A Rule Mining and Bayesian Network Analysis to Explore the Link Between Depression and Digital Behavioral Markers of Games App Usage

Amid the COVID-19 pandemic, spending time on Games increased much, which may impact mental health. While numerous studies were conducted exploring the relation between Games and depression, none of the studies used objective (i.e., actual) Games app usage data which could provide unbiased and real-time insights. To fill this research gap, using our developed app that retrieves the past 7 days’ actual app usage data accurately, we conducted a study on Games app users (N = 60) in Bangladesh. We extracted the behavioral markers from the foreground and background Games app usage events’ data. To explore the relation between Games and depression, we mined rules, did correlation analysis, and built Bayesian networks. Our analyses demonstrated that the students who spent higher time and had a higher launch per Games app on weekends were more likely to be depressed (p < .05). In addition, from the Bayesian analysis, we found that while some usage data impacts depression, depression also impacts some usage behavior such as frequency of launching Games apps. Apart from raising awareness about the negative impact of Games, insights from our study can facilitate the design of systems to improve the students’ mental health.