Citeren meekerdb <[email protected]>:
On 8/14/2013 6:41 PM, [email protected] wrote:
I guess I don't understand that. You seem to be considering a
simple case of amnesia - all purely classical - so I don't see how
MWI enters at all. The probabilities are just ignorance
uncertainty. You're still in the same branch of the MWI, you just
don't remember why your memory was erased (although you may read
about it in your diary).
No, you can't say that you are in the same branch. Just because you
are in the clasical regime doesn't mean that the MWI is irrelevant
and we can just pretend that the world is described by classical
physics. It is only that classical physics will give the same answer
as QM when computing probabilities.
Including the probability that I'm in the same world as before?
With classical I mean a single world theory where you just compute the
probabilities based "ignorance". This yields the same answer as
assuming the MWI and then comouting the probabilities of the various
outcomes.
If what you are aware of is only described by your memory state
which can be encoded by a finite number of bits, then after a memory
resetting, the state of your memory and the environment (which
contains also the rest of your brain and body), is of the form:
"The rest of my brain"?? Why do you suppose that some part of my
brain is involved in my memories and not other parts? What about a
scar or a tattoo. I don't see that "memory" is separable from the
environment. In fact isn't that exactly what makes memory classical
and makes the superposition you write below impossible to achieve?
Your brain is a classical computer because it's not isolated from the
environment.
What matter is that the state is of the form:
|memory_1>|environment_1> + |memory_2>|environment_2>+..
with the |memory_j> orthonormal and the |environment_j> orthogonal.
Such a completely correlated state will arise due to decoherence, the
probabilities which are the squared norms of the |environment_j>'s are
the probabilities. They behave in a purely classical way due this
decomposition.
The brain is never isolated from the environment; if project onto an
|environment_j> you always get a definite classical memory state, never
a supperposition of different bitstrings. But it's not the case that
projecting onto a ddefinite memory state will always yield a definite
classical environment state (this is at the heart of the Wigner's
friend thought experiment).
So, I am assuming that the brain is 100% classical (decoherence has run
its complete course), whatever the memory state of the brain is can
also be found in the environment.
Then the assumption that I'm making is that whenever there is
information in the environment that the observer is not aware of, the
observer will be identical as far as the description of the observer in
terms of its memory state is concerned accross the branches where that
information is different. So, if the initial state is:
|memory>|environment>
and in the environment something happens which has two possible
outcomes, and you have yet to learn about that, then the state will
evolve to a state of the form:
|memory>(|environment_1> + |environment_2>)
and not:
|memory_1>|environment_1> + |memory_2>|environment_2>
because the latter would imply that you could (in principle) tell what
happned without performing a measurement, and I don't believe on
psychic phenomena.
So, the "no-psychic phenomena postulate" would compel you to assume that:
|memory>(|environment_1> + |environment_2>)
is the correct description of the state and that only after you learn
about the fact you become localized in either branch. This applied to
the memory resetting implies what I was arguing for.
Saibal
--
You received this message because you are subscribed to the Google Groups
"Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/groups/opt_out.
