In the Multiple World Interpretation, as I understand it, quantum effects can be caused by multiple worlds interfering with each other. I'm wondering, do all worlds in MWI interfere with each other always? Or do they become unable to interfere because of something, or perhaps in general the interference averages out to nothing measurable. Another thread was just started asking about Interferometer Outcomes Interferometer Outcomes with the following image about Mach-Zehnder Interferometers. This inspired me to ask a MWI question about this which I have been thinking about for a while. Here, in case (a) and (b), in MWI interpretation, in a simplified account, we could say that a photo hits the beam splitter, and goes one way in one world, and another way in a second world. Later the beam is recombined, and the interference of the words causes the photo to always be detected at detector (B) and never (A), consistent with QM. I am wondering what happens in the case if you replace the first beam splitter with a mirror, which can be moved out of position, in a mechanical way in the macro world. However, the decision to move the mirror is caused by a single random quantum event. Now suppose we fire one photo at the mirror, and we time things so that before the photon reaches the mirror, there is a 50% chance that the mirror will have been removed based a single quantum event. I am wondering what would be the expected result in this case. To make things simple, lets please ignore any interference from outside sources that does not include the experiment it's self and the apparatus. First of all, I am not 100% sure what QM predicts, but, my first instinct is that since the mirror is macro, interference would not be observed. In repeated experiments, detector (A) would detect some photos. To make it clearer, lets say that we also take note of whether the mirror was there or not when we detect the photon, this way, I am sure QM would predict no interference, due to which way information, and the photon would have a chase of being detected at (A) Now, MWI, needs to make the same prediction as QM, but, in the simplified account, we could make the same claims as in the first experiment. There are 2 worlds, one with a mirror, one with no mirror, if they interfere, the photon should only be able to appear at (B), disagreeing with QM. I am going to assume these worlds should not interfere in a way that causes MWI to disagee with QM. The question is, why?