Mahjong Attention Training: Four Walls, Zero Mercy, Real Data

The Oasis Health Journal · Submitted August 5, 2026 · 4:30 PM EDT

By Ray 'Sleeves' Okafor · Edited by Mona Trell

They took fifty-six elderly people with mild cognitive impairment. Made half of them play mahjong three times a week for twelve weeks. The other half did nothing special. After three months, the mahjong group scored better on executive function tests. The control group did not improve.

That is the whole finding. No drugs. No supplements. No fancy equipment. Just tiles on a table and mahjong attention training that actually moved the needle on cognitive tests designed to catch early decline.

Now, I have seen a lot of things sold as brain training board games. Apps that make you match colours for nineteen dollars a month, auto-renewing, good luck cancelling that. Puzzles with thirty pieces and a motivational quote on the box. Subscription services that will improve your processing speed and your credit card statement at the same time.

Mahjong costs thirty dollars one time and then you own it forever. Forever! You cannot subscription-model a box of tiles.

What They Measured and Why It Matters More Than You Think

The study out of Nanchong, China used three tests. The Montreal Cognitive Assessment. The Shape Trail Test. The Functional Activities Questionnaire. All three measure executive function, which is the part of your brain that plans, switches tasks, and keeps track of four things at once without losing one.

Executive function is the first thing that slips when cognitive impairment starts. You forget why you walked into a room. You lose the thread of a conversation. You put the milk in the cabinet and find it six hours later wondering what kind of day you having.

After twelve weeks, the mahjong players improved on all three tests. The control group, who just kept doing whatever they normally did, showed no change. None.

The effect size was not huge. But it was real, and it was consistent across every measure. That is rare. Most cognitive skill card games improve the one thing they train you to do and nothing else. You get better at the game. Your actual life stays exactly the same.

Mahjong is different because the game itself is four jobs at once.

Four Piles, Four People, One Deteriorating Attention Span

Here is what makes mahjong hard. You are not just playing your own hand. You tracking what everybody discarded, inferring what that means about what they kept, updating your mental model of all four hands every single turn, and adjusting your strategy in real time based on incomplete information.

That is not a memory test. That is an attention test. And it is relentless.

Most games that improve visual attention make you do one hard thing over and over until you get fast at it. Mahjong makes you do four hard things simultaneously and every decision you make changes the next four decisions for everybody else at the table. You cannot coast. You cannot phone it in. The game will notice.

A meta-analysis published in 2025 looked at twenty randomized controlled trials of leisure activity interventions for mild cognitive impairment. It included mahjong, poker, virtual reality games, and other recreational activities. Across 1,126 participants, leisure activities improved overall cognitive function, memory, attention, and quality of life compared to traditional cognitive rehabilitation training.

The effect was not limited to one test or one population. It showed up on the Montreal Cognitive Assessment, the Mini-Mental State Examination, the Digit Span Backward test, and the Activities of Daily Living scale. That is five different measures saying the same thing.

Traditional rehab is expensive, insurance-limited, and hard to sustain long-term. Mahjong is thirty dollars, infinitely replayable, and you can do it at a folding table in somebody's kitchen with three people who also don't want to pay for rehab.

What This Does Not Prove (But You Gonna Hear Anyway)

This is one study. Fifty-six people. One nursing home. No blinding, because you cannot fake-play mahjong without noticing. No placebo control, because what would that even be, sitting at a table moving tiles around randomly three times a week while pretending it's strategy?

It does not prove it reverses cognitive decline. It does not prove it works in people without impairment, or in younger adults, or in any population that was not represented in this sample.

What it proves is that in a group of seventy-four-year-olds with mild cognitive impairment, playing mahjong for twelve weeks improved their scores on tests of executive function, and the control group's scores did not improve.

That is it. That is the claim. Anything else you hear is marketing, and marketing don't need a control group to say whatever it wants.

Now, a 2025 survey out of China asked 250 older adults about their preferences for haptic feedback in mobile serious card games. Seventy-three percent said combining card games with haptic technology could reduce their reliance on visual information. Seventy-eight percent believed it would help improve cognitive abilities.

That is a preference survey, not an outcome study. What people believe a thing will do and what a thing actually does are two different questions. Only one of them involves a control group and published results.

Why I Think It Works (And Why That Don't Mean You Should Buy It From Me)

I think the reason traditional mahjong for cognitive exercise works is because it is a social game that punishes distraction immediately. You lose track of one discard pile for one turn and somebody just won with the tile you needed three turns ago. And now you gotta hear about it.

The feedback is instant, it is social, and it is mildly humiliating. Which means you pay attention next time.

You cannot half-play mahjong. You cannot check your phone. You cannot let your mind wander. The game will notice and it will punish you and the other three people at the table will remind you that you just threw the tile somebody called two seconds ago. They will remind you loud.

That kind of sustained, divided, socially reinforced attention is exactly the thing that declines in early cognitive impairment. And it is exactly the thing most brain training apps do not require.

An app will wait for you. An app will let you retry. An app does not care if you got distracted because an app does not have feelings or a mortgage.

The seventy-four-year-old woman across the table who just went out on your discard absolutely cares. She remembers. She gonna bring it up next week.

Now look. I sell games that improve visual attention. I got three mahjong sets in stock right now and I would be happy to sell you one. But I am gonna tell you the same thing I tell everybody: if you got a friend who already owns a set and knows how to play, start there. The best version of this intervention is free.

The second-best version is thirty dollars and four people who show up on time.

The worst version is an app that says it is based on mahjong principles and costs you nineteen dollars a month to play by yourself on a screen while the algorithm pretends to care whether you win.

This article is education and reporting on published research. It is not medical advice, and nothing here is intended to diagnose, treat, cure or prevent any disease. Talk to your own clinician about your own situation.

Sources

1. Haptic-Driven Serious Card Games for Older Adults: User Preferences Study, JMIR serious games (2025).
2. Leisure Activity Interventions on Cognition in Mild Cognitive Impairment Patients: A Meta-Analysis, Neuropsychiatric disease and treatment (2025).
3. Playing Mahjong for 12 Weeks Improved Executive Function in Elderly People With Mild Cognitive Impairment: A Study of Implications for TBI-Induced Cognitive Deficits, Frontiers in neurology (2020).