Gambling addiction has no single cause. It develops where four things overlap: gambling products engineered to keep play going, the way the brain learns from unpredictable rewards, individual risk factors such as family history, impulsivity, ADHD and depression, and an environment where gambling is available every hour from a phone. None of the four is sufficient on its own, and most people exposed to any one of them never develop a problem.
That combination is why the condition tracks the product and the situation as closely as the person. The same person can buy a weekly lottery ticket for a decade without harm, then get into trouble within months of installing a fast betting app. The diagnosis itself, and the nine criteria clinicians use, are covered in what gambling addiction is.
If gambling has brought thoughts of suicide or self-harm, treat it as an emergency: call your local emergency number, or find a free crisis line for your country at findahelpline.com. To talk about the gambling itself, the helplines for each country are listed in our helplines directory.
Why gambling is addictive: what the products do
Gambling products are built around a small set of features that keep play going. The World Health Organization, in its December 2024 fact sheet, describes high-intensity products such as electronic gaming machines and high-speed wagering platforms as misleading users with design features that encourage extended use. Six have been studied directly.
| Feature | What it does | Evidence |
|---|---|---|
| Unpredictable payouts (random-ratio reward) | Wins arrive on no fixed schedule, so no moment is an obvious place to stop | Rats on random-ratio schedules responded faster and paused far less after a reward than rats on fixed schedules (Laskowski et al., 2019) |
| Near-misses | Two jackpot symbols and a third just short | Rated less pleasant than full misses, yet they raised the desire to continue and recruited the same striatal and insular circuitry as real wins (Clark et al., 2009, Neuron) |
| Losses disguised as wins | A spin returns less than it cost, but the machine plays the sounds and animation of a win | Skin-conductance responses matched those to real wins and exceeded those to ordinary losses (Dixon et al., 2010, Addiction) |
| Speed and event frequency | Short gaps between bet and result compress hundreds of decisions into one hour | Faster games were rated more exciting by all gamblers, were most appealing to those with a gambling problem, and were linked to more wagers, longer sessions and difficulty stopping (Harris and Griffiths, 2018) |
| In-play betting | Turns one pre-match decision into a rolling series of bets | Australian online sports bettors who bet in-play scored higher on problem gambling severity than those who did not, and frequent users of bonus and money-back offers placed more impulse in-play bets (Gainsbury et al., 2020; Hing et al., 2018) |
| Illusion of control | Choosing numbers or stopping the reels makes chance feel steerable | All six reviewed instruments measuring gambling cognitive distortions separated pathological from non-pathological gamblers with large effects (Goodie and Fortune, 2013) |
Two honest limits apply. The reward schedule explains the pull of the product, not the disorder: in the rat study above, random-ratio training produced fast, persistent responding, but the animals failed the standard tests of addiction-like behavior, leading the authors to call such schedules motivating rather than addictive. Chasing is also more complicated than the folk version. It is one of the nine diagnostic criteria, yet in an analysis of 1.9 million sessions by 15,544 people on a Canadian provincial gambling site, players on most games returned more slowly after a losing session and faster after a winning one (Zhang et al., 2024). Chasing inside a session and returning across days are not the same behavior.
The arithmetic underneath is fixed: every product keeps a set percentage of everything staked, which is why longer play converges on loss (RTP and volatility).
What dopamine actually does
Dopamine is a learning signal, not a pleasure chemical. Recordings from dopamine neurons show a burst coding the discrepancy between predicted and actual reward, plus a slower signal that ramps up before an uncertain outcome and is largest when the chance of winning is around 50/50 (Fiorillo, Tobler and Schultz, 2003, Science). Gambling supplies that condition on demand: unpredictable outcomes, every few seconds.
The strongest human evidence that dopamine drives gambling behavior comes from Parkinson's disease: across 3,090 patients in the United States and Canada, impulse control disorders including problem gambling were two to three times more likely in those taking a dopamine agonist (Weintraub et al., 2010). Nothing in that evidence supports the idea that a person who gambles has a permanently broken reward system.
Individual risk factors, with honest sizes
| Risk factor | What the evidence shows |
|---|---|
| Family history and genetics | In an Australian twin study of 4,764 adults, genetic influences accounted for 49% of the variation in liability to disordered gambling, with no detectable effect of shared family environment and no difference between men and women (Slutske et al., 2010) |
| Impulsivity and behavioral undercontrol | In the Dunedin birth cohort, children assessed as undercontrolled at age 3 were more than twice as likely to show disordered gambling at 21 and 32, independent of IQ and family income (Slutske et al., 2012) |
| ADHD | Across studies of ADHD symptoms and gambling severity, the weighted mean correlation was r = .17: consistent, but modest (Theule et al., 2016) |
| Depression, anxiety, substance use and smoking | A meta-analysis of general-population surveys found all four markedly more common among problem and pathological gamblers; causation runs in both directions (Lorains, Cowlishaw and Thomas, 2011) |
| Childhood adversity | In a survey of 32,000 Japanese adults, a high adverse-childhood-experience score was associated with problem gambling after adjustment for 14 confounders; cross-sectional, so association rather than sequence (Miyahara et al., 2026) |
| Sex and life circumstances | WHO estimates that 11.9% of men and 5.5% of women worldwide experience some level of harm from gambling; separation, bereavement, poverty and discrimination raise vulnerability (WHO, December 2024) |
| An early big win | Usually offered as the explanation in hindsight. A four-wave longitudinal study of young adults tested eleven predictors including an early big win; only impulsiveness predicted how severity changed over time, and average severity fell rather than rose (Edgerton, Melnyk and Roberts, 2015) |
These are population-level associations, not verdicts about a person: many people carry several of these factors and never gamble problematically, and many who develop gambling disorder carry none. They are useful for deciding how much protection to put in place.
Gambling addiction and ADHD
ADHD and problem gambling occur together more often than chance, and the link is smaller than headlines suggest. The meta-analysis above puts the average correlation between ADHD symptoms and gambling severity at r = .17, with a stronger association in older samples; the Japanese survey found ADHD traits associated with problem gambling independently of autistic traits and childhood adversity. Both are cross-sectional: the two conditions travel together, which is not the same as ADHD causing gambling disorder.
Three things follow. Tell whoever treats either condition about the other, because screening for comorbidity changes the treatment plan. Build controls that do not depend on in-the-moment attention: a deposit limit set once, a blocker on every device, money that is not reachable at 2am. Expect uneven progress: in 1,305 adults treated for gambling disorder at a US virtual clinic, an ADHD diagnosis was associated with slower improvement over the first 12 weeks (McAlister et al., 2026). Slower is not the same as ineffective.
Teenagers and young people
Adolescent gambling problems are real and hard to measure. A systematic review of 44 studies worldwide since 2000 found that between 0.2% and 12.3% of young people met criteria for problem gambling, the spread driven mainly by different questionnaires, cut-offs and timeframes (Calado, Alexandre and Griffiths, 2017). A small but significant minority of teenagers are affected; any single headline percentage is an artifact of the instrument used.
Advertising is the environmental factor with the clearest evidence for this age group. An umbrella review of eight systematic reviews covering 74 studies found consistent support for a causal relationship between advertising exposure and more positive attitudes, greater intention to gamble and increased gambling activity, with a dose-response pattern and the strongest evidence among children, young people and those already at risk (McGrane et al., 2023). WHO says the same about sports sponsorship and social-media promotion.
For a parent, the useful moves are specific: look at what leaves the bank and app-store accounts rather than at mood alone, check whether an adult gambling account is being used by someone underage, and raise it without shaming. The markers are in signs of gambling addiction; the conversation itself is covered in how to help someone with a gambling problem.
Why "just stop" usually fails on its own
Stopping by intention alone puts a decision made once against a product available continuously, on the same phone used for work, messages and banking; WHO states that easy accessibility of gambling products increases the risk of uptake. The beliefs that sustain gambling do not dissolve at the moment of decision either: illusion of control and the gambler's fallacy separate problem from non-problem gamblers with large effects.
The measures with trial evidence behind them are structural. Cognitive behavioral therapy has the strongest evidence: a Cochrane review of 14 randomized trials with 1,245 participants found benefits on gambling symptoms and financial loss immediately after therapy, while noting that durability beyond a year is unknown and the trial quality was low (Cowlishaw et al., 2012). Venue self-exclusion also helps, though not for the reason people assume: in a follow-up of Missouri self-excluders up to ten years after enrollment, most had reduced their gambling, but half of those who tried to re-enter a casino managed it, and outcomes were better for those who paired exclusion with treatment or a support group (Nelson et al., 2010). Enrolling, plus a second layer of support, mattered more than enforcement.
What reduces the risk
Prevention means removing the conditions the product depends on: speed, access and credit.
- Set money limits before a session, not during one. Among 49,560 customers of one online operator, the most gambling-intense 10% who had set voluntary deposit limits were spending significantly less a year later than comparable players who had not (Auer, Hopfgartner and Griffiths, 2020). Mechanics by account type: how to set gambling limits.
- Stack more than one barrier. A deposit limit, a device blocker, a bank gambling block and national self-exclusion each fail differently, so together they cover each other. The layers are compared in responsible gambling tools.
- Keep borrowed money out of gambling. Great Britain banned credit-card gambling from 14 April 2020 across all online and offline products, with non-remote lotteries the only exemption (Gambling Commission). Elsewhere a bank card block does the same job.
- Do not gamble to fix money or mood. Gambling to escape distress, and returning to win back losses, are two of the nine diagnostic criteria.
- Know what the product returns. A slot keeps its house edge on every spin, so no session is ever due.
Three things that take ten minutes: set a deposit limit on every gambling account you hold, install a blocker on the phone and not only the laptop, and take the gambling self-test to see where your pattern sits.
Frequently asked questions
Is gambling addiction genetic?
Partly. An Australian twin study of 4,764 adults attributed 49% of the variation in liability to disordered gambling to genetic influences, with no measurable contribution from shared family environment, and the same estimate for men and women. Inherited liability is not destiny: it interacts with what products a person is exposed to and how often. A parent with a gambling problem raises a child's risk, but most children of problem gamblers do not develop one.
Is gambling addiction more common in people with ADHD?
Yes, but the association is modest. A meta-analysis found a weighted mean correlation of r = .17 between ADHD symptoms and gambling severity, and a nationwide Japanese survey found ADHD traits independently associated with problem gambling after adjusting for 14 confounders. Both are cross-sectional, so they show the two conditions travel together rather than proving that ADHD causes gambling disorder. If both apply to you, tell the clinician treating either one.
Why can't I stop gambling even when I want to?
Because a decision made once has to hold against a product available every hour, tuned to unpredictable rewards, near-misses and fast repeat play, and against beliefs about control and about being due that do not disappear on the day you decide to stop. That is why structural barriers, set in advance, outperform intention, and why cognitive behavioral therapy has the strongest trial evidence of any treatment. Wanting to stop and being unable to is one of the diagnostic criteria, not a character failure.
Can gambling addiction be prevented?
Risk can be reduced, though nothing removes it entirely. The measures with evidence behind them are limits set before gambling starts rather than during a session, multiple blocking layers, keeping credit and borrowed money out of gambling, and getting help early rather than after a crisis. At population level, WHO recommends restricting advertising and sponsorship, universal account registration with binding loss limits, and product safety measures such as maximum bet sizes.
Sources
- WHO fact sheet: Gambling (2 December 2024)
- Clark et al., Gambling near-misses enhance motivation to gamble and recruit win-related brain circuitry (Neuron, 2009)
- Dixon et al., Losses disguised as wins in modern multi-line video slot machines (Addiction, 2010)
- Harris and Griffiths, The impact of speed of play in gambling on psychological and behavioural factors: a critical review (2018)
- Goodie and Fortune, Measuring cognitive distortions in pathological gambling: review and meta-analyses (2013)
- Weintraub et al., Impulse control disorders in Parkinson disease: a cross-sectional study of 3090 patients (2010)
- Slutske et al., Genetic and environmental influences on disordered gambling in men and women (2010)
- Theule et al., Exploring the relationships between problem gambling and ADHD: a meta-analysis (2016)
- Calado, Alexandre and Griffiths, Prevalence of adolescent problem gambling: a systematic review (2017)
- Cowlishaw et al., Psychological therapies for pathological and problem gambling (Cochrane Database of Systematic Reviews, 2012)
Need help now? Free, confidential gambling helplines exist in most countries — see the helpline directory. In immediate danger, call your local emergency number.