Categorization of Loot Box Design: A Scoping Review of Problematic Features

Abstract Purpose of Review Loot boxes have become a central form of microtransaction in video games, yet the extent to which specific design features contribute to problematic behaviors remains insufficiently understood. The present review systematically maps and categorizes loot box features studied in existing research, highlights common design patterns and research gaps, enables more precise analysis of player behavior by considering design differences, and aims to provide a foundation for standardized evaluation and evidence-based regulation. Recent Findings A systematic search of four major databases (PsycINFO, Scopus, Web of Science, and DiGRA), followed by an update in March 2026, identified 406 records. Of these, 72 empirically based studies published between 2017 and 2026 met the inclusion criteria. These studies revealed 136 distinct loot box design features organized into 22 subcategories across six broader dimensions: (1) Obfuscation, (2) Psychological Manipulation, (3) Social Influence and Promotion, (4) Value, Data, or Engagement Manipulation, (5) Self-Regulation Obstacles, and (6) Gambling Resemblance. The most prevalent features included cosmetic rewards, in-game currency, pay-to-win mechanics, cash-out options, and rarity. Significant research gaps were identified, especially regarding the impact and interaction of these features. Summary This categorization provides a structured framework for assessing loot box design, helping researchers, policymakers, and industry stakeholders understand the complexities of these mechanisms and their behavioral impacts. The findings highlight the need for further research into emerging trends and feature interactions, as well as the development of improved analytical tools to support evidence-based regulation and responsible game design.