A question about the observer-participancy phenomenon in subatomic physics.

May 30, 2007 1:53 AM

A question about the observer-participancy phenomenon in subatomic physics.

Ok, so basically subatomic particles are in a constant state of limbo, existing in multiple ‘possible’ locations at once, almost like ghosts. It’s only with a human observer that the particle ‘makes up its mind’ and decides where it actually is. This was proved with that experiment, the name of which I forget, in which the beam of light is fed between a slit and are registered on a machine, the name of which I also forget. Anyway, my question in general with regards to the phenomenon is, does a human observer need to be in the room? What if there was a camera set up, recording the meter which registers the direction of the particle… is this a stupid question? If so, how does the particle ‘know’ about the camera, how does it serve as an observer even before it is used as an observational medium, i.e. through the recorded tapes?

posted by heylight to Science & Nature (17 answers total)

There's several ways of looking at what quantum mechanics says. You're pretty much describing the Copenhagen Interpretation which is the most traditional perhaps, but also plagued with unanswered questions like yours.

It's certainly not a stupid question, but it's not clear how one would make an experiment to distinguish between the possibilities.

There are other interpretations which have the same result at the end of the experiment, but with what happens in between being different, especially when it comes to that whole business of 'making up its mind and deciding where it actually is'. The Wikipedia page on interpretations is pretty comprehensive and explains the issues.

In short, it's an unanswered question, perhaps unanswerable, and perhaps not the right question at all.
posted by edd at 2:08 AM on May 30, 2007

When I first read this I immediately thought 'Shrodinger's Cat'. I've always taken this as more of a philosophical conundrum than a purely scientific one. I'm interested to hear what people who know more about such things have to say.
posted by macdara at 2:10 AM on May 30, 2007

What if there was a camera set up, recording the meter which registers the direction of the particle… is this a stupid question? If so, how does the particle ‘know’ about the camera, how does it serve as an observer even before it is used as an observational medium, i.e. through the recorded tapes?

All this stuff confuses me as much as it confuses you, but I'll make these points:

- "Observed" doesn't mean "viewed by a human". I always thought it really means "measured" - after all, these particles can't be seen with our naked eyes anyway, whatever we do with them involves some kind of instrument to "measure" what's going on and report it to us 2nd hand.
- Therefore there isn't a delay between the event happening and it being observed later...it is "observed" by the instrument.
- How does the particle "know"? Herein lies the great question...
posted by Jimbob at 2:48 AM on May 30, 2007

In trying to answer this question, I've gone from "I think I have some understanding" to being confused again. That said, I think Jimbob's on the right track.

As I understand it, it is not the human observer in particular that makes it "choose a path", but the interaction with a photon (which, incidentally, makes it possible to observe it) which does so. In other words, shooting a bunch of high energy photons at the double-slit experiment will cancel the weird interference effect regardless of whether or not you bother to catch the photons and thus figure out which slit the particles passed through. In yet other words, if you set up a situation in which it's possible to observe the effect, it doesn't matter if you actually observe it or not.

Also: as long as it deals with measureable effects, it's certainly a scientific question, not a philosophical one.
posted by alexei at 3:15 AM on May 30, 2007

Jimbob, unfortunately the instrument can be considered a quantum mechanical system. The collapse doesn't necessarily occur there. There's nothing in the measurement process that isn't fundamentally just like what's happening in the experiment.

It's perhaps not Schrodinger's Cat so much as Wigner's Friend.
posted by edd at 3:29 AM on May 30, 2007

We can't come close to understanding quantum mechanics without the math. The words we use cannot capture the meaning of the theory. Even the simplest concepts do not translate properly: for example we typically think of a 'particle' as a 'thing', when perhaps they are not things, but rather 'events'. So asking about the wave function collapse is like asking at what point a car becomes a collision. The question itself muddies the possibility of giving a meaningful answer.


I think the whole 'human observer' concept is a red-herring that is trumped up mainly by ESP gurus and journalists trying to blow your mind (and sell magazines). The theory is unfalsifiable and not parsimonious. Just because decoherence theory isn't perfectly worked out yet doesn't mean we should jump to goofy conclusions (analogous to assuming a UFO must be an alien)


Finally, 'How does the particle know about the camera?' The same way your thermos 'knows' to keep coffee hot and lemonade cold!
posted by Osmanthus at 4:11 AM on May 30, 2007 [2 favorites]

Well, to extend your camera idea:

How would a camera capture a "picture" of the particle? Photons would have to interact with the particle to be observed by the camera, thus measuring the system and collapsing the waveform.

Also, you are thinking of the double-slit experiment. You can actually do it yourself with a good laser pointer and some index cards! It's fun, google it and give it a try.

But yeah, there are two basic ways to interpret quantum mechanics: statistically and probabilistically. The way you are talking about is probabilistically.

I'll expand later, but I'm late right now!
posted by Loto at 5:12 AM on May 30, 2007

Not a stupid question. In fact, I think your experiment has been done, kinda. I remember reading about it in About Time (by Paul Davies), and my memory is more elusive than yours here, but the gist of it was this: Set up the detectors, randomise the data, so that nobody can know about it's quantum state, then to decode the results after the fact.

That may be more wrong than right, but I still remember the take home message: The Heisenberg Uncertainty Principle cannot be beat, no matter how creative we get trying.

It's a great book, btw, written in layman's terms and everything.
posted by kisch mokusch at 5:32 AM on May 30, 2007

@Loto:
How would a camera capture a "picture" of the particle? Photons would have to interact with the particle to be observed by the camera, thus measuring the system and collapsing the waveform.

But wouldn't the photons be interacting with the particle regardless of the presence of the camera (or other observer)?

And the thing with Shrodinger's cat; isn't the cat itself a sufficient observer? Would a person in that situation be a sufficient observer? Or does it only count if the observer is outside the system?
posted by DarkForest at 7:15 AM on May 30, 2007

Unfortunaetly, QM has been hijacked (at least in the popular media) by non-scientific groups using it to push their religious ideologies thus the wavy-gravy rhetoric: "the particles know when youre looking at it," "ghosts" etc. This has happened before everytime something strange is found out by science or the public gets to play with technology. Think 'ghost photography' and hearing voices on staticy tape.

The double-slit is strange but its highly misrepresented. in fact I think its pretty dry and boring. It can be summed up as depending how an experiment is done you get conflicting outcomes. This isnt important because its "ghostly" or because photons "know when you're looking at them," but it is important because it shows the limitations of classical physics.

Also, the wikipedia's write-up on wave-particle duality and double-slit arent bad.
posted by damn dirty ape at 7:20 AM on May 30, 2007

I think edd gets to the heart of it. This particular "paradox" arrises from the chosen interpretation. The Copenhagen Interpretation says that the event is in a superposition of states and is only resolved when it's observed. Various people have proposed definitions as to what constitutes a measurement, but it's a serious problem with the interpretation.

The many-worlds interpretation gets around the problem by positing that all possible quantum states exist in different "worlds."

Are you in a universe where the particle went through the left slit or a universe where it went right? You're in both, and the various "yous" aren't differentiated until you conduct an experiment which reveals the particle's path.
posted by justkevin at 8:02 AM on May 30, 2007

Nice article in the current Believer on the Many-Worlds Interpretation - you can read the beginning here.
posted by nicwolff at 8:17 AM on May 30, 2007

justkevin: Various people have proposed definitions as to what constitutes a measurement, but it's a serious problem with the interpretation.

The many-worlds interpretation gets around the problem...

the various "yous" aren't differentiated until you conduct an experiment

How does MWI get around the problem?

In both interpretations, there's some apparent criteria for a (supposed) state transformation

1)In Copenhagen, it's from superposition to decoherence
2)In MWI, it's the selection of a subset of potential futures and discarding the rest (within the current continuum)
posted by Gyan at 8:30 AM on May 30, 2007

Ok, so basically subatomic particles are in a constant state of limbo, existing in multiple ‘possible’ locations at once, almost like ghosts. It’s only with a human observer that the particle ‘makes up its mind’ and decides where it actually is.

This is a terrible way to think about it, brought about by terrible woo-woo mystical quantum meta physics writing.

In the real world, what happens is averaging. Each particle of billions upon billions of particles falling into one of billions of states. Considered all together you get a random distribution. Anybody who's seen Las Vegas knows that random distributions aren't all that random.

When you manage to isolate a single photon or electron, it stops looking like a distribution, and results start looking funny. Do you say that a coin is in limbo while it is in the air, mid flip? Start flipping a coin, and recording the outcomes.. Very odd patterns can emerge for a while. Once you've flipped the coin 6.02x10^23 times (the number of particles in a mole - almost a million billion billion particles, in American billions, yet not much 'stuff' at all, on a human scale)..

More is Different is worth checking out.
posted by Chuckles at 8:58 AM on May 30, 2007

I don't think it is quite as straightforward as Chuckles makes out. A coin isn't in limbo in mid-air. You can see its movement through the air and its rotation, and all these things contribute to the final way it lands.

A particle in quantum mechanics doesn't act like a coin for which you simply can't see the way it is moving - it can't have properties like that associated with them without doing other funny things. Such classical but unobserved behaviour would be a 'hidden variable', and if you want to explain quantum behaviour in that way you have to sacrifice locality. This is not something to be done lightly.

Something very odd is going on down at those scales.
posted by edd at 9:17 AM on May 30, 2007

Here is an interesting link.
posted by Caper's Ghost at 10:50 AM on May 30, 2007

Chuckles is thinking of Quantum Statistical Mechanics, and in that sense he is mostly correct. When you have 10^23 particles, things just don't look as interesting.

Darkhorse: Welcome to the joy that is quantifying what constitutes a measurement. This issue is dealt with by quantum decoherence.
posted by Loto at 2:19 PM on May 30, 2007