Last time the lever was belief. Teenagers who had never gambled were more likely to start if they believed more of their peers gambled, and correcting what students believed about their peers cut how much they lost. This entry follows the question into games.
A loot box is a paid item in a video game with a random reward inside. You pay first and find out what you got after. The question is whether buying them leads people into gambling, or whether the two tend to show up in the same people.
Start with gamers who didn't gamble
In December 2020, researchers at the University of British Columbia surveyed gamers aged 18 to 25 in the UK, North America and Australia through an online research panel, and asked again six months later. About six in ten were men. Of the 291 who did not gamble at the start and answered both times, 33 had started gambling by the second survey, about one in nine.
The more someone had spent on loot boxes in the year before, the more likely they were to be one of the 33. Spending on ordinary in-game purchases, where you pay for an item you can see, didn't predict starting once loot box spending was counted. What predicted starting was paying for a random reward, not spending money in games.
Like the Spanish study in the last entry, this one starts with people who weren't gambling, so their gambling can't explain their spending. It shows what came first. It doesn't show why, and something else could drive both. The main test was registered before the data came in. The research centre that ran it is supported by the British Columbia Lottery Corporation, a government-owned gambling operator.
A different kind of study points the same way. In an Australian survey of 1,026 gamers aged 18 to 25, 59% of them women, people recalled the age at which they first did each thing. Paid loot boxes tended to come before sports betting, casino games, esports betting and skin gambling, though after scratch-cards and lottery tickets.
The repeat was too small to tell
The same lab tried again with gamers aged 19 to 25 in British Columbia, recruited through Reddit and an esports Discord. After removing responses that looked like bots or careless answers, 54 non-gamblers were followed for six months and seven started gambling. With so few, the standard test was not significant. A second kind of analysis, which used the first study's results as its starting point, did support the link. Same lab, same funder, so it is not an independent check.
The other direction
The first study also ran the question backwards. Among gamers who had not bought loot boxes, those who held more of the beliefs common among gamblers, such as thinking they can predict or control chance outcomes, spent more on loot boxes six months later. So did those who spent more on ordinary in-game purchases. The authors label this analysis exploratory.
That is a second belief, and it is not about friends. It is about luck and control, and it came before the loot boxes, not after.
A 2021 review that pooled 15 studies found that people who spend more on loot boxes tend to have more symptoms of gambling problems. Its authors judged the reverse explanation, people with gambling problems buying more loot boxes, "equally, if not more likely." The studies it pooled were correlational, so none of them could say which came first.
And gambling problems?
Starting to gamble is not the same as being harmed by it. Two studies have followed loot box buyers to see whether problems came later.
In Britain, 2,080 young people aged 16 to 24, drawn from an online panel, were surveyed in 2019 and again a year later. Buying loot boxes at the first survey did not predict problem-gambling scores at the second. But in that analysis no single form of gambling predicted later scores, so it is not evidence that loot boxes are harmless. It was funded by Wellcome, the medical research charity.
In Spain, 2,213 students aged 11 to 17 at seven schools in Valencia were surveyed in December 2020 and again in May 2021. Those who had bought loot boxes were a little over twice as likely to be gambling online six months later; for boys, about twice. That count includes teenagers who were already gambling at the start, and nothing else was accounted for. On a questionnaire that screens for online gambling disorder, buyers were more likely to screen positive later, but the link was clear for girls and not for boys. Only 25 of the 2,213 screened positive, and the authors say this part needs further study. The study was funded by the university that led it.
The one study here that reports young men as a group asked once, not twice. In New South Wales, Australia, in 2019 and 2020, among 335 men aged 18 to 24 from an online panel, those who had bought loot boxes in the past year had higher problem-gambling scores. The link was a little stronger among young women. In the two studies here that split their results by sex, this one and the Spanish one, it was stronger for girls and women. That is a pattern I see across two studies, not something either of them tested.
Where friends come in
So does the belief about peers carry into games? In a survey of 1,472 students aged 11 to 17 at 13 Flemish secondary schools, taken in 2021 and 2022, teenagers who bought loot boxes felt more pressure toward gambling from the people around them. They were more likely to agree that most people important to them gamble and would approve of them gambling. It is the same kind of measure the last entry rested on. The pressure was low overall, and the survey was taken once, so it can't say which came first.
In interviews with 28 loot box buyers in the UK aged 18 to 56, two thirds of them men, people described buying for status, bragging and keeping up. One put it this way: "I'm then comparing myself to him, because he's got it and I don't, so I want to get something."
Followed for a year, the pressure wasn't the route
The same Flemish researchers followed 561 students aged 10 to 17, 43% of them boys, for a year, from 2021 to 2023. Taking part in gambling-like activities in and around games, from paying for loot boxes and spinning prize wheels to watching others gamble on stream, predicted more real-money gambling a year later. It did not run the other way.
What carried it was attitude. The teenagers who did more of these things saw gambling more favourably, and that carried through to gambling a year later. The pressure they felt from the people around them did not.
That complicates the last entry's story without undoing it. That study measured starting, among teenagers who had never gambled. This one measured gambling more, and counted seven activities together, so loot boxes on their own can't be picked out. Most of the real-money gambling was scratch-cards and small bets among friends.
Which way does it run?
One more study turns the question round. At one university in Singapore, 252 students, 71% of them women, answered a short survey every week for 13 weeks. Weeks when they worried more about missing out on friends ("I felt worried when I found out my friends were having fun without me") were followed by more risky loot box buying. Weeks of more risky buying were followed by more of that worry, and that second link was the larger, though still moderate.
That raises a question this series hasn't asked yet: does what a young man spends change how he sees his friends, as well as the other way round? It is my inference from one study, mostly of women, at one university, which didn't check whether the students played games at all. Later entries take it up.
Where this leaves the question
For starting, the best evidence puts loot boxes first. Among young gamers who didn't gamble, more loot box spending predicted who started, and ordinary in-game spending didn't once loot boxes were counted. The one repeat was too small to confirm it.
It also runs the other way. In an exploratory analysis, beliefs about luck and control came before loot box buying in the same people, and a review of 15 studies judged the reverse at least as likely.
For problems, the two studies that followed buyers didn't find the link in British young people or in Spanish boys. Neither rules it out: the follow-ups were short and the numbers with problems small.
The belief about friends is there in loot box buyers, but followed over a year it wasn't what carried play into gambling. Attitude was. Last time the lever was what young people believe about others. This time it looks more like what they come to believe about gambling itself.
Sources and how I checked
Every number in this entry was opened at its original source and checked against it again before the entry was written. The table lines up what AI search summaries said against what the sources actually say.
| What the search said | What the source says | How I checked |
|---|---|---|
| “A 2025 replication in British Columbia found that increased spending on loot boxes, but not direct-purchase microtransactions, at baseline predicted an increased likelihood of initiating conventional gambling at the 6-month follow-up” | The failed test was left out. In the replication's standard test the link was not significant: seven people started gambling. Only a second analysis that used the first study's results as its starting point supported it. | Opened at source (full text), 2026-11-01 |
| “There was no evidence that earlier experiences with loot boxes predict later gambling problems” | Right null, wrong design. The study was a one-time survey that compared how many years people had used loot boxes with their gambling now. Nothing was measured later. | Opened at source (full text (via summary of the PMC full text)), 2026-11-01 |
| “Loot box purchasing did not increase PGSI scores in young people aged 16 to 26” | Right as far as it goes. In the year-long analysis, loot box buying didn't predict later problem scores, but no single form of gambling did. Among those already gambling, loot box buying went with higher scores. | Opened at source (full text (PMC author manuscript)), 2026-11-01 |
Sources
- Brooks, G.A. & Clark, L. (2023). The gamblers of the future? Migration from loot boxes to gambling in a longitudinal study of young adults. Computers in Human Behavior 141:107605. doi 10.1016/j.chb.2022.107605
Measured on: Online-panel (Prolific) gamers in the UK, North America and Australia who did not currently gamble at baseline; 62.4% men · 18-25 (two aged 26) · UK, USA/Canada, Australia · What I read: Opened at source (full text (published version, CHB 141, open access CC BY-NC-ND) and accepted manuscript), 2026-11-01 · Funding: OPERATOR-FUNDED: Centre for Gambling Research at UBC is supported by the Province of BC and the British Columbia Lottery Corporation (BCLC, a Crown gambling operator); LC also NSERC Discovery Award. - Russell, A.M.T., Hing, N., Newall, P., Greer, N., Dittman, C.K., Thorne, H. & Rockloff, M. (2023). Order of first-play in simulated versus monetary gambling. Journal of Behavioral Addictions 12(4):992-1005. doi 10.1556/2006.2023.00069
Measured on: Video gamers in an Australian online panel; 59.1% women, 40.2% men (Table 2) · 18-25 · Australia · What I read: Opened at source (full text), 2026-11-09 · Funding: Victorian Responsible Gambling Foundation, Round 10 Early Career Researcher grant (GR10/19/03); funder had no role in the study. - Palmer, L., Brooks, G.A. & Clark, L. (2025). A longitudinal replication study testing migration from video game loot boxes to gambling in British Columbia, Canada. BMC Psychology 13:459. doi 10.1186/s40359-025-02766-1
Measured on: Gamers playing 3+ hours a week in British Columbia, recruited via Reddit and an eSports Discord; non-gamblers 75.9% men · 19-25 · Canada · What I read: Opened at source (full text), 2026-11-01 · Funding: OPERATOR-FUNDED: Centre for Gambling Research at UBC is supported by the Province of BC and the British Columbia Lottery Corporation (BCLC, a Crown gambling operator); LP holds a BCLC-supported fellowship; GB BCLC speaker honoraria; LC paid by gambling-related bodies incl. ICRG and GambleAware, and BCLC legal consultancy. - Garea, S.S., Drummond, A., Sauer, J.D., Hall, L.C. & Williams, M.N. (2021). Meta-analysis of the relationship between problem gambling, excessive gaming and loot box spending. International Gambling Studies. doi 10.1080/14459795.2021.1914705
Measured on: Mixed: mostly adult gamers, some adolescent samples, across the included studies · mixed, not pooled by age · multiple · What I read: Opened at source (full text (accepted manuscript)), 2026-11-01 · Funding: Marsden Fund Council from Government funding (Royal Society Te Aparangi), MAU1804. No gambling-industry funding named. - Wardle, H. & Tipping, S. (2023). The relationship between problematic gambling severity and engagement with gambling products: Longitudinal analysis of the Emerging Adults Gambling Survey. Addiction 118(6):1127-1139. doi:10.1111/add.16125
Measured on: YouGov online panel, Great Britain; sex a covariate only · 16-24 at wave 1 (16-26 across waves; mean 20.5) · UK (Great Britain) · What I read: Opened at source (full text (PMC author manuscript)), 2026-11-01 · Funding: Wellcome (Grant 200306). Author declares one GambleAware-funded study (gambling and suicide) and past paid advisory role to the Gambling Commission - Gonzalez-Cabrera, J., Basterra-Gonzalez, A., Ortega-Baron, J., Caba-Machado, V., Diaz-Lopez, A., Pontes, H.M. & Machimbarrena, J.M. (2023). Loot box purchases and their relationship with internet gaming disorder and online gambling disorder in adolescents: A prospective study. Computers in Human Behavior 143:107685. doi:10.1016/j.chb.2023.107685
Measured on: Students at seven schools in Valencia province who completed both waves; 48.9% boys · 11-17 (mean 13.89) · Spain · What I read: Opened at source (full text (published PDF, open access)), 2026-11-01 · Funding: Universidad Internacional de la Rioja (UNIR) 'Cyberpsychology' project; open access paid by UPV/EHU. Authors declare no competing interests - Rockloff, M., Russell, A.M.T., Greer, N., Lole, L., Hing, N. & Browne, M. (2021). Young people who purchase loot boxes are more likely to have gambling problems: An online survey of adolescents and young adults living in NSW Australia. Journal of Behavioral Addictions 10(1):35-41. doi 10.1556/2006.2021.00007
Measured on: Online-panel (Qualtrics) adolescents and young adults living in NSW; 59.9% female overall · 12-24 (adults 18-24) · Australia · What I read: Opened at source (full text (via summary of the PMC full text)), 2026-11-01 · Funding: NSW Government Responsible Gambling Fund, supported by the NSW Office of Responsible Gambling. Authors declare no conflicts. - Grosemans, E., De Cock, R., Bradt, L. & Zaman, B. (2024). More than loot boxes: the role of video game streams and gambling-like elements in the gaming-gambling connection among adolescents. Journal of Gambling Issues 54:68-97
Measured on: Students from 13 Flemish secondary schools who completed all gambling items; 50.7% boys · 11-17 (mean 14.02) · Belgium · What I read: Opened at source (full text), 2026-11-01 · Funding: Research Foundation Flanders (FWO-SBO S006821N), government - Close, J. & Lloyd, J. (2021). Lifting the Lid on Loot-Boxes: Chance-Based Purchases in Video Games and the Convergence of Gaming and Gambling. GambleAware (commissioned report)
Measured on: UK loot box buyers recruited purposively; 19 men, 9 women · 18-56 (mean 28.9) · UK (England, Scotland, Wales, Northern Ireland) · What I read: Opened at source (full text), 2026-11-01 · Funding: GambleAware (commissioned) - Grosemans, E. (2024). Gam(e)(a)ble: doctoral dissertation, KU Leuven (Institute for Media Studies). Chapter 6: From simulated gambling in video games to monetary gambling: Testing a pathway model among adolescents
Measured on: Flemish secondary-school students who answered both waves and passed attention checks; 43.3% boys, 85.9% academic track · 10-17 at wave 1 (mean 13.78, SD 1.42) · Belgium · What I read: Opened at source (dissertation chapter (pre-publication version of the published article); published article not read), 2026-11-01 · Funding: Research Foundation Flanders (FWO), Gam(e)(a)ble SBO project (government) - Grosemans, E., De Cock, R., Bradt, L. & Soenens, B. (2025). From gambling-like elements in and around video games to monetary gambling: testing a pathway model among adolescents. International Gambling Studies. doi:10.1080/14459795.2025.2567896
Measured on: Flemish secondary-school students who answered both waves and passed attention checks; 43.3% boys, 85.9% academic track · 10-17 at wave 1 (mean 13.78, SD 1.42) · Belgium · What I read: Not opened: the published article is paywalled; its abstract was read at the publisher (2026-11-01) and its full content in the pre-publication dissertation chapter listed above · Funding: FWO (per Crossref funder field) - Hartanto, A. & Kasturiratna, K.T.A.S. (2024). Longitudinal bidirectional relation between fear of missing out and risky loot box consumption: Evidence for FoMO-Driven loot boxes spiral hypothesis. Computers in Human Behavior Reports 16:100535. doi:10.1016/j.chbr.2024.100535
Measured on: Students at one Singapore university recruited through a subject pool for a larger wellbeing study; 180 of 252 women (71%); NOT screened for gaming or loot box use · mean 21.83 (SD 1.75) · Singapore · What I read: Opened at source (full text (published PDF)), 2026-11-01 · Funding: Singapore Ministry of Education Academic Research Fund Tier 1 (via Singapore Management University); authors declare no conflicts
Corrections
None yet. If something here changes, the change is logged here with its date.
How this was made
This series is researched with an AI desk: it searches, reads, pulls the numbers and drafts, including this entry. I edit and decide what is published.
How I check sources · The research library
If gambling is costing you or someone close to you more than it should, the US National Problem Gambling Helpline is free and confidential: call or text 1-800-MY-RESET (1-800-697-3738), or chat at ncpgambling.org.