According to University of Toronto psychologists, Playing Video Games Reduces Gender Differences In Spatial Cognition. According to Feng, Spence, and Pratt's paper, the experimental population played Medal of Honor: Pacific Assault. The control population played Ballance. The tests were useful field of view (by Ball and Roenker) and mental rotation. The experimental population improved in spatial cognition; most notably female participants improved more, approaching the levels of performance of the males. Ian Spence links much press and discussion.
Of educational interest, spatial cognition correlates to performance in science and math classes. The benefits for experimental population, who had little or no recent FPS experience, was achieved after 10 hours of play. I personally suspect that there is a rapidly diminishing margin of return on additional hours of play that correlate to educationally-applicable improvements to spatial cognition. There is a paradigm shift in visual cognition when one adapts to a first-person shooter, in part due to decoupled vestibular and visual feedback.
A problem I personally have is a mild form of FPS-induced motion sickness. When the camera rotates from left to right, my visual stimulus indicates the environment (or my head) is spinning. But my vestibular system (in my inner ear) indicates that my head is not rotating. This conflict induces disorientation and mild nausea. This problem is exacerbated by a large TV screen and a dark room. The problem doesn't happen when the camera pans, tracks, or zooms. The problem also doesn't happen during a third-person game with a mobile camera. I bring this up, because the motion-sickness suggests there is an adaptation process that is occuring for FPS players who do not experience discomfort.
Admittedly in wild speculation, this first-person adaptation process might correlate to educationally-applicable improvements in spatial cognition. For example, the ability to decouple vestibular and motion feedback might correlate to improvements in mental rotation, since mental rotation necessarily decouples the stable (non-rotating) vestibular state from the rotation task.
Anyway, I'd like to learn more of the cognitive mechanisms at work here. Do you have any leads to suggest?