In June 2018, a commentary article entitled “Video game loot boxes are psychologically akin to gambling” by Aaron Drummond and James D. Sauer (2018) was published in Nature. The authors assess if loot boxes (i.e., purchasable randomized rewards in video games) are analogous to gambling and, if so, whether or not they should be regulated under the same legal terms.

In order to assess the similarity between loot boxes and gambling, the loot box elements of 22 select video games were evaluated against the unique criteria that distinguish gambling from other risk taking behaviour as defined by Griffiths (1995).

These criteria are: 1. The exchange of money or valuable goods; 2. An unknown future event determines the exchange; 3. Chance at least partly determines the outcomes; 4. Non-participation can avoid in incurring losses; 5. Winners gain at the expense of the losers.

The authors added an additional criterion: 6. The ability to “cash out” winnings.

This sixth criterion is important, as the ability to “cash out” is often a differentiating criterion for gambling as defined by various regulatory bodies.

For each game, the authors assessed the characteristics of the in-game loot boxes by watching videos of players opening the loot boxes. The authors found that just under half (45.45%) of the 22 games analysed met all five of Griffiths’ criteria for gambling, with another four allowing for players to “cash out” their items via third-party marketplaces. As a majority of the 22 games analysed in their sample met at least two of the six criteria, the authors conclude that the loot boxes within these games meet the psychological and legal definitions of gambling. They go on to recommend that the Entertainment Software Ratings Board and other regulatory bodies should consider restricting access to loot box content in video games for players under legal gambling age.

While the importance of determining whether or not video games are integrating unregulated gambling features in their games cannot be understated, there are several limitations of this analysis that need to be considered before accepting the authors’ conclusions.

Lack of systematic selection of video games

The small number of games analysed in this article is problematic. 22 games are not a representative sample of the range of video games available today (or even of games released in the past two years). To put this into perspective, in Q4 of 2016 (September 2016 – December 2016), 142 video games were released (Espineli, 2016).

The only information given in the article as to why these particular games were chosen was that they were games “released in the past two years” (p.2). Furthermore, no justification is given as to why the list of 22 games includes multiple versions of the same game. 27% of the games included were from the same franchise (i.e., Call of Duty: Infinite Warfare, Call of Duty: WWII; FIFA 17, FIFA 18; and Madden NFL 17, Madden NFL 18).

Perhaps because of the inclusion of multiple iterations from the same gaming franchises, the list of analysed games does not span the range of video-game genres or differentiate between the use of different gradations of loot boxes for different types of games.

Video games can differ widely in gameplay, setting and content. These differences, can impact a user’s experience with any particular individual game, in relation to player motivation, retention, and engagement. For example, players of first-person shooters (a genre that includes games where one shoots at, or destroys a series of objects or opponents through a first-person perspective, such as Call of Duty or Unreal Tournament; Wolf, 2001) are more likely to report being motivated to play to achieve a sense of achievement, while players of role-playing games (a genre that includes games which involve taking on the role of a player and progressing through a narrative, such as the Final Fantasy Series or World of Warcraft; Wolf, 2001) are more likely to be motivated by a desire to be immersed in a virtual space (Ghuman and Griffiths, 2012; for more on video game genres and game classifications see Wolf, 2001).

These variations in genre are important to consider when drawing conclusions about the uses and effects of video games in a generalized form (as this article does). For example, the single difference in play motivation discussed above reflects the competitive, rather that cooperative nature of shooter games as compared to games in the role-playing genre (Zhbek & Khoo, 2002). It is not much of a leap to suggest that players may be more inclined to purchase loot box content in games played competitively.

The lack of differentiation between gradations of loot boxes is also problematic as some loot boxes can provide clear strategic advantages (e.g., better weapons) whereas others are simply cosmetic (e.g., different costumes). Using the above example, players may be more motivated to purchase loot boxes with strategic advantages in games played competitively whereas cosmetic prizes may be more desirable in role-playing environments. Without any differentiation across game genres or the contents of loot boxes, the authors imply that all loot boxes are equal in their appeal and in-game use, as well as their regularity in video games across genres.

Application of criteria The authors determine whether or not loot boxes constitute gambling by utilizing Griffiths (1995) criteria. Whilst these criteria have been widely used, it is difficult to use them to differentiate between what constitutes an in-game gambling mechanism and what constitutes the use of randomized rewards. For example, using the Griffiths (1995) criteria, it is unclear how the characteristics of loot boxes differ from the random rewards in a host of chance based games (e.g., baseball, Magic the Gathering, or Pokémon). If the authors aim to understand the unique features of loot boxes within video games and how they may or may not constitute gambling, they need to first demonstrate how the integration of loot boxes is not simply just an extension of the chance based play mechanics that have been evident in video games (and other chance based analog games) for decades.

Looking bigger picture, as it has been widely publicized, the World Heath Organization has included gaming disorder in their latest version of its disease classification menu. Many experts have described the decision to class gaming addiction as a mental health disorder as “premature” and based on “moral panic” more than hard science . Putting this hotly disputed debate aside, it may be time to pose the question whether it is out-dated to consider games through a gambling paradigm rather than the developing arena of video game analysis?

The ability to “cash out” is driven by third-party services The authors argue that the ability to “cash out” is an important differentiation between gambling and non-gambling activities. However, their evidence for loot boxes meeting this criteria is entirely reliant on the use of third-party sites where players sell/trade virtual goods. It is important to note that the ability to “cash out” is through the use of third-party (non-affiliated) websites, not through in-game features. Most game developers are not creating these items with the intention for them to be sold/traded for “real world” money or goods. The terms and services of many video games explicitly state that the items should not be sold or traded in secondary markets. There is a complex legal question here centred on agency – and that should be given due recognition by the authors.

Are we talking about loot boxes; or are we talking about secondary markets? There is conceptual confusion here - on the one hand, the authors argue that the definition of gambling should be broadened to include loot boxes; yet they focus much of their argument the existence of secondary markets to generate a classic “cash out” definition of gambling. Concerns of secondary markets (e.g., “third-party” websites) are not unique to the loot box controversy – they are not the silver bullet that mean loot boxes are gambling in contrast to a myriad of other in-game mechanics found in video games.

No countervailing consideration The methodology utilized within lacks systematic rigor as the authors fail to explore any countervailing considerations as to why games may not satisfy the Griffiths (1995) criteria. Without this, their methodology is a simple box checking exercise rather than a true analysis of each particular game.

Lack of scientific support On page 3, the authors state, “The literature on those at risk of developing problematic gambling behaviours suggests that the video-gaming population might be a ripe breeding ground for such issues”. The authors not only provides no support for this claim, but fail to recognise that some research in this area has actually suggested the opposite: that online games can be used as avenues to reduce gambling cravings and activities among disordered gamblers (see Gainsbury et al., 2015; Hollingshead et al, 2016; Parke et al., 2013; Wohl et al., 2017).

Concluding Thoughts As debates surrounding the use of loot boxes continue to unravel across the globe, we must stay up-to-date with the latest research about their uses and implications, especially among younger players. However, the analysis by Drummond and Sauer is being interpreted as scientific evidence that loot boxes constitute gambling and need to be regulated. The limitations of this paper cannot be understated. An analysis of a small handful of video games against out-dated and vague criteria does not constitute scientific evidence.

The repercussions of publishing an unsystematic and under-referenced article such as this one in a well regarded scientific journal cannot be understated. Shortly after publication, Senator Jordan Steele-John of the Australian Greens Party used this article as evidence to justify an inquiry into loot boxes in the Australian senate. Specifically, he presented a motion pertaining to the use of loot boxes in video games, whether they constitute gambling and whether they are appropriate for younger audiences. In supporting the motion in the Senate he said “(ii) that a paper published in Nature Human Behaviour on 18 June 2018, entitled ‘Video game loot boxes are psychologically akin to gambling,’ recommended that games that use loot boxes ‘appear to meet both the psychological and legal definitions of gambling’ and that ‘ratings agencies and gambling regulatory bodies [should] consider restricting access to people of legal gambling age.’” The motion was supported unanimously by the Senate with no debate or vote required. The Senate has referred the matter to the Environment and Communications References Committee for inquiry and report by 17 September 2018.

Without maintaining rigorous standards, articles such this may promulgate moral panic rather than balanced discussion of video games. Future research in this area should be more systematic in their analyses, including the assessment of a wider range of video games (across genre) and consideration for different types of loot boxes (strategic verse cosmetic). It is suggested that researchers in this area should begin to consider developing new comparison criteria all together in fitting with our fast changing conceptualisation of video games and their associated risks - Griffiths’ (1995) criteria was a ground-breaking step in assessment of generalized gambling behaviour; but we should avoid doing it disservice through haphazard application to the study of loot boxes in video games.


Aarseth, E., Bean, A. M., Boonen, H., Colder Carras, M., Coulson, M., Das, D., … & Haagsma, M. C. (2017). Scholars’ open debate paper on the World Health Organization ICD-11 Gaming Disorder proposal. Journal of Behavioral Addictions, 6(3), 267-270

Drummond, A., & Sauer, J. D. (2018). Video game loot boxes are psychologically akin to gambling. Nature Human Behaviour, 1.

Espineli, M. (2016, November 30). Game Release Dates 2016. Retrieved from

Gainsbury, S. M., Hing, N., Delfabbro, P., Dewar, G., & King, D. L. (2015a). An exploratory study of interrelationships between social casino gaming, gambling, and problem gambling. International Journal of Mental Health and Addiction, 13, 136–153. doi:10.1007/s11469-014-9526-x

Ghuman, D., & Griffiths, M. (2012). A cross-genre stud of online gaming: Player demographics, motivation for play, and social interactions among players. International Journal of Cyber Behavior, Psychology, and Learning, 2(1), 13 – 29.

Griffiths, M. (1995). Adolescent Gambling. New York: Routledge.

Hollingshead, S. L., Kim, H.S., Wohl, M.J.A., & Derevensky, J. L. (2016). The Social Casino Gaming-Gambling Link: Motivation for Playing Social Casino Games Determines Whether Self-Reported Gambling Increases or Decreases Among Disordered Gamblers. Journal of Gambling Issues, 33, 52 – 67. doi: 10.4309./jgi.2016.33.4

Kowert, R. (2015). Video Games and Social Competence. New York: Routledge.

Parke, J., Wardle, H., Rigbye, J., & Parke, A. (2013). Exploring social gambling: Scoping, classification and evidence review. Final report submitted to the UK Gambling Commission. Retrieved from 20Gambling.pdf

Skoric, M. M., Teo, L.L.C., & Neo, R. L. (2009). Children and Video Games: Addiction, Engagement, and Scholastic Achievement. Cyberpsychology & Behavior, 12(5), 567 – 572.

Wohl, M. J., Salmon, M. M., Hollingshead, S. J., & Kim, H. S. (2017). An examination of the relationship between social casino gaming and gambling: The bad, the ugly, and the good. Journal of Gambling Issues, (35).

Wolf, M. J. (2001). The medium of the video game. University of Texas Press.