Understanding the complex temporal and spatial correlations of ions in disordered perovskite oxides is critical to rationalize their functional properties. Here, we provide insights into the longstanding controversy regarding the off-centering of transition metal (TM) ions in the archetypal ferroelectric alloy KTa[Formula: see text]Nb[Formula: see text]O[Formula: see text] (KTN). By mapping the full energy ([Formula: see text]) and wavevector ([Formula: see text]) dependence of the dynamical structure factor [Formula: see text] using neutron scattering, and rationalizing our observations with atomistic simulations leveraging machine learning, we fully resolve the static