In my favorite kind of "Take That!", people who argue that video games are making people more violent now have to contend with the country of Somalia:
(Link to video) Somalian teenagers are playing video games, and it seems to be taking over the free time that would otherwise have been spent fighting for radicalized militias. They are, apparently, playing mostly nonviolent games, like soccer or racing.
It's possible that all this proves is that video games tend to pull people towards a mean level of violence that's less than normal in Somalia, but still more than in America. But I don't think that's likely to be true. I think that one of the major causes of violence by teens is excess energy or aggression, and spending that energy on video games rather than fighting (a.) reduces the amount of energy available to throw into violence, and (b.) teaches teens to channel their aggression into a system designed for it, and which can handle the aggression without harming anyone.[1. I feel it's important to state that this theory does require assuming that people have different sorts of energy, and that energy that's good for, say, violence can't be used on, or exhausted by, doing math homework. As far as I know, this hypothesis doesn't have explicit backing in psychology.]
I wonder if this news could be used to make the case for government-funded arcades?