In this paper, we investigate the independence between perception and action that has been reported in the previous psychophysical studies. The independence, however, might not be directly investigated because illusion tasks, such as the Ebbinghaus illusion, are used in those studies, although those findings are quite attractive. Thus, the independence may include specific characteristics in the illusion tasks. From this point of view, we focuses on grasping positions when grasping an object, such as a tool, which is used in our daily life. In the first experiment, we investigate the independence of the grasping positions by using the following measurement tasks: 1) a visual-estimation task that the grasping positions are visually estimated, 2) a pinch task that only grasps an object without lifting-up its grasped object and 3) a lift-up task that grasps an object and lifts it up. As a result, both the grasping positions of the visual-estimation and lift-up tasks are significantly different. Thus, these results indicate the independence of perception and action for grasping movement in daily life. In addition, those of the pinch and lift-up tasks are significantly different amazingly, although both the tasks can be considered as an action task. In the second experiment, for the pinch and lift-up tasks, we examine the difference of both the trajectories of the finger tip and moreover the influence of visual feedback for the grasping positions. As a result, we confirmed that the above results are not affected by the visual information perceived from own hand and arm during the movement. Moreover, these results indicate that grasping positions are determined before movement because both the trajectories are different just after movement onset. Finally, our findings are quit attractive because the difference of the grasping positions may be explained based on the Goodale΄s hypothesis (˵how system″ and ˵what system″). [ABSTRACT FROM AUTHOR]