X-Message-Number: 3720
From:  (Joseph J. Strout)
Newsgroups: sci.cryonics
Subject: Re: Uploading (not just explicit memories!)
Date: 21 Jan 1995 18:35:36 GMT
Message-ID: <>
References: <3fldch$>

In article <3fldch$>, "Bruce Zimov"
<> wrote:

> Here is why uploading the subjective circuit is important. Suppose you 
> could not upload the subjective circuit, but suppose that everyone's
> subjective circuit was qualitatively identical, but not numerically
> identical. One might get the bright idea of uploading just the memories
> in a kind of memory swap between the 2 individuals. This is Locke's
> example.  In this case, rather than being transferred to the other
> body, you would in fact, be staying in your own body and just thinking
> you were the other person, even to the point of believing you had 
> transferred since that's how it seemed to you.  This is the case
> of keeping the identity that matters with the subjective circuit.
> I think if you clear your mind introspectively you can feel that
> your memories aren't contributing to your subjective state. By
> clearing introspectively I mean eliminating all intrusive thoughts
> and recollections meditatively, and using sensor deprivation to 
> suppress object recognition.

People have been talking about this "subjective circuit" in curious 
terms.  It almost sounds like some folks still think there's a little 
person in their heads who watches what's going on.  You can swap the 
memories, the knowledge, even (presumably) fears, aspirations, and 
personality traits, but if you don't swap the little person watching the 
screen, then you haven't REALLY swapped people.

Clearly this is ridiculous (is in not?), so I must wonder what is 
intended by the "subjective circuit".  Since many theories of personal 
identity are based on memory (and generally only explicit memory is 
discussed), perhaps the term refers to those aspects of mind not covered 
by explicit memory.  Let's consider what those are, and whether they 
might be transferred or missed by uploading.

Explicit memory refers to memories we recall consciously, e.g, what did 
you have for lunch?  Since the 1980s or so, implicit memory has gotten a 
lot of research as well -- this refers to memories which we do not recall 
consciously, but nonetheless affect our behavior.  There are many types 
of implicit memory, including motor skills, habits, priming effects, and 
so on.  The difference is most striking in severe amnesic patients, such 
as the famous H.M.  Such patients have drastically impaired explicit 
memory, but retain normal implicit memory.

[Example (feel free to skip ahead): if we talk about reed instruments 
for a while, the different types, how they work, etc., and somewhat 
later someone asks you to quickly write down the word "read", there will 
be a higher probability that you will write "reed" instead.  If I ask 
why, you'll suppose that it's because we were talking about them earlier 
so it was "on your mind."  H.M. will show the same effect, but he'll 
have no idea why he wrote "reed" rather than "read", as he has no memory 
of the conversation.  Thanks to S. Zola-Morgan for this example.]

In addition to both types of memory, there are also personality traits, 
abilities, and such which are probably a combination of genetics, 
learning, and other environmental effects.  Clearly, all of these 
factors are important parts of who we are.

Would these factors be transferred properly in uploading?  Our knowledge 
of the underlying mechanisms is incomplete, but we must work with what 
we have.  Explicit memories appear to work through a centralized system, 
of which the hippocampal formation is a key component.  Implicit 
memories appear to be maintained by a wide variety of structures such as 
the cerebellum, basal ganglia, etc.  A number of deep brain structures 
have been shown to be involved with personality traits, and of course 
the cerebral cortex serves important functions related to motivation, 
sensation, motor pattern initiation, and language.

It's conceivable, then, that one could perform a "partial" upload, which 
would transfer some of these structures, but not all of them.  To me, 
this sounds more difficult than ("simply") uploading the entire brain.  
You'll still have interface issues with the senses and muscles, but 
those should be much simpler than interfacing one brain system to 
another.  Of course, if the upload could fail to replicate the hormonal 
balance, etc. of the original, which might produce personality changes 
until the problem is fixed.  But medicines can cause the same changes, 
yet we generally assume the patient is the same person.

Conclusions: (1) when discussing personal identity, we must consider all 
the brain's functions, rather than just explicit memory; (2) whatever 
"subjective circuit" really means, it's in the brain somewhere, and 
there's no reason to imagine that it won't be transferred by a 
full-brain upload.

-- 
Joseph J. Strout                   Dept. of Neuroscience
                   Univ of CA, San Diego

Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=3720