4 ms·

A single photon will register one click on the detector, it’s only with an ensemble of many photons that an interference pattern emerges.

roywiggins 8y ago
You will build up an interference pattern if you fire single photons, one at a time, spread out over as long a time period as you like.
- SAI_Peregrinus 8y ago
  You can even do it with electrons. It's easier to fire single electrons, so that tends to be what's done for such an experiment.
  - goldenkey 8y ago
    Everything. Isolation/coherency, of course, becomes harder with larger. https://medium.com/the-physics-arxiv-blog/physicists-smash-record-for-wave-particle-duality-462c39db8e7b https://medium.com/the-physics-arxiv-blog/physicists-smash-r...
- axilmar 8y ago
  Since the interference pattern is built over time, how do we know a particle is actually a wave that interferes with itself and not a particle that selects a different path from a set each time it is fired?
  - ben_w 8y ago
    If it was acting like a particle and selecting a different path each time, the expectation would be only two peaks, each following line-of-sight to the source: —————|—|————— What is seen is many peaks, as if it was a wave: —+—+—|—|—+—+—
    - TheOtherHobbes 8y ago
      That doesn't solve the problem, because the question is really how the set of paths available to the particle is constrained. If you go Full Copenhagen, there are no waves, no particles, and no paths. There are only evolving probabilities - which act in unintuitively non-local but predictable ways - and events which appear to sample one possible result from the current state of a probability density. The probabilities can be composite, which allows for entanglement and superposition. The probability part is fairly well understood, or at least fairly easy to calculate. The exact nature of an event/measurement/whatever is still a complete mystery. But there is nothing in a naive wave/particle/path model that makes it any less mysterious or easier to understand.
  - roywiggins 8y ago
    There are lots of ways to explain the interference pattern. Probably the most famous "renegade" theory is Bohmian mechanics, which suggests that there's a real particle that is guided by a wave that we can't see. The particle doesn't interfere with itself, but it's guided by a wave that does. I don't think the particle can select a path from a set at the time it's fired. That sounds very much like a local hidden variable, which is more or less strictly ruled out by violations of Bell's inequality. Unless you allow causality to go back in time (in which case, it can make its decision on where to land based on, for instance, how it will eventually be measured). You can do things to the particle (like trying to measure it, or entangling it with other particles and measuring them, etc) which make it seem less and less likely that the particle could have selected a "which-way" as it was fired. For instance: suppose you put a detector on one of the slits as the particles are "in flight". We know this results in no 'interference pattern' observed. How do the particles know to scramble themselves as soon as the detector is put on? It doesn't seem like a local effect, since all the particles that go through the other slit never interacted with the detector at all and therefore shouldn't have been affected. But they are anyway. So either there's some information being passed back in time along the particles' history, or there's some strange nonlocal effect.

roywiggins 8y ago

You will build up an interference pattern if you fire single photons, one at a time, spread out over as long a time period as you like.

SAI_Peregrinus 8y ago
You can even do it with electrons. It's easier to fire single electrons, so that tends to be what's done for such an experiment.
- goldenkey 8y ago
  Everything. Isolation/coherency, of course, becomes harder with larger. https://medium.com/the-physics-arxiv-blog/physicists-smash-record-for-wave-particle-duality-462c39db8e7b https://medium.com/the-physics-arxiv-blog/physicists-smash-r...
axilmar 8y ago
Since the interference pattern is built over time, how do we know a particle is actually a wave that interferes with itself and not a particle that selects a different path from a set each time it is fired?
- ben_w 8y ago
  If it was acting like a particle and selecting a different path each time, the expectation would be only two peaks, each following line-of-sight to the source: —————|—|————— What is seen is many peaks, as if it was a wave: —+—+—|—|—+—+—
  - TheOtherHobbes 8y ago
    That doesn't solve the problem, because the question is really how the set of paths available to the particle is constrained. If you go Full Copenhagen, there are no waves, no particles, and no paths. There are only evolving probabilities - which act in unintuitively non-local but predictable ways - and events which appear to sample one possible result from the current state of a probability density. The probabilities can be composite, which allows for entanglement and superposition. The probability part is fairly well understood, or at least fairly easy to calculate. The exact nature of an event/measurement/whatever is still a complete mystery. But there is nothing in a naive wave/particle/path model that makes it any less mysterious or easier to understand.
- roywiggins 8y ago
  There are lots of ways to explain the interference pattern. Probably the most famous "renegade" theory is Bohmian mechanics, which suggests that there's a real particle that is guided by a wave that we can't see. The particle doesn't interfere with itself, but it's guided by a wave that does. I don't think the particle can select a path from a set at the time it's fired. That sounds very much like a local hidden variable, which is more or less strictly ruled out by violations of Bell's inequality. Unless you allow causality to go back in time (in which case, it can make its decision on where to land based on, for instance, how it will eventually be measured). You can do things to the particle (like trying to measure it, or entangling it with other particles and measuring them, etc) which make it seem less and less likely that the particle could have selected a "which-way" as it was fired. For instance: suppose you put a detector on one of the slits as the particles are "in flight". We know this results in no 'interference pattern' observed. How do the particles know to scramble themselves as soon as the detector is put on? It doesn't seem like a local effect, since all the particles that go through the other slit never interacted with the detector at all and therefore shouldn't have been affected. But they are anyway. So either there's some information being passed back in time along the particles' history, or there's some strange nonlocal effect.