Abstract: Central-slotting box girders are widely employed in long-span bridges due to its advantageous flutter stability. However, the existence of the central slot would potentially worsen the performance of vortex-induced vibrations (VIVs). Aiming at a typical central-slotting box girder of a long-span bridge, wind tunnel tests of the pressure distribution and displacement response of a spring-suspended sectional model were conducted. The surface pressure distributions in the whole amplitude-developing lock-in period during the torsional VIV process were measured, including the beginning-VIV point, midpoint of the ascent stage, amplitude extreme point, midpoint of the descent stage, and ending-VIV point. It is concluded that the aerodynamic effects of the model have obvious evolutionary characteristics in series of VIV developing stages, the aerodynamic characteristics and the VIV response during the VIV process are strongly correlated with each other and even synergistic. The contribution of distributed aerodynamic forces to the general vortex-excited force (VEF) are positively correlated with the VIV amplitude, both reach the maximum at the amplitude extreme point. The distributed aerodynamic force along the rear region of the upper and lower surfaces of the downstream separated box and the front region of upper surface of the upstream box has a greater contribution to the general VEF, though contributing positively and negatively to the VEF. These areas are the main exciting source of the VIVs of central-slotting box girders. Compared with a closed-box girder, distributed aerodynamic forces in the downstream region of upper and lower surfaces usually weaken the contribution to the VEF from the viewpoint of correlation, while central slotting keeps the contribution values in the upper and lower surfaces of the downstream region positive and enhances the general VEF. Therefore, the VIV effects of central-slotting box girders are usually stronger than those of closed-box girders.