Are motor images based on kinestheticvisual matching?