Shared randomness allows violation of macroscopic realism using a single measurement

Abstract: Macro-realistic description of systems is based majorly on two basic intuitions about the classical world, namely, macrorealism per se, that is, the system is always in a distinct state, and non-invasive measurements, that is, measurements do not disturb the system. Given the assumption of no-signalling in time, one utilizes Leggett-Garg inequalities to observe a violation of macroscopic realism which requires at least three measurements. In this work, we show that if one has access to shared randomness then one can observe a violation of macroscopic realism using a single measurement even if no signalling in time is satisfied. Interestingly, using the proposed scheme one can also rule out a larger class of models, which we term "macroscopic no-signalling" theories which can not violate the no-signalling in time conditions. We further construct a witness to observe the violation of macroscopic no-signalling.