Can Brain Games Slow Aging? What a 20-Year Study Found

For years, the honest answer to "do brain games keep you young?" was "not really proven". In 2026 that answer got more interesting. A 20-year follow-up of one of the largest brain-training trials ever run found that one specific kind of training was linked to fewer dementia diagnoses — and two other kinds were not. Here is what the study did, what it does and does not mean, and how to use that when you pick a puzzle game.

This article explains research for general interest. It is not medical advice; if you are worried about your memory or thinking, talk to a doctor.

The ACTIVE Trial

ACTIVE (Advanced Cognitive Training for Independent and Vital Elderly) enrolled 2,802 adults aged 65 and older in 1998–1999 and randomly assigned them to one of three kinds of training or to a control group:

• Memory — strategies for remembering word lists and stories.
• Reasoning — finding patterns in letter and number series.
• Speed of processing — computer exercises that ask you to identify and locate things on a screen faster and faster, getting harder as you improve.

Training was about ten one-hour sessions, and some participants received extra "booster" sessions later.

What the 20-Year Follow-Up Found

The 2026 analysis, published in Alzheimer's & Dementia: Translational Research & Clinical Interventions, linked 2,021 of the participants to their Medicare records from 1999 to 2019. In the control group, 48.7% received a diagnosis of Alzheimer's disease or a related dementia over those two decades. Participants who did the speed training plus boosters had about a 25% lower risk of such a diagnosis. The memory and reasoning groups showed no meaningful difference.

The researchers' leading explanation is that the speed training was adaptive: it kept adjusting to each person's level, so it stayed hard enough to work.

What It Does Not Mean

• It does not show that brain games prevent dementia. Diagnoses came from insurance records, not from testing everyone.
• The effect was in one group — speed training with boosters — not in brain training generally.
• The training was a specific research program, not an ordinary game. A puzzle app, including ours, is not that program.
• One study, however long, needs confirming before anyone can call it proven.

What to Take From It

The most useful lesson is about which mental activity seems to matter. The training that stood out had three features:

1. Visual speed and attention — scanning a busy scene and picking out the right thing quickly.
2. Difficulty that keeps up with you — always slightly too hard, never on autopilot.
3. Coming back to it — the boosters months later were part of the effect.

That is a practical checklist for choosing everyday puzzles too: prefer games that make you search and decide quickly, that get harder as you get better, and that you will actually return to week after week. And keep the rest of the picture in view — physical activity, sleep, hearing care and time with other people all have their own evidence behind them.

Puzzles That Fit the Checklist

Arrow puzzles (Arrow Jam Go — our game). Each board is a crowd of arrows, and the whole game is visual search: find the arrow whose path is clear, tap it, scan again. Boards grow from a few arrows to more than a hundred, and Nightmare mode keeps going when the main levels feel easy. You can try 300 levels free in your browser, and there is a new daily puzzle every day — a simple reason to come back.

Picture puzzles.Puzzles that reveal an image as you solve them — nonograms, jigsaws and Arrow Curve's curved-arrow pictures — add pattern recognition and a satisfying finish. See our picture puzzle games guide.

3D and spatial puzzles. Rotating shapes in your head is one of the most trainable mental skills. Our spatial reasoning games guide covers the research and the games, including a free 3D arrow cube.

Reaction games. Quick-response games train speed directly; our reaction time games article looks at what improves and what does not.

FAQ

Can brain games prevent dementia? No game has been shown to prevent dementia. The strongest evidence so far is the ACTIVE trial, where adults over 65 who did a specific speed-of-processing training, with booster sessions, had about 25% fewer dementia diagnoses over 20 years than a control group. Memory and reasoning training in the same trial showed no such link.

What kind of brain training worked in the ACTIVE study? Speed-of-processing training: computer exercises that ask you to spot and locate things on a screen faster and faster, with the difficulty adjusting to your performance. Participants who also did later booster sessions showed the link to fewer dementia diagnoses.

Are puzzle games good for older adults? They are a good way to keep using attention, visual search and planning, and many people find them more enjoyable than exercises. They are not the same as the training tested in ACTIVE, so treat them as one healthy habit alongside physical activity, sleep and social contact — not as medicine.

How much time should I spend on brain games? The ACTIVE training was ten sessions of about an hour, plus boosters months later. For everyday puzzles, short regular sessions — 10 to 20 minutes most days — are easier to keep up and leave room for other activities.