The present study attempted to investigate the effect of position on the perception of angle of trunk flexion while standing. For this purpose, the range effect was factored out by setting the constant target angle at 10 degrees, with varied starting positions of trunk flexion. We found that subjects underestimated angle of trunk flexion when the starting position was close to a quiet standing posture, overestimated when close to maximum trunk flexion, and correctly perceived it when at the middle position. Less perceptual distortion was observed at the positions close to maximum trunk flexion in the present study than in our previous one, in which various target angles of trunk flexion were reproduced from a quiet standing posture. The reduced distortion in the present study was believed to have resulted from factoring out the range effect. The flexion angle of the hip joint changed in tandem with that of the trunk, while very little movement was observed in the ankle, knee, and neck joints. Judging from the changing pattern of hip-joint angle, the muscle activity of the erector spinae and biceps femoris increased gradually to 90 degrees trunk flexion. In contrast, the actual increment of muscle activity reached zero or a minimum value at the middle angles as the angle of trunk flexion increased. It was assumed that the abrupt change in kinesthetic information associated with muscle activity exerted a great influence on the perception of trunk flexion.