||This paper investigates the variation of root-mean-square values of acceleration responses and corresponding aerodynamic dampings of two square cross-sectional high-rise buildings with low Scruton numbers due to interference effects caused by neighboring identical buildings. Aero-elastic vibration test is adopted for displacement and acceleration measurements of a rigid, base-pivoted aero-elastic principal building model by a two-directional sway gimbal system. The principal building is manufactured in the shape of a square prism model with an aspect ratio of eight. Twenty close interference locations are of interest, with a special focus on the across-wind responses. Results show that the variations of interfered acceleration responses and estimated damping ratios exhibit different tendencies for different interference locations at different reduced velocities. For the typical upstream interfered effect, the interfering building located at the oblique-upwind locations in this study, the acceleration response with lower Scruton number is suppressed at lower reduced velocities and is amplified at higher reduced velocities. On the other hand, for the secondary downstream interference effect, i.e., the interfering building located at downwind locations, the amplification or reduction in acceleration response is less sensitive to Scruton number but is related to the gap between two buildings. The corresponding response spectra and resultant damping ratios are then calculated to enhance the understanding of different interference mechanisms. Finally, an idealized two-dimensional CFD simulation technique is adopted to further explain the differences between the upstream and downstream interference mechanisms from the viewpoint of the vortex movement between two buildings.