AI and cognitive psychology rant (getting more and more OT - tell me if I should shut up)

[Stephen Horne]

On 29 Oct 2003 05:23:26 -0800, mis6 at pitt.edu (Michele Simionato)
wrote:

>Stephen Horne <steve at ninereeds.fsnet.co.uk> wrote in message news:<ff7tpvknjm2fa99ab6068lilthn8h0dmhr at 4ax.com>...
>> Not exactly. When we look at superpositions of subatomic particles,
>> there are observable artifacts of the interactions between
>> superpositions - the interference patterns. Without those
>> interactions, the theory of superpositions would be pointless as there
>> would be no effects of superposition to observe - the theory would
>> have no predictive or explanatory power.
>> 
>> My point is that the cat is superposed in the same way as the
>> subatomic particle, and yet we are unable to observe any artifact of
>> that superposition. *All* we can see is a single state resulting from
>> the waveform collapse when we observe the cat, but this is
>> emphatically not the case with subatomic particles where we can
>> observe artifacts of the superposition itself.
>> 
>
>There is no such a big difference between the Scroedinger's cat experiment
>and the narrow slits experiment:
>
>
>               |           | 
>               |           |
>                 A         | S
> P             |           |
> ->            |           |
>                 B         |
>               |           |
>               |           |
>
>The particle P passes through one of the slits A,B and ends its life on the 
>screen S, producing a spot. Getting the spor corresponds to opening the box in
>the cat experiment. If you repeat the experiment many times 
>with many particles, the distribution of the spots is an interference 
>pattern. But if you perform the experiment with a single particle, then
>you have 50% of probability of getting the spot in the upper side of the 
>screen, and 50% in the lower side (assuming a symmetrical experimental
>disposition). This corresponds to have the cat  50% of times alive
>and 50% of times dead when we open the box. Having only seen the spot in S, 
>it does not make sense to ask whether the particle passed through A or through 
>B, exactly in the same sense that it does not make sense to ask if the cat 
>is dead or alive before opening the box (this according to the
>orthodox interpretation). If you watch to see if the particle really passes 
>for A you necessarely perturb the system. For instance, you can close the 
>slit B, and make sure that the particle passes trought A, but then the 
>interference disappear.
>When the particle reach S and you get the spot, then the collapse
>of the wave function happens. It is less spectacular than opening
>the box, but in principle it is the same operation. I will not
>claim that the collapse is understood, not that it is an easy
>task to understand it :-(

OK, but there is still an artifact. In order for a sufficiently large
number of repetitions to build an interference pattern, the points on
S cannot be simple projections of P through A or B.

With the cat, the there are two possible final states - either the cat
is alive or dead - essentially the 'projections' of the initial
radioactive decay through the cause-and-effect chain with the detector
and poison.

There is still a clear artifact at the microscopic scale (the point
detected on S is NOT a simple projection through either A or B)
whereas no such artifact can be detected at the macroscopic scale (the
cat ends up either alive or dead, just as it would without any quantum
considerations).

I don't claim to understand quantum theory by any means, but I do know
that in principle if there were no artifacts of quantum effects no
sane human would have invented them.

So why don't we find cats that are half-dead and half-alive (between
the projections of A and B), or cats which are more dead than dead
(further to the A side than the projection of A)?

Perhaps cats simply don't have a particle/wave duality issue to worry
about.


-- 
Steve Horne

steve at ninereeds dot fsnet dot co dot uk