Loot boxes are the focus of growing research and regulatory attention. While they are frequently treated as a monolithic feature of games by researchers and policymakers, loot box implementations are not uniform: the features of loot boxes vary widely from game to game in ways that may have important consequences for player spending and behaviour. Despite this, previous attempts to classify loot boxes have either not focused on the impact of loot box features on player behaviour and spending, or have not attempted to fully map the different forms that loot boxes currently take. In this work, we attempt to illustrate the nuance present in loot box implementation in a featural model. Using our lived experience, a qualitative coding exercise, and consultation with an industry professional, we identify 33 features of loot box-like mechanics that might be expected to influence player behavior or spending, which we group into 6 domains: point of purchase, pulling procedure, contents, audiovisual presentation, salience, and social. Each feature is broken down into two or more categorization tags for a given loot box, and illustrative examples of each feature are provided. This work may serve to guide researchers in studying how different types of loot boxes may affect players, aid regulators in ensuring that any proposed legislation is sufficiently nuanced to handle the wide variation in loot box design, and help parents and players to better understand the inner workings of loot boxes during play.
Cite this work
@inproceedings{ballou2020hidden, author= {Ballou, Nick and Gbadamosi, Charles and Zendle, David}, title= {{The hidden intricacy of loot box design: A granular description of random monetized reward features}}, year= {2020}, publisher= {PsyArXiv}, abstract= {Loot boxes are the focus of growing research and regulatory attention. While they are frequently treated as a monolithic feature of games by researchers and policymakers, loot box implementations are not uniform: the features of loot boxes vary widely from game to game in ways that may have important consequences for player spending and behaviour. Despite this, previous attempts to classify loot boxes have either not focused on the impact of loot box features on player behaviour and spending, or have not attempted to fully map the different forms that loot boxes currently take. In this work, we attempt to illustrate the nuance present in loot box implementation in a featural model. Using our lived experience, a qualitative coding exercise, and consultation with an industry professional, we identify 33 features of loot box-like mechanics that might be expected to influence player behavior or spending, which we group into 6 domains: point of purchase, pulling procedure, contents, audiovisual presentation, salience, and social. Each feature is broken down into two or more categorization tags for a given loot box, and illustrative examples of each feature are provided. This work may serve to guide researchers in studying how different types of loot boxes may affect players, aid regulators in ensuring that any proposed legislation is sufficiently nuanced to handle the wide variation in loot box design, and help parents and players to better understand the inner workings of loot boxes during play.},
}