In a double slit experiment a wave can become what seems to be a
particle by being "observed" at the slit. What would happen if such
a particles were NOT observed thorough the second series of slits?
Would they still be particles or would they turn back into waves?

This experiment would determine
whether the observation "took
something away"

Drinking Tim Horton's double double during the experiment is
optional

That's the plan ... first pair of slits collapses, and the second pair of slits will help us determine whether we have a wave or whether we have a particle.

What I'm trying to get at is this... right now we assume that photon is changed to a particle because the interference pattern disappears. That's a reasonable conclusion. But it doesn't answer this question: Does that mean it lost it's wave property forever? Or will the interference pattern come back once the photons continue through the second pair of slits? (no observation at the second pair)

Here is a fairly simple explanation for what goes on
during a two-slit experiment.

Start by realizing the space through with a photon
moves is not empty; it is full of "virtual particles"
popping into temporary existence, and vanishing again.
And it is known that real particles like photons can
interact with those virtual particles.

A photon carries momentum. If it hits a virtual particle,
it is possible for some of the photon's momentum to be
temporarily transferred to the virtual particle. The one
virtual particle might even transfer it to another.
Remember, the virtual particles MUST disappear, and
the Law of Conservation of Momentum requires that the
photon get its momentum back, eventually.

During a two-slit experiment, just imagine the photon
as a particle passing through one slit. However, some
of its momentum can get carried by virtual particles
through the other slit. THAT suffices to lead to an
interference pattern.

Do remember that the two-slit experiment has been
performed with electrons and even whole atoms. That
should make it easier to imagine some of their
momentum getting carried by virtual particles through
the slit that the real particles don't go through.

// is it possible that light is a wave, or waves, of
statistical likelihood// That's exactly what a photon
is. In fact it's exactly what everything is.

// A wave of probability of particles materialising?//
Not really. It's a wave of probability of particles
being detected.

[WcW], since this is the HalfBakery, I'm free to make up
any
stuff I want. In this particular case, though, the "stuff"
FITS.

It is most certainly a fact that real particles constantly
interact with a surrounding sea of virtual particles. And
logical consequences are logical consequences.

The way momentum can be distributed/passing
through the two slits can perfectly explain the
interference
pattern, without needing to care which slit was used by
the real/moving particle.

This thing about atoms behaving in quantum ways is very
interesting. Is it correct to deduce that the result of this
double double slit experiment would be that photons
would behave as particles through the first slits and as
waves through the second pair of slits?

//Is it correct to deduce that the result of this
double double slit experiment would be that
photons would behave as particles through the first
slits and as waves through the second pair of
slits?//

Photons behave as photons. Which behave very
much like all other things. If you ask where
something is, you will be given an answer, plucked
out of nowhere on the basis of a probability
density function. As soon as you stop asking for a
definite position, the thing will not be anywhere -
it will just be a probability density function.

Ultimately, it comes down to the computational
capacity of spacetime. If you consider all the
factors that could influence the position of, say, a
photon, it's obvious that spacetime cannot possibly
calculate exactly where that photon is in realtime.
So, spacetime works like an overworked
accountant - it can't keep track of every pound,
but if you say "give me a figure for revenue", it will
spit out a plausible and precise number.

And when spacetime does pluck that precise
number out of thin air, it only does it for the
person who asked the question. If the person asks
for the revenue, the accountant will say "£345.67";
but if another person asks the same question, and
if they're out of earshot of the first person, the
accountant will grab another number out of the air
and will say "£389.23".

To summarize:
(a) quantum fuzziness is spacetime's way of
keeping up with all the maths.

(b) asking for a number will produce a number,
arbitrarily chosen from the fuzzy range of available
answers. This is "collapsing".

(c) Quantum collapse is relative, like everything
else. Just because spacetime has given you a
definite answer only means that there's a definite
answer for you; I may not be given a definite
answer, or if I ask for one it might be a different
answer.

So can you DoS spacetime by collapsing spacetime way
too frequently?

What if we built a machine that collapses spacetime
trillions times a second in a certain area? Would it slow
down time in that region relative to others?

Maxwell, that way of looking at uncertainty is hubristic. Our primary problem is the presumption that we already understand all of the discrete components. When you pound a nail into dry wall you don't get to claim "computational conservation" when the nail goes through more easily in one place than the other, you basically have to concede that there are other things under the drywall we don't have a model for as yet. Tagging out with a complicated dodge like "it's all a big computer simulation" is simply a way to say "we think math can help us understand this".

//the presumption that we already understand all
of the discrete components.// But it has been
proven pretty conclusively that no "hidden
variables" model is valid.

//a complicated dodge like "it's all a big computer
simulation"// No no no. I'm not a Matrix freak,
and physics is not a computer simulation.
Nevertheless, the actual physical behaviour of
things is complex, and spacetime itself has to
compute that behaviour. Think of spacetime as an
analogue computer, where the computation *is*
the phenomenon. It is pretty clear that if it takes
five pages of maths to describe the behaviour of a
proton in a magnetic field, spacetime has to work
equally hard to make it behave like that. _That_
is the point I was making.

There is a limit to the ability of spacetime to
figure out what everything is meant to be doing all
the time, so it gets by with averages until you
press it for a specific answer.

From my perspective that's a paradoxical view because the lively nature of the thing is very much the hidden variable that I am talking about, less a computer and more a writhing explosive underpinning to the makeup of the universe that seems uncertain and chaotic from the front side and looks orderly and predictable from the back. Spooky action at a distance is only confusing from the side that experiences distance, particle uncertainty only seem uncertain when you cannot see the fabric that twists together to make the particles and the places where they are not. I suspect that we will eventually be able to make models that allow for meaningful predictions that go much further than "measure closely and the quantum computer goes gonfable and spits out random data". It seems that true random behavior is abhorred which fits very well with concrete, "computational", models.

One thing that I still don't understand. Is it true that with
a single photon you can't determine if it's behaving as
particle or wave? .. because to have confirmation of a
wave you need interference pattern which means a large
amount of photons?

Quantum entanglement is a lot like marriage. The state
of one spouse's correctness on any subject can be
immediately be determined by the state of their other
spouse's, even though no communication has taken place
between them. If, for example, a man observes a tree
falling in the forest and tells his wife, then it is
immediately obvious that the tree did not fall, because
his wife's opinion must collapse to the negative upon the
revelation of her husband's informing her that it did.

Incidentally, this is why lesbian marriages should be
impossible; they violate the laws of physics in that one
partner must be wrong.

As we know, a "double-double" is an order at Tim's in Canada; your (usually) large coffee comes with 2 cream, 2 sugar already added.

Combined with a simple dyslexic misreading of slit as 'silt', you have the image of a mad physicist peering into the cup, watching the Brownian motion, refusing to observe the wavicles in the second set of slits, contemplating what might be found at the bottom of the cup when the experiment is over.

Tasty method of refusing to observe wavicles...unless the undulating coffee freezes in place as you take the Chi-Cheemaun to Manitoulin.

Please note that the social atmosphere aboard any public conveyance can be considered to be a vacuum for the purposes of this experiment.

//One thing that I still don't understand. Is it true
that with a single photon you can't determine if it's
behaving as particle or wave? .. because to have
confirmation of a wave you need interference
pattern which means a large amount of photons?//

No.

First, you _can_ generate interference patterns
with single photons. If you set up the double-slit
experiment and fire one single photon through it,
that photon will be detected hitting the screen at
some point. You come back next week and fire
another photon through it, and again it hits the
screen at some point. If you do this over many
many weeks, you will notice that the points where
the photons hit make an interference pattern -
bands of very few photon hits, and bands of many
photon hits.

The photon behaves like a wave when you look for
wavelike properties (like, through two slits). It
behaves like a particle when you look for particle-
like properties (like, when it hits the screen).

If you ask it to behave like a wave, but then set
out to detect a particle, you will therefore detect
a particle that could only have got there by acting
as a wave.

If you put your "passing photon monitors" on the slits, then you are
looking for a particle at the slits, so you will see a particle. That will
abolish the interference effect, because particles can't interfere.

To come back to the original question: after the photon has gone
through one of the slits (and has been detected as it goes by, so it's a
particle), it will _thereafter_ behave just like any other photon. So, if
you had a second double-slit (this time with no passing-photon-
montitors) set up after the first double-slit, you would see interference
in the second (un-monitored) double-slit but not the first (monitored)
double slit.

Feynman was absolutely right when he said that the double-slit
experiment is really the embodiment of quantum mechanics.
Everything else is detail. Truly "understanding" what's happening in the
double-slit experiment is like truly "understanding" a hypercube:
however much you know about it, it's still not remotely graspable.

I belive the fundamental problem is that we still imagine that the photon takes a discrete course rather than that the photon exits in a field state not unlike the electron does, only this field state radiates away from the source until the photon field collapses. Instead of discreet points or pinpricks it is better to imagine that every photon emission is a spherical emission from the source that expands until it is collapsed. In one lightyear the photon field for an individual photon random emission is a sphere two light years across and could, thus be interacted with anywhere in a surface that is more than four square lightyears in size.

//the photon exi[s]ts in a field state not unlike the
electron does//

There's no real difference between a photon and
an electron, but that's not the problem.

The real problem is that the "field" (probability
density function) can be made to collapse into a
particle at one, and only one, point. Suppose I
send a photon out from Earth. Its probability
density function spreads out (not necessarily
evenly, but it spreads) until it's, say, a light year
across.

Now I set up photon detectors in a big sphere
around Earth, with a radius of one light year. One
of those detectors will detect a photon - it will go
click. But only _one_.

Compare this with what we normally think of as
radio waves. A radio station broadcasts a very
weak "beep", and detectors in a 1 light-year sphere
listen for the beep. Either all of the detectors
will hear the "beep", or maybe only a few (if it's
very weak), or maybe none - but it's not as if ONLY
one detector can hear the "beep".

So, photons and electrons do _not_ behave like a
classical field.

No, you see I meant exits, the vocabulary for emission is confusing. I wasn't trying to suggest that fields and photons shared other similarities than the peculiar way in which they propagate across what we measure as distance and time. The photon field is also prone to peculiar distortions in time which other fields are seemingly immune to. We need to stop teaching a vocabulary of the photon fired as a tiny cannonball though because that is clearly slowing how quickly people grasp photonics.

//stop teaching a vocabulary of the photon fired
as a
tiny cannonball//

We don't teach that, do we? Even in my day I don't
think it was taught beyond primary school, if then.
We're generally taught that photons have
"wave/particle duality", which is still not perfect
but
is a better approximation. I presume nowadays
that children are gently introduced to quantum
mechanics in their early teens.

I think that still fails. The photon has features of all
three; particle (mass), wave (frequency, polarity)
and field (non discrete properties). This means that
wave and particle still causes the student to
imagine that each photon wiggles away in a
specific vector.