This study examines the facilitative effects of embodiment of a complex internal anatomical structure through three-dimensional (3-D) interactivity in a virtual reality (VR) program. Since Shepard and Metzlers influential 1971 study, it has been known that 3-D objects (e.g., multiple-armed cube or external body parts) are visually and motorically embodied in our minds. Such findings confirm the theory that our mental images, and rotations of these images, are in fact confined by the laws of physics and biomechanics, because we perceive, think and reason in an embodied fashion. With the advancement of new technologies, virtual reality programs for medical education now enable users to interact directly in a 3-D environment with internal anatomical structures. Given that such structures are not readily viewable to users and thus not previously susceptible to embodiment, coupled with the VR environment affording all possible degrees of rotation, how people learn from these programs raises new questions. If we embody external anatomical parts we can see, such as our hands and feet, can we embody internal anatomical parts we cannot see? Does manipulating the anatomical part in virtual space facilitate the users embodiment of that structure and therefore the ability to visualize the structure mentally? Medical students grouped in yoked-pairs were tasked with mastering the spatial configuration of an internal anatomical structure; only one group was allowed to manipulate the images of this anatomical structure in a 3-D VR environment, whereas the other group could only view the manipulation. The manipulation group outperformed the visual group, suggesting that the interactivity that took place among the manipulation group promoted visual and motoric embodiment, which in turn enhanced learning. Moreover, when accounting for spatial ability, it was found that manipulation benefits students with low spatial ability more than students with high spatial ability.