The study examined belt effects on the change of lumbosacral angle (LSA) and back muscle activity in postures of standing, erect sitting, and slump sitting. We thought that the resulting changes of LSA and back muscle activity when wearing belts with different mechanical characteristics should be different. Eighteen healthy male subjects participated in this study. Though we failed to identify a significant belt effect on the back muscle EMG, the radiographic data revealed an interactive effect of postures and belts on the change of LSA. In standing, the belts increased LSA by increasing almost every lumbar vertebral angle. In erect sitting, the lumbar belt had no effect but the pelvic belt decreased LSA through a decrease in the L1/L3. While sitting slump with a trunk flexion of 15 degrees, both belts increased LSA by restricting the movement of the pelvis. Belt effect on LSA was accompanied with a change of pelvic angle. Significant correlation was found between the backward rotation angles of the pelvis and the angles of LSA (r = 0.692, p < 0.0001), also between the decrease of pelvic angles and the increase of back muscle EMG (r = -0.4, p = 0.017). A change in LSA and pelvic angle after wearing a belt along with posture change seems lead to an increase of the myoelectric activities on the back.