So first off, Ill admit Ive forgotten most of the finer details of the bell inequality so I apologize if Ive gotten something wrong. From how I remember it, the bell inequality is a test of any local hidden variable theory vs. QM. It uses the fact that the inequality measured if a classical, local hidden variable theory were true is necessarily different than the inequality given by QM. As far as I know, the nature of decoherence is somewhat of a mystery. Personally, I like Everett's interpretation that decoherence is just an illusion caused by the entanglement of a macroscopic object (the observer) with the measured quantum object. I was thinking of this and the other theories (Penrose's specifically which I really dislike) and it seems to me like these are more than just interpretations. Although MWI is clearly an interpretation, Everett's view of decoherence should be testable. So heres the experiment I thought of that should test it: Put a photon into an equal superposition between up and down spin (or polarization w/e). Have a detector in a vacuum (space) measure this photon. Then have the detector shoot a new photon with identical spin to the measured photon, alone with a tennis ball with some property determined by the measured spin (e.g. angular momentum, speed). Now do the Bell inequality test on the photon and the tennis ball (repeated many times of course). The detectors must be isolated enough from the first detector that no information about the projectiles can reach them (or else the systems wave function would collapse). If a macroscopic object is capable of entangling with a quantum object, the measured inequality should be that predicted by QM. However if some theory like penrose's is right, the tennis ball would not be capable of entangling with the photon and the measured inequality would be that of a hidden variably theory. Did I miss something or is this actually testable?