Observation of non-classical correlations in sequential measurements of photon polarization

TL;DR

This paper demonstrates that sequential measurements of photon polarization reveal non-classical correlations, described by an imaginary joint probability distribution, across different measurement strengths, highlighting quantum measurement back-action and error correlations.

Contribution

It introduces a method to describe error statistics in sequential quantum measurements using an imaginary correlation, validated through experimental realization with variable measurement strengths.

Findings

Imaginary correlation describes error statistics in sequential measurements.

Complex joint probability distribution remains consistent across measurement strengths.

Experimental setup successfully characterizes polarization states with controlled measurement back-action.

Abstract

A sequential measurement of two non-commuting quantum observables results in a joint probability distribution for all output combinations that can be explained in terms of an initial joint quasi-probability of the non-commuting observables, modified by the resolution errors and back-action of the initial measurement. Here, we show that the error statistics of a sequential measurement of photon polarization performed at different measurement strengths can be described consistently by an imaginary correlation between the statistics of resolution and back-action. The experimental setup was designed to realize variable strength measurements with well-controlled imaginary correlation between the statistical errors caused by the initial measurement of diagonal polarizations, followed by a precise measurement of the horizontal/vertical polarization. We perform the experimental characterization of an elliptically polarized input state and show that the same complex joint probability distribution is obtained at any measurement strength.