One randomized controlled trial of 110 nursing home residents with dementia found that mahjong players scored 4.5 points higher on the Mini-Mental State Examination than the control group by 9 months. That is the strongest single result in this literature. It is also close to the only randomized evidence that exists, and it was conducted on Chinese mahjong in Hong Kong.
Mahjong is regularly described as good for the brain, usually without a citation. The research is real but narrower than the claims made about it, and it is worth knowing exactly where the line sits. Everything below cites a specific published study, and every limitation is stated. This article is not medical advice, and nothing here should be read as a claim that mahjong prevents or treats dementia.
Does Playing Mahjong Improve Brain Health?
The honest answer is that mahjong appears to help maintain cognitive function in older adults, that the effect is best documented in people who already have cognitive impairment, and that most of the supporting evidence is observational rather than experimental.
A 2024 scoping review published in the journal Gerontology and Geriatric Medicine, titled "Does Playing Mahjong Benefit Older Individuals?", gathered the available studies and reached a cautious conclusion: the correlational design of most of them means the underlying neural mechanism is still unclear, and more randomized controlled trials are needed to establish effects on higher-level cognitive functioning. When the review paper written specifically to assess this question says the evidence needs strengthening, that is the appropriate starting point.
Nearly all of the mahjong research is conducted in China, Hong Kong, or Taiwan, on Chinese mahjong, with participants who have played the game for decades. No published randomized trial has studied American mahjong specifically. That gap matters and it is not usually mentioned in articles on this subject.
What Does the Only Major Randomized Trial Show?
The strongest study is Cheng and colleagues, published in the American Journal of Geriatric Psychiatry in January 2014. It is the first randomized controlled trial to test mahjong as a therapeutic activity.
The design was a cluster-randomized open-label trial across nursing homes. Participants were 110 residents with at least very mild dementia, scoring 10 to 24 on the Mini-Mental State Examination, most of whom had never taken cholinesterase inhibitors. Homes were randomized into three conditions: mahjong, tai chi, and simple handicrafts as the control. Activities ran three times a week for 12 weeks.
The results are worth reading carefully:
| Group | MMSE difference vs control at 9 months | 95% confidence interval | Effect size |
|---|---|---|---|
| Mahjong | 4.5 points | 2.0 to 6.9 | d = 0.48 |
| Tai chi | 3.7 points | 1.4 to 6.0 | d = 0.40 |
The pattern the authors describe is the important part. Control participants deteriorated over time while the mahjong and tai chi groups held steady, so the gap between them widened as the study went on. The benefit was preservation of function rather than improvement of it.
The trial also found no treatment effect at all on immediate recall or backward digit span. The authors' own conclusion is measured: mahjong and tai chi can preserve functioning or delay decline in certain cognitive domains, even in people with significant cognitive impairment. Certain domains, not all of them.
What this study does not establish
It was conducted on nursing home residents with diagnosed dementia. It says nothing directly about whether a healthy 55 year old who takes up mahjong will have a different cognitive trajectory at 75. Extrapolating from a dementia population to a healthy one is the most common error made when this study is quoted.
Is Mahjong Linked to Lower Rates of Cognitive Impairment?
In observational data, yes, and consistently. A study published in the Journal of Aging and Physical Activity in 2022 recruited 489 healthy older adults and 187 with mild cognitive impairment, and examined mahjong playing alongside leisure physical activity.
After adjusting for a range of covariates, years of mahjong playing was associated with reduced odds of having mild cognitive impairment, with an odds ratio of 0.595 and a 95% confidence interval of 0.376 to 0.961. Physical activity showed a comparable association at an odds ratio of 0.572. The two also interacted, producing a combined effect larger than either alone, which led the authors to suggest that combined cognitive and physical interventions may outperform either on its own.
Related work points the same direction. Wang, Liu and Zhao published an analysis of playing cards or mahjong and cognitive function in Chinese older adults in the International Journal of Environmental Research and Public Health in 2022. A broader systematic review and meta-analysis by Yates, Ziser, Spector and Orrell in International Psychogeriatrics in 2016 examined cognitive leisure activities generally and future risk of cognitive impairment and dementia.
Every one of these is an observational finding, and observational findings cannot separate cause from selection. People who play mahjong weekly at 75 may have better cognition because they play, or they may play because they already have the cognition, mobility, hearing, and social network required to sit at a table for three hours. This is called reverse causation and it is the central unresolved problem in the field.
Does Mahjong Help With Depression in Older Adults?
There is some evidence here, and it is arguably better supported than the cognitive claim. Cheng and colleagues published a study in the American Journal of Geriatric Psychiatry in 2012 finding that leisure activities alleviated depressive symptoms in nursing home residents with very mild or mild dementia. In discussing the later trial, the same research group noted that mahjong, but not tai chi, showed short-term effects on reducing depressive symptoms in that population.
This is intuitive in a way the cognitive claim is not. Mahjong is four people at a table for several hours, talking. Isolation is a well-established risk factor for depression in older adults, and a standing weekly game is a structural fix for isolation. Whether the benefit comes from the tiles or from the company is not something these studies can separate, and it may not matter much practically.
Why Might Mahjong Do Anything at All?
The proposed mechanism is that mahjong loads several cognitive systems at once, rather than exercising one in isolation the way a crossword does.
- Working memory. Holding 13 tiles, a target hand, and a mental record of what has been discarded.
- Attention and inhibition. Tracking three opponents' discards while managing your own hand.
- Executive function. Deciding when to abandon a hand that is not coming together.
- Visuospatial processing. Reading tile faces and matching them against a printed card.
- Social cognition. Reading three other people in real time.
- Motor engagement. Handling, pushing, and racking tiles for hours.
American mahjong adds a further load that most variants do not have. Because the National Mah Jongg League issues a new card of legal hands every spring, players cannot rely on patterns memorized in previous years. The task is genuinely re-learned annually rather than practiced into automaticity, and novelty is one of the features cognitive reserve research tends to favor. This is a plausible argument rather than a tested one; no study has compared American mahjong to Chinese mahjong on this basis.
Does American Mahjong Count, or Only the Chinese Game?
Nobody knows, because nobody has studied it. The published trials and cohorts all concern Chinese mahjong played by Chinese-speaking populations. American mahjong differs in tile count, in the Charleston passing ritual, in the use of jokers, and in the annual card system.
The reasonable position is that the cognitive demands overlap heavily and the social structure is identical, so the findings probably transfer in broad terms. But "probably transfers" is not a research finding, and Land of Mahjong will not claim otherwise. Anyone citing this literature as evidence for American mahjong specifically is extending beyond what has been measured.
How Often Would You Have to Play?
The Cheng trial used three sessions per week for 12 weeks. The observational studies measure years of playing rather than sessions per week, which suggests that duration over time matters more than intensity in any given month.
Practically, a standing weekly game is the format that the research population actually lived, and it is the format that lasts. The obstacle is almost never enthusiasm; it is logistics, which is the subject of the guide to hosting a mahjong night. A group that meets once and never rebooks accrues no benefit of any kind.
What the Research Does Not Show
Stating this plainly, because it is routinely blurred:
- Mahjong has not been shown to prevent dementia.
- Mahjong has not been shown to reverse cognitive decline.
- No trial has tested American mahjong.
- No trial has tested healthy middle-aged adults.
- The randomized evidence base is essentially one trial with 110 participants.
- Observational associations cannot rule out that healthier people simply play more.
What can fairly be said is this: in older adults with dementia, structured mahjong sessions three times weekly preserved global cognitive function relative to a control activity over nine months in one randomized trial, and mahjong playing is consistently associated with lower odds of mild cognitive impairment in observational studies. That is a genuine finding and a modest one.
If you are considering mahjong for a parent or for yourself on cognitive grounds, discuss it with a physician rather than with a blog. If you are considering it because four people sitting at a table every week is good for a person, the evidence for that is thinner in the journals and fairly obvious in life.
The Studies Cited in This Article
- Cheng ST, Chow PK, Song YQ, Yu ECS, Chan ACM, Lee TMC, Lam JHM. Mental and physical activities delay cognitive decline in older persons with dementia. American Journal of Geriatric Psychiatry. 2014 Jan;22(1):63 to 74. PMID 23582750.
- Cheng ST, Chan ACM, Yu ECS. An exploratory study of the effect of mahjong on the cognitive functioning of persons with dementia. International Journal of Geriatric Psychiatry. 2006;21(7):611 to 617. PMID 16779765.
- Cheng ST et al. Leisure activities alleviate depressive symptoms in nursing home residents with very mild or mild dementia. American Journal of Geriatric Psychiatry. 2012;20(10):904 to 908.
- Mahjong playing and leisure physical activity alleviate cognitive symptoms in older community residents. Journal of Aging and Physical Activity. 2022;30(1). PMID 34388703.
- Wang J, Liu N, Zhao X. Association of playing cards or mahjong with cognitive function in Chinese older adults. International Journal of Environmental Research and Public Health. 2022;19(15):9249.
- Does playing mahjong benefit older individuals? A scoping review. 2024. PMID 39350382.
- Yates LA, Ziser S, Spector A, Orrell M. Cognitive leisure activities and future risk of cognitive impairment and dementia: systematic review and meta-analysis. International Psychogeriatrics. 2016 Nov;28(11):1791 to 1806.
- Verghese J, LeValley A, Derby C, et al. Leisure activities and the risk of amnestic mild cognitive impairment in the elderly. Neurology. 2006 Mar 28;66(6):821 to 827.
- Liu X, Ruan Y, Huang L, et al. Cognitive leisure activity and all-cause mortality in older adults: a 4-year community-based cohort. BMC Geriatrics. 2021;21(1):236. PMID 33836660.
Shop Starter Kits
The Weekly Game Starts With One Box
Land of Mahjong starter kits include 160 engraved acrylic tiles, four racks with pushers, an 80cm neoprene mat, dice, and a carry tote, so a new group can sit down and play the same week.