Investigation of the role of embodiment and action knowledge in 3-D object recognition

本研究旨在探讨具身性和动作知识在三维物体识别中的作用。结果表明：（1）主动探索新的三维物体视图有助于随后的视图概括在右螺旋方向，这是最容易旋转的手。(2)当在非直立视图中识别常见对象时，在可抓握对象的情况下，在围绕水平轴的右螺旋方向上旋转的视图比在左螺旋方向上旋转的视图识别得更快，并且在不可抓握的大对象（例如，G.床，沙发），反之亦然。这表明，运动模拟，即想象旋转的手旋转可抓握的物体，被用来识别非直立的意见。这些结果是第一个实验证据表明，手的生物力学约束和物体的动作知识的影响，三维物体识别。

This study was conducted to clarify the role of embodiment and action knowledge in 3-D object recognition. The following results were obtained :(1) Active exploration of novel 3-D object views facilitated subsequent view generalization in the right-screw direction that was easiest for rotation of the hand.(2) When common objects are recognized in non-upright views, in the case of graspable objects, views rotated in the right-screw direction about the horizontal axis was recognized faster than those rotated in the left-screw direction, and in the case of non-graspable large objects(e. g. bed, sofa), vice versa. This suggests that motor simulation, that is, imagined rotation of graspable objects by hand rotation, was utilized to recognize non-upright views. These results are the first experimental evidences that showed the effect of the biomechanical constraints of hands and action knowledge of objects on 3-D object recognition.