The uncertainty principle is certainly one of the most famous and important
aspects of quantum mechanics: Heisenberg first suggested a limitation of
joint measurements of canonically conjugate variables due to the back action
of measurement. Nevertheless, the prediction, that the product of
the measurement error and disturbance caused by the
measurement is not less than a bound set by the commutator,
is justified only under limited circumstances. Recently a universally valid relation
between the error and the disturbance has been derived by Ozawa[1].
In my talk,
a neutron optical experiment is reported that measures the error of a spin-component
measurement and the disturbance caused on another spin-component
measurement. The experimental results exhibit that the error and the
disturbance completely obey the new relation but violated the old one in a
wide range of experimental parameters [2]. This experiment stimulates
further measurements, for instance, using photon's polarization [3].
The solution of a long-standing problem to describe the relation between the
measurement accuracy and the disturbance caused by that measurement is adressed.
[1] M. Ozawa, Ann. Phys. 311, 350 (2004).
[2] J. Erhart et al., Nature Phys. 8, 185 (2012).
[3] L.A. Rozema et al., Phys. Rev. Lett. 109, 100404 (2012).