're ignoring most of the brain

also, just because the brain has a connection to information doesn't mean it can process it or that you'd be aware of it.. most of what the brain does is non-conscious and much of the information coming into the brain isn't made a part of consciousness

but also, I already have a connection to all the inteernet information, the pathway is internet->computer->eyes/ears->brain... in order for the information to be understandable, it has to at least go through sensory pathways, if you just run an html file into your cortex you won't suddenly know all the information in it, you have to read it and understand it the old fashioned way.. the way our brains work seem to make knowledge downloading impossible


Thomas: No you would not. I can very well relate to
your concern as I do creative work myself. Each time
you practice or even perhaps think about your music it
would be gradually bit by bit uploaded to the
computer. Some time ago I proposed what I called'
"selective mind uploading". That means if asked which
aspect of my mind I would want to upload if I had to
pick only one - my answer would be my talents in
mechanical design.


what makes you think your talents in mechanical design are somehow seperable from other parts of you?

I think for some reason you picture this all being a one-way transfer.. you think about something and the computer picks up on it and it all just slides up into the computer and disappears from the brain, then you serially go through everything you know and eventually it's all in the computer and the brain isn't needed anymore


Thomas: Don't worry about your mind's imperfections -
we all have them. Your left - right problem is a very
good example how computer augmented mind can be
improved. It would take a very simple function to
communicate the left and right to your brain. One step
further the computer can be equipped with a magnetic
compass solving your east - west problem.


oh, I don't worry about them

what makes you think it'd be so easy to communicate left and right to the brain? it seems like that'd be a big problem to me.. for instance, let's say I'm in a car and there's a turn ahead, I want to tell the driver which way to turn.. I can picture the direction in my head, but I have to think for a couple seconds about which word means that direction, usually I do the L on the left hand thing, or draw a line down the middle of the view in my head and figure out which is which.. or I'm left-handed, so if I have a pencil or something I can get it quickly.. a "normal" person on the other hand I would assume has simply associated the sense of leftness with the word left, so you know right away that left is that way and right is the other way.. how would a function in a program instill that association? it sounds like another attempt at knowledge downloading.. but maybe you can think of a better way to do it than I can..

Thomas: I am not exactly sure why extending the brain
should result in autism, but I do realise it can be a
risky business. The computer - brain communication may
be difficult to learn. Especially in the beginning
there could be serious coherence problems causing
confussion and in the worst case the experiment may
result in insanity. Autism could be one of the
problems.


no no, exactly as I said: I didn't mean to say they were the same, I was just offering another example... all I meant is that people should be free to do what they want to do, but that they should know what the results may be.. your idea is one example, voluntary autism which I got from a greg egan book is another example, replacing your body with a 50 foot solid gold robot is another possibility


Thomas: It doesn't have to be like digital photograph
as long as the cortex activity is consistent with the
objects seen. We can talk some symbolics here. After
some long enough observation we can create a cortex
dictionary - a particular activity will represent a
particular picture.


I had this same idea a year ago or so for a better interface with a computer, to scan out this information from the brain.. but there are many problems.. for instance, it would quite likely have to be seperately calibrated for each individual.. and that's *if* visual mental imagery actually shows up as an object shape somewhere in the brain, and personally I don't think it does, but I could be wrong

in addition, each person would have to seperately train the computer to identify each "symbol" seperately.. just to do an alphabet for text entry would take a significant amount of time.. 52 letters (or 27 if you have the alphabet and a prefix for capitalization), plus 10 digits, plus special symbols


Thomas:
I have seen that or similiar photo in one of the books
on the brain I have read. It was a monkey's brain. The
monkey was made stare at a light pattern and then it
was sacrificed. Corresponding dots were located on the
occicipal lobe. I don't remember what technique was
used to make these dots vissible.


it sounds like you're leaving something out, cause I don't see how that would be possible.. unless they immunofluoresced some protein that gets created when a cell is stimulated for an extended period of time.. but I'm not sure if that's possible..

-Eric

(oh.. I never told you my name.. you can call me that instead of dfe:)


---------------------------------
Do you Yahoo!?
Yahoo! SiteBuilder - Free, easy-to-use web site design software
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://askja.bu.edu/pipermail/murg/attachments/20030920/141d9c42/attachment.htm