Psychology's Enduring Problem With Gaming

So, today I read a study (that I was alerted to via Kotaku) about a meta-study into potential causal links between videogames and aggression. Note that they were looking at aggression, and not violence. That bird flew years ago, but aggression studies linger and many have not wised up to how bad most of these studies are. I think this is indicative of a big problem that exists between psychologists and gaming: a certain snobbery that just leads to a lack of credability. In an effort to counter this, the American Psychology Association created a task force of 7 scientists to find out, using the best science avaliable, if there is a link between videogames and aggressive behaviour. They found that there was, and as someone who A) plays violent videogames and B) is rarely aggressive I wondered how good a job they had done. Oh dear.

Psychology is tricky business. There are a lot of variables involved in the human mind and it is near impossible to isolate factors that can influence the outcome. But, apparently, the team conducting the APA study did the best they could do with the best science available. Unfortunately, most of the science available is crap – skewed heavily by personal interests or by negligence in failing to account for such basic influences on aggressive behavior as competition. Let’s be serious, a bit of GTA isn’t nearly as frustrating as a Halo match where your K/D ratio is 0.4. But I am yet to find a single study into videogames and aggression to take account of this basic facet of gaming.

My first problem with the study is that when searching through other studies for their meta-meta-analysis, they used ‘violence’ as a keyword, which tends to turn up studies that are unreliable; either due to conflict of interests or bad science. So, many of their sources suffer from those problems because the team used violence as a search term. Their review of previous meta-studies is frankly as barebones as it gets. ‘Given the breadth and publication dates of these studies, the task force determined that they sufficiently covered the existing literature through 2009.’ Then they identified 6 problems with that literature, which perhaps suggests that they are not great source material.

In fairness to them, they then put the post-2009 studies through a screening process, making sure they had been independently reviewed and ensuring that they had had considered outside variables in their tests. They also made sure studies contained at least one credible empirical test. Out of 170 studies considered, only 31 were considered credible enough to work with. Even then, only 3 (yes, 3) studies considered the interaction between outside influences on aggression and playing videogames - a dire demonstration of psychology’s poor treatment and understanding of the medium. Of 170 studies, 3 are properly fit for purpose.

Once they aggregated the results using methods I don’t fully understand (I recognize a few of them from Statistics textbooks but that’s about it) they found that there was some link between videogames and aggression. But no link between videogames and violence, or delinquency. So at least those two accusations have been firmly put to bed. But the link they identified did not account well for competition, or gender, or age, or population demographics. Those are some serious holes in the research and they admit as much in their conclusions.

The APA’s analysis also found that most psychologists used the ESRB ratings as an indicator of how violent a game is, which the study describes as ‘adequate’ owing to the fact that the ESRB is at least independent and tends to focus on violence above swearing or sexual content. I cannot disagree, I’d rather something better, but ESRB does the job.

The analysis also acknowledges that most of the 31 credible studies only showed correlation, and did not distinguish between potential other factors, notably the competitive element of videogames. Dying in Assassin’s Creed isn’t nearly as annoying as failing to beat my high score in Tetris – but most studies did not take this into consideration. It is also telling that the analysis found that ‘no single method or study is conclusive in this field’ – though it noted that the weaknesses of some studies were made up for by the strengths of others, but that this was still flawed. Then there is the minefield of how to properly measure aggression.

That is extremely problematic, and the analysis glosses over the issue. For reference, some of the studies that passed their screening process had these tests:

• The “short story” test, where the subject is given the beginning of a writing prompt and told to fill in what happens next. A more aggressive story is indicative of more aggression
• The “noise” test, where a subject is asked to press a button that delivers an annoying sound to another subject, then evaluated based on the duration and intensity of the noise
• The “hot sauce” test, where a subject is asked to give hot sauce to another subject and is evaluated based on how much sauce they give and how spicy it is

Many just used questionnaires, which are highly subjective and frankly not that useful unless the sample size is gargantuan.

Then the analysis goes really downhill. After identifying that studies don’t take account of third factors, the interaction of third factors on how a person plays videogames, or the background of participants, the study then calls the research ‘robust’ – I’m not sure anything with that many flaws can be called robust. The study then noted that ‘competition may provide an additional independent influence’ – no shit guys. For people with PhDs, they sure do fail to deal with the obvious.

I should also point out that the analysis is incredibly short and light on depth, apart from when it delves into statistical analysis. I’ve written more detailed research pieces than this as one 18 year old boy, not ‘7 senior scientists with exemplary methodological and scientific expertise’. For reference, my piece (which was a study into the Harappan Civilization) was longer, had more sources and had more credible sources. I got an A grade, not the A* I was hoping for. Frankly, if I’d submitted the study of these 7 scientists I would have gotten a low B at best. It’s really quite pathetic. Granted, history is easier than psychology to do meta-analysis on, but still – 7 of the USA’s best scientists should be able to outperform an A-Level student.

So – most studies into videogames are crap. We all knew that, but it seems the APA actually knows it now. But still, they glossed over the fact. It’s really quite frustrating as a gamer who plays a lot of violent games, but is only very rarely aggressive, to see my hobby treated so poorly by psychologists. Studies need to isolate gaming as the only variable – and they especially need to find violent games that do not have a component of competition. Studies should sit a large sample size in a room, do a wide array of tests on their aggressiveness before and after gaming sessions, then take account of the subjects’ backgrounds. The fact that only 3 of 170 studies in the last 6 years even tried to do that is shockingly bad.

I guess we’ll be waiting a little longer before psychology takes gaming seriously.

LOOK WHO CAME:

James Internet Ego   

Dreggsao   1

Forsakeneyes   1

Jinx 01   1

Logeon   1

Gamemaniac3434   1

extatix   1

Gajknight   1

OverlordZetta   1

GodEmperor Paige   1