If two people perform the double slit experiment, why does the pattern disappear for both if only one observes which slit the electron goes through?

Question

In the double slit experiment, one can observe which slit the electron goes through with a flash of light as the electron is scattered by a light source behind the screen.

Whenever we know which slit the electrons go though, we see no interference pattern.

But, if these measurements are done secretly by one person, and the second has no knowledge even that the second person exists (but can see that there is a light source, they may just no have the detector to see the flash themselves), why does the interference pattern disappear for both, rather than just the active observer?

In theory, even when both people don't actively detect the flash of light that gives away which slit the electron passes through, there are still ways the electron gives away its position, its just both people don't work to measure it. So it is "always" being measured in some sense (perhaps via very slight gravity effects we don't record).

Answer 1

What a person knows or doesn’t know is not a factor in the appearance or not of interference. The rule is: if it is possible in principle to derive the which path information from the setup, there will be no interference. Nobody actually needs to know anything. This has experimental support, see for example:

https://sciencedemonstrations.fas.harvard.edu/files/science-demonstrations/files/single_photon_paper.pdf

”An apparatus for a double-slit interference experiment in the single-photon regime is described. The apparatus includes a which-path marker that destroys the interference as well as a quantum eraser that restores it. We present data taken with several light sources, coherent and incoherent and discuss the efficacy of these as sources of single Photons.”

In the experiment, the which path information is encoded (or not) using polarization. There is no need to actually decode it to see the expected effect.

Answer 2

What happens is well captured by the delayed choice quantum eraser. This is an even more brutal version of the scenario you describe, where the decision to observe or not observe path information occurs after the observation at the screen.

The result is very paradoxical, unless you properly account for quantum effects, but the part you care about is surprisingly natural. Remember that a single electron does not make an interference pattern. It makes a dot. You need to repeat the experiment many times with many electrons to build up the interference pattern. And when you do, you see no particularly interesting results. In particular you see no interference fringes.

However, the delayed choice quantum eraser has a behavior similar to the second observer observing in "secret." The second half of the quantum eraser, or the second observer, can provide path information. If one uses that to select only electrons that went through the left slit, and plot only the strikes on the screen from those electrons, you see the single band you would expect from a single slit. If one selects electrons which don't provide path information, the resulting dots form interference patterns.

Why does this happen? Well... because it does. Nature does what it does. We formulated quantum mechanics to properly predict these behaviors. And if you run the numbers, you indeed see the results we see in the eraser.

Now all of this is for the delayed choice quantum eraser, not a conscious observer observing "in secret." I find this to be a useful way to approach the problem because the delayed choice quantum eraser is real. We've made them, they work. A "observer observing in secret" is a bit less clear. It turns out to be rather easy to come up with an "observer" that observes something that is intuitively easy to observe but turns out to be non-observable in the real world. Some things in quantum mechanics are observable and others simply aren't. What is observable and what isn't isn't intuitive until one is comfortable with QM, at which point one isn't asking this question =) So I like this experiment because we know for a fact that it doesn't do anything illegal -- we've executed the experiment and found the results.

Answer 3

Consider the following related thought experiment. We have an interferometer with a polarizing filter in each path. Assume when the filters are orthogonal to each other we can potentially deduce which path an individual photon took. Each polarising filter is in a sealed black box (sealed except for a path in and out for the photon). Attached to each filter is a mechanism to randomly alter the orientation of the filter once per day. E.g. the orientation can be based on the decay of a radioactive substance. There is no communication between the two filters and no recording devices to record the positions of the filters on a given day. We can be pretty sure on some days we will consistently see an interference pattern and on other days we will not. No sentient being in the universe knows the positions of the filters on a given day and there is not even a computer or some such device that 'knows' the relative orientations of the filters on a given day, but still the interreference pattern disappears on some days. Now unless the universe has decided that on e.g. Mondays it will have a non interference day, we can conclude that the breakdown of the interference is not due to knowledge of some being, sentient or otherwise. It is simply that any potential measurement device in the path of the particle disturbs the trajectory, wavelength or polarisation etc of the particle and this can cause the breakdown of the interference pattern (as Bill Alsept mentioned in the comments).

It is also worth noting that the early universe evolved according to the rules of quantum physics, long before any sentient beings or observers evolved. they are not required. An observation is simply an interaction of a quantum particle with other particles. The rule is that if an observer could potentially deduce the path of a quantum particle, then the interference pattern breaks down, even if there is no actual sentient observer "knowing" the path.

Another example of an experimental setup is to consider the mirrors that are conventionally placed at the corners of an interferometer to redirect the photons. If these mirrors are made small enough and mounted in such a way that they can move, then when a photon bounces off the mirror, the mirror recoils. The recoil of the mirror alters the wavelength and trajectory of the photon and the interference pattern can break down. This is "because" the recoil of the mirror could potentially be used to detect which way the photon went, even if no sentient being makes that observation or conclusion.

In conclusion, any "measurement" requires something physical to interact with the quantum particle being "observed" and in so doing can potentially alter its path or nature (eg polarisation) which can lead to breakdown of the interference pattern. Any pop-science books or videos that claim a conscious or sentient being is required for quantum effects to take place are simply nonsense.

EDIT: To more directly address your particular thought experiment, consider that to see an electron pass through a slit, photons have to bounce off it ang go to a detector. These photons bouncing off the electron alter its momentum. We can reduce the alteration of the the electrons momentum by reducing the intensity of the light source (amount of photons). Now imagine the detectors are CCD chips. When there is absolutely no illumination, individual pixels of the CCD light up randomly due to quantum noise, so it is difficult to tell if a single pixel lighting up is random noise or an actual detection of an electron. We can reduce the uncertainty by increasing the intensity or frequency of the illumination but in so doing, we progressively increase the deflection of the electrons momentum in turn increasingly breaking down the interference pattern. This is basically the Heisenberg uncertainty principle at work.