Long time followers of mine may remember a blog post I did dissecting some of the scientific papers that were published around video games causing violence. Unfortunately the wayback machine seems to have missed that post (or at least, it was too big a pita to find it) so... let's start over.
Scientific studies have consistently shown that there is an increase in transferable skills, but the effect is very localized. Moving thumbpads or a mouse to shoot someone? That's not going to make you a more effective killer. Flying a plane with a flight stick, throttle, and rudder pedals in a simulation that accurately represents wind? That will help some. Here is a meta analysis that cites problems with most of the studies done up to that point, and here are studies that show how video games can help improve non-specific, super general skills such as fine motor coordination, spacial reasoning, and general education (when designed by a good teacher who understands the advantages of the medium).
So, I claim that flight simulators with proper equipment could make you a better pilot. What about shooters with a toy gun? My personal intuition on the subject is that toy guns like the ones you'll find in expensive arcade machines do not relate very well to actual guns. There's never a safety. There's never a jam. There's never a misfire. There's never a realistic recoil. There's never a realistic sound, at least in terms of volume. I can't think of an example where the weight of a video game rifle was close to the weight of a real world pistol. I own a gun. I take it to the range. I have a healthy fear of the device, and I treat it with caution and respect. It's presence in my hand fills me with a variety of emotions. Toy guns never do that.
But a paper disagrees with my personal experience. It studied 151 participants over three video games, Resident Evil 4, Wii Play, and Super Mario Galaxy. In RE4 and Wii Play participents were given either a "traditional" controller (it's not specified, but I suspect it was a Wii pro controller from the description) or a "pistol" controller (the wii-mote in a plastic shell). They played these games for 20 minutes, then went to the range with an airsoft pistol to fire rounds at a mannequin.
That all sounds fine, except you don't have a round where the participants shoot at the mannequin to establish a baseline. Also, airsoft pistols don't address a couple of the differences I highlighted between real guns and toy guns - principally the volume. Then you get to the results. First, there was no significant change detected between players who had firearms training and those who had not. Then there's no significant change between players who were rated higher for aggression and those who were rated lower. That doesn't make a lot of sense to me.
The paper does not provide raw data for the study. It has a chart. And the numbers on the chart, to me at least, don't make a lot of sense. Maybe I just don't know how to read this type of chart. Thankfully, there are professionals for that. While the paper was in peer review questions were raised about the numbers, a request for the raw data was put in. The authors of the study never got back to it. It was later revealed the data was lost, which raises questions about the validity of the study, such as "Did you cherry pick or massage the results a little bit?" A few weeks ago, the University of Ohio issued a retraction for the study. Justice is served.
I'm not sure why there is such an impulse to blame video games for violence. The numbers show a coloration between an increased percentage of households that own a video game system and a decrease in murder rates. If nothing else, murder (and violent crime in general) are way down. There doesn't seem to be much need for a scapegoat.