The DSM-5 formally added Internet Gaming Disorder to its diagnostic appendix in 2013 — the first time gaming behaviour was treated as a candidate clinical category. The research base has accumulated since then, and the documented endpoints are specific enough to map directly to design decisions at the operator level.
The loot-box problem
A "loot box" is a randomised in-game purchase: the buyer pays for a chance at a high-value reward, with most outcomes delivering low-value common items. The mechanism is structurally identical to a slot machine pull. A 2019 Frontiers in Psychiatry paper found statistically significant correlations between loot-box engagement and problem-gambling indicators in older-adolescent cohorts.
The regulatory response varies: Belgium and the Netherlands have classed certain loot-box formats as illegal gambling; the US and Canadian regulatory positions remain inconsistent across jurisdictions. The industry-side preferred classification is "surprise mechanic" — explicitly distinct from gambling terminology — which the EA executives' public framing has emphasised.
The displacement pattern
A 2024 PLOS One paper documenting heavy-gaming cohorts found statistically significant displacement of in-person social contact and elevated reported loneliness at sustained-use thresholds. Sub-threshold use (a few hours per week) didn't show the displacement pattern; the effect appears above ~2 hours daily.
A 2023 Frontiers in Psychiatry review aggregated the associations between gaming disorder and anxiety, depression, and ADHD diagnoses — bidirectional in some cases (the conditions and the gaming behaviour interact rather than running in one causal direction). A 2019 Aggressive Behavior meta-analysis documented elevated aggression measures in violent-video-game user cohorts, particularly in adolescent samples.
The harassment baseline
The ADL's 2021 Hate Is No Game survey found 83% of online multiplayer gamers reported experiencing harassment within the previous six months — sexist, racist, anti-LGBTQ, and identity-targeting verbal harassment normalised within the operating environment.
Beyond the harassment itself, the research base documents a moral disengagement pattern — sustained immersion in operations where in-game harm has no real consequence weakens the user's sensitivity to comparable behaviour in actual social contexts. This isn't the older "video games cause violence" debate; it's a narrower documented effect on social norms within and around the gaming community.
Three focus tools that actually work
Telling someone with a gaming addiction to stop is not a useful intervention. Three pragmatic tools that operate at the access-friction level:
Cold Turkey — full website and app blocking; timer-locked once initiated, with no undo until expiry.
Freedom — cross-device blocking with schedulable sessions; iOS, macOS, Windows, Android, Chrome.
FocusMe — the high-friction option; documented as the hardest to bypass at a technical level.
What follows
The relationship between gaming engagement and life outcomes is not a simple causal story. Moderate gaming is consistent with healthy functioning. The structural problem is at the engagement-engineering layer — the operators are optimising the same dopamine variable as the slot-machine industry, with substantially fewer disclosure and harm-reduction obligations than equivalent gambling operations.
For adjacent reading, see our piece on the broader attention-economy pattern, our look at mental-health access infrastructure, and our piece on Google's data-collection model.