From: Damien Broderick (thespike@earthlink.net)
Date: Mon Jan 20 2003 - 13:04:48 MST
Anders Sandberg writes:
> Hmm, winning the lottery is such a large scale effect that decoherence
> would get you.
Is that necessarily the case? Certainly your state would decohere from those
where you chose a different set of numbers. I'd been thinking of the choice
as, finally, a 0/1 decision: my selected numbers = 1, any other result = 0.
Of course all those n million universes must differ minimally at the moment
when the macroscopic randomizer tumbles and selects its six balls. (In fact,
maybe this is never a suitable case for MWI, since perhaps there's no
quantum-level contribution to the outcome, which always happens the same way
in all `adjacent' universes? If so, we do have to adopt a `quantum Lotto'
generator for this gedanken.
> As far as I understand things, then
> you would get the fringes change as if the probability amplitude of the
> potentially blocked photon was decreased (how much depends on your
> setup). So for a very unlikely blocking you would get the fringes nearly
> as in the typical two-slit experiment, and for a more likely blocking
> they would move towards a no-fringe pattern.
That's one of the points I remain unclear about. In the Deutsch formulation,
interference occurs because of the interplay between measurable photons
(etc) and their `shadow' other-world counterparts (which I sometimes think
of, provincially, as `fauxtons'). If this actually happens, I suppose a
reduced number of worlds with the 2 slits open should cause reduced rather
than obliterated fringes. But I was under the impression that experiments
(either real or gedanken) by Chiao and colleagues had shown that even *in
principle* slit closure was enough to block interference. (Serafino kindly
sent me some references to, e.g., interaction-free measurements --
http://arxiv.org/PS_cache/quant-ph/pdf/0102/0102049.pdf -- etc, which are
damaging my brain even as we speak.)
> You still need the photons to get to the photographic film to get
> something measurable, so they will have to take the day to pass down the
> beam path.
Yeah, I guess so, in which case there's no obvious way to *use* such an
effect. Sigh. (Unless Elitzur/Vaidman etc, cited above, helps out...)
Damien Broderick
This archive was generated by hypermail 2.1.5 : Tue Jan 21 2003 - 17:10:21 MST