The mediolateral balance assessment method (MELBA) consists of tracking a sinusoidal or multisine target with the center of pressure as feedback (CoPfb). The aim of the CoP trajectory is to elicit weight-shifting, i.e. movement of the center of mass (CoM). However, it is not known whether CoPfb elicits consistent mediolateral displacements of the CoM, whether CoM feedback (CoMfb) is required to achieve this and whether CoPfb or CoMfb elicit different kinematic strategies. The aims of this study were to determine (1) the extent to which CoP imposes CoM displacements (CoMd) during CoPfb, (2) whether larger CoMd are elicited by CoMfb and (3) whether different kinematic strategies arise when using CoPfb or CoMfb. Nineteen young adults performed MELBA with CoPfb and CoMfb from which coherence, gain and phase-shift between CoP-CoM and leg-trunk kinematics were calculated. CoMd and CoPd and leg and trunk excursions were also calculated. Results show that for CoPfb tasks, CoP-CoM coherence was high, while the gain dropped with increasing frequency. The drop in gain was highly consistent between subjects. Reasonable trunk-leg coherence (≈.6) was found over all frequencies and tasks. The leg-trunk angle gain increased with frequency in all tasks and was significantly higher in the CoMfb compared to the CoPfb. Significant interaction indicated that this difference increased with frequency. CoPfb in MELBA elicits consistent CoMd. However, different kinematics are employed in CoMfb with more trunk movement and an ankle-to-hip shift as frequency increases. Hence CoMfb may be preferable over CoPfb despite the larger measurement effort involved.