As for video games… that's a tricky one. I recall an anecdote about two teens holding up a bakery at gun point. The baker at some point made a move that was a bit too sudden, and the teen shot him in the forehead. One possible facilitator is very similar to a soldier's training: simple conditioning about shooting the bad guys as soon as they threaten your avatar in your favourite FPS. For all I know the teen didn't even intend to shoot.
Of course, it takes more than a lifetime of violent video games to actually point a gun at someone. But a tiny nudge can sometimes push you over the edge. By the way, I expect the nudge will get bigger with VR. (So will the benefits, I expect.)
The host of scientists who conducted their studies know this. Which is why they did many such studies (some of them very large scale), controlled for confounding variables, modified the parameters across time in both direction (watching more and watching less this and that kind of show), and of course reproduced most of the results.
Go tell them "correlation does not prove causation", and they will laugh in your face. We're long past correlation by now. Causation itself has been thoroughly demonstrated.
I can think of arguments for why white-nationalist forums would decrease the likelihood that the white nationalists who find them would go and kill non-whites. (a) Such a person might feel relieved to find that there are others who feel similarly, and therefore not feel like their cause has no one working for it and they need to do something spectacular and terrifying to attract attention. (b) They might have few friends in real life, and they can't fully show themselves to the ones they do have, so they have no truly close friends. But on the forum, they meet fellows who share their beliefs and feel they can trust, and then they make what feel like close friends. And now they'll know that, if they kill someone, that carries a high risk of getting thrown in jail and cut off from their friends; that extra disincentive might be enough to stop them. (c) On that note, the most visible and admired participants in the forum are not going to be those who go out and kill people, because (I assume) people can't continue posting when they're in jail. The general incentives of the forum probably point towards being a blowhard who doesn't do anything.
This is why I will only accept the conclusion "Letting white nationalists gather on a particular forum will cause more white-nationalist-perpetrated killings" if I see a decent study showing it.
(I think one of the official basic arguments for freedom of association and of the press is that, if you prevent people with grievances from associating and publishing their views, they will go underground and then get violent.)
(Incidentally, I do like it when one of my possible conclusions is that, in order to reduce the likelihood of people doing terrible things, the solution is to work to raise everyone's standard of living, so they'll be too comfortable to want to risk losing what they have.)
The list of murderers who've listened to 2Pac is quite long, for example.