Background/Aims: It is a common concern in the research field and the community that habitual violent video gaming reduces empathy for pain in its players. However, previous fMRI studies have only compared habitual game players against control participants cross-sectionally. However the observed pattern of results may be due to a priori differences in people who become gamers and who not. In order to derive the causal conclusion that violent video game play causes desensitisation, longitudinal studies are needed. Methods: Therefore we conducted a longitudinal fMRI intervention study over 16 weeks. Participants were randomly assigned to 1) play a violent video game (Grand Theft Auto 5), 2) perform a social life simulation game (The Sims 3) 30 min/day for 8 weeks, 3) serve as passive control. To assess empathy processing, participants were exposed to painful and non-painful stimuli (e.g. someone cutting a cucumber with or without hurting herself) either as real photographs or video-game like depictions in a 3T MRI scanner before and after the training intervention as well as two months after training. Results: We did not find any evidence for desensitization in the empathy network for pain in the violent video game group at any time point. Conclusions: The present results provide strong evidence against the frequently proclaimed negative effects of playing violent video games and will therefore help to communicate a more realistic scientific perspective of the effects of violent video gaming in real life.