3 185 IN-GAME PLAY BEHAVIOURS PREDICT IMPROVEMENTS IN ANXIETY DISCUSSION The video game MindLight translates evidence-based techniques into game mechanics that teach children how to cope with anxiety in a playful manner. The aim of the present study was to investigate whether children with elevated levels of anxiety improved in their anxiety levels through these game mechanics that were explicitly designed into MindLight. Based on the anxiety literature, two types of in-game play behaviours (i.e., “engaged” and “avoidant/safety”) that are most relevant to the intervention goals of MindLight were distinguished and coded during gameplay. First, contrary to what was expected, pre-test anxiety scores were not associated with the different in-game play behaviours during the first play-session. Second, in line with our predictions, increases in one of the in-game engaged behaviours and decreases in the two avoidant/ safety behaviours predicted lower anxiety scores at 3-months follow-up. Together, these findings suggest that mechanics related to exposure techniques predicted improvements in children’s anxiety symptoms. Associations between Pre-test Anxiety Scores and In-Game Play Behaviours Our finding that none of the in-game play behaviours, and the avoidant/ safety behaviours in particular, were associated with pre-test anxiety scores was unexpected. Because safety behaviours are an important maintenance process in anxiety disorders (Clark, 1999), it was expected that children higher in anxiety would try to create more light by turning on ceiling lights and to hide more inside chests, because these objects provide a way to reduce and/or avoid exposure to fear events. The current study is part of an indicated prevention trial in which selection was based on scoring one standard deviation above the mean and clinical cases that already received mental health care were excluded. Therefore, the current finding might be due to a restricted range in anxiety scores. It might also be that the first play-session was standardized in such a way (with a lot of cut-scenes explaining the game) that little room was left for children to show very different in-game play behaviours. Nevertheless, it is promising that irrespective of their pre-test anxiety scores, children started playing MindLight in a similar way because this strengthens the findings for the second hypothesis; changes in in-game play behaviours and the predicted improvements in anxiety symptoms are not due to associations between initial levels of anxiety and in-game play behaviours during the first session. Furthermore, the finding that none of the in-game play behaviours during the first session was associated with behaviours during the last session (except
RkJQdWJsaXNoZXIy MTk4NDMw