Medical learning and training are fields in transition. Catalyst in this change is the introduction of digital technology, for example in the form of simulator technology in surgical training, and virtual learning environments in anatomical learning. The primary aim of this thesis is to help understand and optimize this new situation. More specifically, the mediating role of cognitive abilities such as visuo-spatial ability in this learning and training is investigated. Visuo-spatial ability refers to the human cognitive ability to form, retrieve, and manipulate mental models of a visual and spatial nature The first three empirical chapters report experiments investigating the effects of stereopsis, dynamic exploration, and visuo-spatial ability on virtual anatomical learning. Stereopsis refers to the visual experience of space derived from slightly different patterns of light being simultaneously received by the eyes. One can dynamically explore virtual objects if it is possible to interactively change one's viewpoint with respect to such objects in real time. In virtual anatomical learning environments, people of low visuo-spatial ability benefit from a combination of computer-implemented stereopsis and dynamic exploration. The last two experimental chapters investigate how the development of endoscopic skill correlates with the major five individual factors that together form visuo-spatial ability. Endoscopy refers to those surgical techniques that are characterized by the use of a small camera introduced to the body to provide visual feedback, in order to minimize surgery related trauma. The visuo-spatial factor of Visualization is more important to endoscopic than other visuo-spatial factors. Visualization is characterized by a high sensitivity for visuo-spatial complexity. More work is needed to understand visuo-spatial complexity in this context (or, indeed, any context).