Violent vs. non-violent video games’ improvement on cognitive performance (Barlett et al., 2009)

sleeping_chiakiSince my last post, I’ve been feeling lethargic. Trying to find out the cause(s), I underwent through introspection. At first, it might’ve been I was eagerly waiting for the first episode of Minami-ke: Okaeri, which distracted me from what I wanted to do. Then it might’ve been the annoying peccadilloes from my laptop, my procrastination on the internet where I spent an inordinate amount of time on the net, or my self-loathing attitude about my independent project which was severely setback on its timeline. So I was thinking of reviewing a journal article might bring me out of my lethargy. Here’s one from Barlett et al. published in Computers in Human Behavior.


Two studies were conducted in order to determine the impact computer games had on the cognitive performance. Study 1 evaluated a measure of cognition, which incorporates aspects of short-term working memory, visual attention, mathematical decision making, and auditory perception. Study 2 measured the cognitive performance between those who did not play video games versus those who played either a violent or non-violent video game. Results from Study 1 indicate participants needed approximately four trials to reach asymptotic performance on the cognitive measure. Results of Study 2 showed that participants who did not play any video game did not have a change in their cognitive performance, while those who played either a violent or non-violent video game had an increase in their cognitive performance.

I know I should take a break because of the winter break, but…

In their introduction, they reviewed several cognitive abilities and the research literature as it relates to video games. Visualization where the mental rotation task falls in; Concentration or selective attention, efficient visual focus on relevant visual information and screening out distractors; visual scanning and visual tracking. However, the research on video games’ effect on specific cognitive abilities received mixed results, as argued by the authors (see my previous post on Boot et al., 2008). They cited several methodological flaws in previous studies, such as sampling, games used, video game experience, violence, practice effect from the cognitive measures, etc. They especially focused on Kearney (2005) of which this study is based upon (i.e. replication study).

So for their study, they started examining general cognitive abilities instead of specific abilities. Why not? Science can be compared to an essay: start with a broad statement, then funnel it until you formulate your thesis statement. They examined whether non-violent or violent content in video games has any bearing on cognitive performance. They also addressed some methodological flaws from previous studies, namely video game experience in order to be applicable to the general population and practice effects from the cognitive measures.

Study 1: it’s a pilot study and I will not write anything about it. Please refer to the abstract.

Study 2


Participants: 113 (57 male), mostly white, average age 19, average play time is 0.77 hours during a weekday and 1.78 hours during weekends. So a week total of around 5.63 hours.  It should be noted that the distribution is large, so I’m guessing it’s statistically close to the population average of 7 hours (Is it? Don’t remember my source).


Cognitive measure: SynWin is a cognitive computer program used to measure general cognitive abilities. It was also used in Kearney (2005). SynWin measures working memory, auditory perception, visual attention and mental addition:

Working memory: you have to remember this set of letters and then later we show you a letter and you have to respond whether this letter was one of the letters you’re supposed to remember. You get it right, plus ten points, you get it wrong, minus ten points.

Auditory perception: press a button if you heard a high tone, but ignore a low tone. You get it right, 20 points, you get it wrong, minus 20 points.

Visual attention:  You have a gas gauge and you have to keep the gas meter from being empty. You get minus 10 points if that needle isn’t moving for every ten seconds. Huh, I need more info on this task because the second part of this task is confusing me.

Mental addition:  three digits additions. 20 points for getting it right and minus 20 for getting it wrong.

Here’s a twist: all these four tasks are done simultaneously, IMO which means SynWin also requires multi-tasking, although this is not stated explicitly.

Games used:

Non-violent games: two games are used: the Tile game, the closest description is Mah-jong solitaire, remove all tiles by matching identical twin tiles. The other game  is the marked numbers game, identify numbers that are in a range, say even numbers between 10-36.

Violent game: Red Alert 2.

Let’s see if I could justify this selection of games. They require a good of deal of working memory, especially the mah-jong solitaire (I played it before), you have to remember where this tile was and spot its identical twin or remember the enemy positions and attack vectors. They require visual attention, more specifically scanning and tracking, you have to be efficient in spotting your needed tile or units of interest in Red Alert 2. Both types have variety, the tiles game have a lot of twins, the numbers game got a lot of numbers, and Red Alert have infantry, tanks, aircrafts, ships and so on. They both require a lot of thinking to put it bluntly.

However, differences between the games may arouse suspicions from gamers and readers. For one, Red Alert 2 requires players to manage resources which put them on a different spot than those in the non-violent game. Nevertheless, I haven’t played Red Alert 2 for a long while and I don’t know the A.I.’s tactics on resource disruption. Without resources disruptions, then players would have a lighter cognitive load (i.e. thinking on spending wisely minus planning ways to protect resources gatherers). Another difference is that the non-violent game is a tile-based game whereas Red Alert 2 is not. A pseudo chess game could be appropriate as an alternative. However, this game would certainly be made out of scratch and would be dissimilar from a common video game. Another detail is the learning curve, the non-violent games are easy to learn, but hard to master, whereas Red Alert 2 is more difficult to learn and is also hard to master. Finally, the non-violent games are not of common video game stock (i.e. it’s not Mario or Katamari Damacy), although it’s quite hard to find an equivalent non-violent video game. I guess there are enough justifications for their selections.


Participants are randomly assigned to be in the control group (no play), non-violent video game group or the violent video game group. They do four SynWin sessions, each session last 5 minutes. After that, the non-violent game participants played the numbers game for 4 minutes, then the tile game for 10 minutes, and then the numbers game for 4 minutes. As for the violent game participants, they played Red Alert 2 for 18 minutes (in order to be equivalent in time with the non-violent games). The control group are asked to do some internet searches on some boring topic for 18 minutes, too. After that, participants go through a fifth and final SynWin session.


For people who know statistics, something about violations of assumption of multicollinearity and assumption of normality.

For those who don’t (i.e. me), analysis is done with repeated ANOVA. Using the SynWin session scores (session 4 and 5) as the dependent variables. Regardless of content, participants in the video game condition showed improved cognitive performance in relation to the control group who showed none. Further analyses were conducted for any possible “third variables“, such as gender, video game experience, SynWin session differences, etc. None were found.


I’ve said everything I could for this study. From a general standpoint, I guess we’ll have to investigate long term effects on cognitive performance, minimal required time for playing video games before improvements on cognitive performance are observed, whether it applies to real life situations and more research.

The authors, however, covered pretty much what I would say. General applications in the workplace training or “warm ups”; less reason to use violence as means to improve cognitive performance when you could do it without, just a matter of making it fun and challenging which is in itself a challenge given how violence can be an easy seller for excitement; their research methodology allow a clear reason to attribute video games in improving general cognitive performance.

IMO, I’m impressed that after 18 minutes of play they showed improvements. But then, participants went through 4 SynWin sessions before playing. Does it have any effects on their cognitive performance? No because they’ve considered it into account. So I’m good.

Some limitations the authors mentioned:  they didn’t measure participants’ mood, arousal and aggression, so interpret results with caution (so they say). The non-equivalence between the video games (I mentioned that earlier), such as amount of action, concentration (I’d say the non-violent game requires more focused attention whereas RTS requires rapid eye movement), fun level and pace (Well I guess modding the game’s pace can be achieved since we have various examples of RTS with different pacing). Video game performance was not 1231091499658measured because scoring can’t be done until the match is complete. A time limit mod on Red Alert 2 might do the trick, but who does that?

Just yesterday, I got to watch the first episode of Minami-ke: Okaeri and I’d say it’s better than the second season, but somewhat less than the first season.

Barlett, C. P., Vowels, C. L., Shanteau, J., Crow, J. & Miller, T. (2009). The effect of violent and non-violent computer games on cognitive performance. Computers in Human Behavior, 25, 96-102


2 thoughts on “Violent vs. non-violent video games’ improvement on cognitive performance (Barlett et al., 2009)

  1. Thanks for all the interesting items that you post on this site.

    As a researcher, I’d like to think that I do a good job of staying on top of this area. However, you’ve introduced me to other really helpful pieces of research.

    Mucho thanks!

  2. Pingback: Cognitive performance in games: violence vs. non-violence? « Vincto :: Game Harder!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s