Our paper “Failure to Demonstrate That Playing Violent Video Games Diminishes Prosocial Behavior” went live yesterday morning. Here’s a link to the PLoS One page.
There are three key things we report in paper:
- We were unable to replicate findings by Greitemeyer & Osswald (2010) that prosocial games increase prosocial behavior.
- We were unable to find a reason for Greitemeyer & Osswald’s failure to demonstrate a detrimental effect of violent games.
- While were unable to reveal an effect of video game content on behavior, we were able to demonstrate that slight differences in the administration of the experiment led to large changes in behavior
On a personal note, I’m really happy to see this work published. When we started writing, we were resigned to the fact that we would have trouble getting this manuscript accepted, given that it’s essentially three failed replication attempts. Failures to replicate have typically been difficult to publish because they are often viewed as less interesting or unfairly labelled as difficult to interpret. In a literature as divided as the violent video game literature, I think null effects are valuable. If we never report difficulty in reproducing past research, then we are only seeing a partial account of the true nature of the effects.
Some people might wonder why we report the direct replication last. We initially wanted to ask a different series of questions using the pen-drop task and felt it was important to replicate the Greitemeyer and Osswald studies as a base from which to test our ideas. What followed was a number of surprising failures to conceptually replicate, which culminated in a last-ditch attempt to reveal the effect by conducting a replication as faithful to the original as we could run. The order of experiments in the paper reflects this thought process.
Tear, M. J., & Nielsen, M. (2013). Failure to demonstrate that playing violent video games diminishes prosocial behavior. PLoS One, 8(7): e68382. doi:10.1371/journal.pone.0068382