Safe and effective arthroscopic surgery requires ambidextrous motor skills. The current study examined dominant versus non-dominant hand performance on a virtual reality serious game in a group of expert arthroscopic surgeons (n=15) compared to a group of orthopedic surgery residents (n=10). A virtual reality Tetris game was performed with the arthroscopic camera and arthroscope in one hand, using an arthroscopic grasping tool in the opposite hand to manipulate the virtual Tetris blocks onto the game grid. A second run was performed after swapping instruments between hands. The order of hand testing was randomized. There was no statistically significant difference in exercise time, grasper path length, or camera path length between the right and left hands of the expert surgeons. In contrast, there were statistically significant differences in all of these parameters between the two hands for the orthopedic surgery residents, with better performance when the grasping tool was used in the dominant hand. The findings of this study suggest that virtual reality games which incorporate progressive cognitive loading could be used to facilitate training, automation, and objective assessment of surgical motor skills.