Do video games train for violence?

At this year’s Human Brain Mapping meeting in Toronto on 12-16 June, Klaus Mathiak from the University of Aachen in Germany presented a study of brain activity recorded while volunteers played a violent video game. The study involved 13 men aged 18-26 who were regular game players. The researchers gave them a game that required them to navigate a bunker, killing terrorists and rescuing hostages. Once the players were proficient they had their brains scanned to reveal areas that became more or less activated during the moments of virtual violence.

The major findings of the study included increased activation in areas of the brain believed to be involved in thought, followed by suppression of activation in areas of the brain believed to be involved in emotion. Mathiak and his colleagues concluded that playing violent video games can train brains into aggressive thinking and violent behaviour. These findings, and the interpretation, were recently reported by the New Scientist (1).

I have several problems with this study, and with the way it has been framed. Firstly, let’s look at the science itself. The research is plagued by the usual limitations of studies involving brain imaging: small numbers and a contrived scenario.

The 13 men involved could easily be unrepresentative of video game players as a whole, and are almost certainly unrepresentative of non-game players and men as a whole. To demonstrate that violent game playing creates the pattern of brain activity observed it would be necessary to have a group of players naive to video games also taking part in the study. If their pattern changes during the course of the experiment to match that observed in the regular players, it might be fair to say the game is producing the observed pattern. Otherwise, it might be a peculiar feature of the brains of the violent video game players selected, existing in advance of any game play.

There are also the peculiarities of the experiment. Having your brain scanned with a magnetic resonance scanner is an odd experience. You lie inside a narrow and claustrophobic tunnel with instructions to hold your head extremely still. When the scanner is recording brain activity it makes a very loud banging noise. During this, the subjects in the Mathiak study will have observed the video game through a mirror pointing at a projection screen. While this setup provides some access to the brain during game play, it is likely to be quite different from game play that takes place at home or in an arcade. Some of the intricacies of strategising and thinking about the game might be abandoned just to get the game over quickly, and some of the usual absorption into the game may be lost.

The experimental set-up can be very different from the real world. It is therefore irritating that the authors have pursued a pat moral interpretation of their findings. The suppression of activity in the anterior cingulate and amygdala is taken to mean that the emotive centres of the brain were switching off, and that this is a bad thing. The suggestion is that playing violent games will undermine empathy in the real world, and provide for an escalation of violent behaviour in game players.

There is, however, little evidence that playing violent video games increases violent behaviour. Large numbers of children play games with violent content without going on to perform actual acts of violence (2). The suggestion that video games cause real acts of violence was aired last year when the parents of a 14-year-old blamed the game Manhunt for his death. Police investigating the murder, however, dismissed the involvement of the game. The subsequent increase in sales of Manhunt is not yet to correlate with any increase in violent acts.

Mathiak and his colleagues could have provided a very different interpretation of their findings. It could be, for example, that the suppression of emotive centres is advantageous before completing a highly skilled motor action. Indeed, a very large number of studies, also presented at Human Brain Mapping meeting, have shown that the emotive part of the anterior cingulate cortex regularly reduces in activity during performance of almost any complicated task. There is no reason to believe this alteration in brain pattern represents a flight of empathy and a precursor to violence.

The strength of functional imaging is that it can provide us with a more refined and sometimes surprising view of brain function. The brain, however, is a complex organ and changes in brain activation can be interpreted in many different ways. We must be cautious in deciding what these colourful patterns of activation really mean. Mathiak and his colleagues have completed an interesting pilot study into the effects of video games on brain function, but their suggestion that video games train the brain for violence is premature and reaches beyond the data. If their interpretation is accepted it will not be because of the strength of their study but because of the strength of moral aversion to virtual violence.

Stuart Derbyshire is a senior lecturer at the School of Psychology at the University of Birmingham.

(1) Do games prime brain for violence?, New Scientist, 23 June 2005

(2) Parents ‘ignore game age ratings’, BBC News, 24 June 2005