Hacker News new | ask | show | jobs
by bhattid 1252 days ago
What you're describing is called the "observer effect", which is different from the "measurement effect" that's used to describe the quantum mechanics problem. The misunderstanding is understandable though, because it's difficult to properly explain why 'observation' in quantum mechanics is so weird. What constitutes observation is a bit controversial, but you can more or less interpret it as taking a measurement - measuring voltage with a voltmeter, looking at something with your eyes, touching something, etc.

I feel like Schrodinger's cat is used as an example a lot for this, but imo it's a bad example because it doesn't properly distinguish between our classical intuition (the cat is either alive or dead) and the quantum interpretation (the cat is in a superposition between being alive and dead until observed). If I recall correctly, when Schrodinger originally proposed the thought-experiment, it was more of a jab against quantum theory, since the concept of a cat being in a superposition of being alive and dead sounds nonsensical (and probably is, since most would agree that a cat, or any conscious entity, measures things constantly).

Also, in case it's not clear, saying an object is in a superposition between X or Y does not mean that the object is either in a state of X or Y. I don't think there's an intuitive way to describe it without referencing some math. If you've taken some linear algebra, imagine that X and Y are linearly independent vectors in a vector space. Then classical mechanics says that an object can either be in state X or state Y. Quantum mechanics says that the object can be in X, Y, or a linear combination of the two vectors.

To work with something concrete, let's say that our object is an electron and X is spin-up and Y is spin-down (disclaimer: spin is bad name since they don't correspond physically to something spinning). I'm hoping this might be familiar to you since you like electronics, but let's just say that we've created a context where these are the only two states the electron is ever observed in.

In the classical interpretation, the electron is only ever in a spin-up or spin-down position, regardless of whether we're observing it or not. In the quantum interpretation, it's possible for the electron to be in a superposition of spin-up and spin-down when we're not observing it, and when we observe it, it "collapses" into either spin-up or spin-down. Put this way, it sounds like cheating; quantum mechanics is saying we can only observe spin-up or spin-down anyway, so what's the difference! Well, fortunately, there ARE experiments that can distinguish between the classical and quantum based on what they're doing 'behind the scenes' when we're not observing them.

Imagine now that we have photons of light. Instead of spin-down and spin-up, these photons are either horizontally polarized or vertically polarized. The experiment I'm about to explain would also work for the electron example above, but I'm only switching to photons since I know experiments for this have been performed (https://en.wikipedia.org/w/index.php?title=GHZ_experiment&ol...).

Suppose that we've entangled three photons of light together. If you're unfamiliar with entanglement, it just means that we've produced the photons in such a way that they're either all horizontally polarized or all vertically polarized. We can confirm this by using a horizontal polarizer (or vertical polarizer if you prefer). Whenever we shoot the horizontal polarizer with the three photons, they either all go through or none of them go through. Maybe we switch the horizontal polarizer with a vertical polarizer just to be sure, and indeed, we observe the exact same thing happen. Right now, the classical and quantum interpretations agree that this is what we should observe.

Now let's do something that sounds a bit silly. Horizontal and vertical polarization aren't absolute things, they're relative. What this means is that we're testing for polarization at angles, say 0 degrees and 90 degrees. This also means we can rotate our polarizer to a 45 degree angle.

Just for fun, let's say we shoot our three polarized photons through the polarizer which is now at a 45 degree angle. If you're thinking classically, you might think that maybe all will go through or all won't go through. Maybe some will go through sometimes and others will go through other times (probabilistic).

The standard classical interpretation says that you'll observe either: 1. All three photons go through. 2. None of the photons go through.

This is where the classical and quantum disagree. The quantum interpretation also says you'll observe one of two scenarios as well, but those scenarios are: 1. Two photons will pass through, one photon will not 2. One photon will pass through, two photons will not

And lo and behold, experiments show (within experimental error) that the quantum interpretation is correct!

There's still plenty of room for disagreement. Maybe you or someone might argue that the photons are interacting with each other or something funny is going on with the polarizer in question. However, we still observe results aligned with the quantum interpretation regardless if we use different polarizers for each photon, have them sent on a delay, or so on (although, I don't know how many variations have been tested by others for this specific experiment).

Hopefully, I haven't been much of a bore, or wasn't overly confusing. :)

There are ways to "save" classical mechanics using non-local hidden variables and other fancy things, but (if you can take my word for it) at that point, classical mechanics starts losing its intuition anyway. I'm not very knowledgeable about these alternate theories of classical mechanics, but my impression is that they don't make strong predictions, which I'm guessing is why quantum mechanics is more heavily favored.

If you're interested in reading more on the topic, an experiment related to Bell's Inequality was a major piece of evidence in favor of the quantum model. It's similar to the GHZ experiment I described, but simpler. The tradeoff is that its predicted result is inherently probabilistic.