This isn't a story about whether or not games can make people violent. It's about how science tries to test the effects during research. Clever psychiatric stuff? Delicate machines that measure brainwaves? Nope: chilli sauce and air horns.
As reported in The Guardian the evidence "simply doesn't support [the] claims" that violent games can lead to actual violence. What's incredible though are the means by which it's all tested. Here's how the tests used are described:
"For example, one experimental test that’s often used is a modified version of the Taylor Competitive Reaction Time Task. Here the participants are first asked to play either a violent or non-violent video game. Afterwards, they’re asked to play a reaction time game against another, fictional player. If they win a particular encounter, they get to blast their opponent with a loud noise. The key manipulation is that the participants choose how loud the noise is, and how long it lasts for. Longer, louder noises are taken as a measure of increased aggression.
Another task, called the "hot sauce paradigm", measures aggression by having participants prepare a cup of chilli sauce for another (again, fictional) participant. The more hot sauce they put in the chilli, the more aggressive they are deemed to be, and some studies have shown that people who are asked to play violent video games beforehand use more hot sauce."
So to clarify: tests that 'prove' videogames make people violent do by using noises and condiments before equating that to actual physical aggression. It's worth reading the whole article as, despite the wealth of research and number of studies being conducted, the field sounds like a mess of poor practice and conflicting interests