What keeps a mind engaged, since “use it or lose it” is only half right

Key takeaways

• In the ACTIVE trial’s 20-year follow-up, published in 2026, 40% of people who did speed training plus booster sessions were diagnosed with dementia, compared with 49% of an untrained group¹.
• That result rests on Medicare billing codes and one subgroup, a co-author owns stock in the company that sells the speed task, and outside experts warn it could be a chance finding²³.
• A 2016 review found that brain-training programs mostly make you better at the trained tasks, with little evidence of gains in everyday thinking⁶.
• In the Synapse Project, older adults who spent three months learning demanding skills such as digital photography improved their memory, while a social group and a group doing easy tasks did not⁸⁹.
• Helping others for 2 to 4 hours a week tracked with 15% to 20% slower cognitive decline in a 2025 study of more than 30,000 Americans, though it can’t prove cause¹⁵.

In 1998 and 1999, 2,802 adults aged 65 and older joined a study called ACTIVE, short for Advanced Cognitive Training for Independent and Vital Elderly¹. They were assigned by chance to memory training, reasoning training, speed-of-processing training or no training at all¹.

The training was modest. Each group got up to 10 sessions of 60 to 75 minutes over five or six weeks¹. Some people were also picked at random for a few “booster” sessions about one and three years later¹.

Twenty years on, researchers checked the volunteers’ Medicare records. Among people who did speed training with boosters, 105 of 264 had been diagnosed with dementia, or 40%. In the untrained group, 239 of 491 had, or 49%¹. In raw numbers, that’s about 9 fewer diagnoses per 100 people, by our math¹.

No other group stood out. Memory and reasoning training, with or without boosters, showed no clear link to fewer diagnoses².

So “use it or lose it” is half right. Some ways of using your mind seem to matter, and many don’t. Here is what the best studies show about brain games, learning, work and helping others, plus a weekly menu you can try.

One kind of training stood out in the biggest trial

Speed training wasn’t a crossword. People looked at a screen and made quick decisions about images while also spotting objects at the edge of their vision³. The task got harder as they improved³.

That design set it apart. Johns Hopkins researchers note that speed training adjusted to each person’s performance that day, while the memory and reasoning groups learned fixed strategies¹.

The signal had shown up before. At the 10-year mark, ACTIVE researchers reported 29% lower dementia incidence after speed training, and each booster session was linked to a further drop¹.

In the 2026 analysis, published in Alzheimer’s & Dementia: Translational Research & Clinical Interventions, speed training with boosters carried about a 25% lower risk of a dementia diagnosis than no training¹². The National Institute on Aging paid for the follow-up¹.

Read the fine print before you buy the game

That headline number comes with several catches.

The benefit showed up only with boosters. Speed training without boosters showed no benefit, and none of the training groups stood out when boosters were ignored². Only people who finished at least 8 of the first 10 sessions could get boosters, so they may have been healthier or more motivated from the start².

The diagnoses came from billing codes. Dementia was identified from Medicare claims, not from doctors examining each person². People in Medicare Advantage plans were left out because their claims were incomplete². The records covered 2,021 people, about 72% of the original group¹. The authors note that the people left out tend to be healthier, which could have hidden some benefit².

Many tests were run. Sarah Ackley, a Brown University researcher who wasn’t involved, noted that the team ran more than 20 statistical tests and only two suggested a benefit, which raises the odds of a false alarm³. Baptiste Leurent of University College London questioned whether the booster group was comparable to the controls³. The study’s authors acknowledged the risk of a false alarm but judged it low².

Money is involved. The speed task is now sold as an exercise called Double Decision in BrainHQ, a subscription program from Posit Science, which owns the task³⁴. Co-author Karlene Ball helped develop the original training⁴. In the paper’s disclosure, she reported that she consults for Posit Science, owns stock in it and sits on its scientific advisory board².

The company’s own announcement said the game “may help protect against dementia,” and it didn’t mention those ties⁴.

None of that makes the finding false. Marilyn Albert of Johns Hopkins, the study’s corresponding author, said more research is needed to understand why only speed training showed this long-term link¹. For now, it is one promising result from one trial, not proven prevention.

Scientists split over brain games, and regulators stepped in

In October 2014, 75 scientists signed a statement issued by the Stanford Center on Longevity and Germany’s Max Planck Institute for Human Development⁵. They said the evidence didn’t show that brain games improve everyday thinking or prevent brain disease, and they objected to ads that play on older adults’ fears⁵.

Months later, 133 scientists and practitioners signed a rebuttal saying many studies do show benefits⁶.

A team led by psychologist Daniel Simons then reviewed the studies that leading brain-training companies cited. Their 2016 verdict, in Psychological Science in the Public Interest: training reliably improves the trained tasks, does less for closely related tasks, and shows little effect on unrelated tasks or daily thinking⁶. Many of the studies had design flaws⁶.

The Federal Trade Commission (FTC) stepped in, too. In 2016, it charged that Lumosity’s maker lacked the science for claims that its games could slow memory decline and ward off dementia, and the settlement requires solid evidence behind any such claims⁷. Our article on testimonials covers that case.

Learning a hard new skill beat staying busy

If games mostly train games, what about real-life learning? Psychologist Denise Park’s Synapse Project, at the University of Texas at Dallas, tested that question⁹.

Her team placed 221 adults aged 60 to 90 in activity groups for three months, about 15 hours a week⁹. Some learned digital photography, including cameras and editing software. Some learned quilting on computer-controlled sewing machines. Some did both⁹.

Others got easier assignments. One group met for social activities such as games, storytelling and museum trips. Another listened to music, watched videos or did easy word games at home⁹.

Memory improved in the demanding groups compared with the easy ones⁸. The university singled out the photography groups for the clearest gains⁹. The social group was the surprise: friendly company on its own did little for memory in three months⁸.

Park’s advice was to try something unfamiliar and mentally hard. She also said no one yet knows whether such activities slow brain aging⁹. The National Institute on Aging funded the study⁹.

That doesn’t make friendship unimportant. Our articles on loneliness and social support cover why connection matters for health. Challenge and company may simply do different jobs, and a good class gives you both.

Language lessons: one trial says yes, another says no

Learning a language late in life sounds like the perfect brain workout. Two trials disagree.

In one trial, researchers randomly assigned 160 healthy adults aged 65 to 75 to 11 weeks of Italian lessons or relaxation training¹⁰. The language group learned some Italian. But its gains on memory and intelligence tests were about the same as the relaxation group’s¹⁰.

A Hong Kong trial ran longer. It assigned 153 adults aged 60 to 85 to up to five hours a week of computer-based language lessons for six months, to brain-teaser games, or to watching music videos¹¹. The language learners and game players improved on overall thinking scores, and the music listeners did not¹¹.

The longer course may explain the gap, but nobody knows for sure. The fair summary is mixed. Learn a language because you’ll use it, with grandchildren, on a trip or in your family’s heritage, and count any brain benefit as a bonus.

A demanding job leaves a trace decades later

Some engagement builds up over a lifetime. Researchers call one version of this idea cognitive reserve: a brain shaped by years of mental challenge may cope longer with the damage that comes with age.

A Norwegian study looked at one source of that challenge: work. Researchers traced the job histories of 7,003 people from their 30s into their 60s, then checked their thinking at ages 69 and older¹². Jobs were rated by how routine their tasks were¹².

The gap was large. About 42% of people in the least demanding jobs had mild cognitive impairment, an early stage of memory and thinking trouble, compared with 27% in the most demanding jobs, the Fisher Center for Alzheimer’s Research Foundation reported¹³. After accounting for age, sex and education, the low-demand group had a 37% higher risk of dementia¹².

There’s a catch. When the team also accounted for income and health, the dementia link was no longer statistically clear, though the link with mild impairment held¹². Part of the pattern may run through money and health rather than the work itself.

You can’t rewrite your work history. But the finding suggests a goal for retirement: replace the problem-solving your job used to give you.

Helping others tracked with slower decline

Volunteering is one way to keep that challenge, and it comes with people. Our article on purpose covers the Experience Corps school-volunteer trial, and our article on social support covers volunteering and longevity. Newer data look at thinking itself.

In 2025, University of Texas at Austin researchers analyzed Health and Retirement Study data on more than 30,000 Americans aged 51 and older, followed for about two decades¹⁵. They counted formal volunteering and informal help, such as rides to appointments, child care, yard work or preparing someone’s taxes¹⁵.

People who helped for about 2 to 4 hours a week showed 15% to 20% slower age-related cognitive decline¹⁵. Informal help counted about as much as formal volunteering, and stopping altogether was linked to worse thinking¹⁵.

The study was observational and funded by two National Institutes of Health institutes¹⁵. People with sharper minds may simply help more. Still, the result fits the Synapse lesson: regular, purposeful effort seems to count.

Hearing matters here, too. If you can’t follow the conversation, you may drop the very activities that keep you engaged. Our loneliness article covers the large ACHIEVE hearing trial.

What the evidence shows

Claim	Evidence level	What it rests on
Brain-training games improve the tasks you practice	Strong evidence	Review of studies cited by brain-training companies6
Brain-training games improve everyday thinking	Not enough evidence	Little transfer to untrained tasks56
Speed training with boosters lowers dementia risk	Promising but early	One trial, claims-based diagnoses, many tests123
Memory or reasoning strategy training lowers dementia risk	Not enough evidence	No significant link at 20 years in one trial2
Learning a demanding new skill improves memory	Promising but early	One three-month trial of 221 adults89
Learning a language sharpens thinking	Promising but early	Two trials with opposite results1011
Mentally demanding work tracks with less cognitive impairment	Promising but early	Observational; the dementia link faded after adjusting for income and health12
Helping others tracks with slower decline	Promising but early	Large observational study15
Social activity alone sharpens memory	Not enough evidence	No memory gain in the Synapse social group89

Your brain-engagement menu

This menu is our own, built from the studies above. Pick two or three items that fit your life, and give each at least a few months.

Activity	What research shows	Rating	How to start
A hard, hands-on new skill	Memory gains after three months of demanding classes such as digital photography89	Promising but early	Choose a class with homework and a finished project. Synapse used about 15 hours a week, so build up
Helping others	2 to 4 hours a week tracked with slower decline15	Promising but early	Offer a neighbor rides to appointments, or take one weekly shift somewhere
A language you’ll use	Mixed trial results1011	Promising but early	Pair lessons with real conversation, such as a weekly group
A role with problem-solving	Demanding work tracked with less impairment12	Promising but early	Be the club treasurer, a tutor or a project organizer
Adaptive speed training	Link to fewer diagnoses only with boosters12	Promising but early	If you try it, keep returning for refresher sessions and ignore dementia promises
Familiar puzzles and games	Mostly improve the game itself6	Not enough evidence	Keep them for fun, not as medicine

Here is one sample week, for illustration only: a two-hour photography class on Monday with practice at home, a two-hour volunteer shift on Wednesday, a language conversation group on Friday, and calls or visits with friends in between.

Engagement asks for nothing you have to buy

Natural health has long trusted the body’s own capacities, and the mind is no exception. The best-supported options here need no pill and often no fee: a class at a library or senior center, a craft learned from a neighbor, a shift at a food pantry, a ride offered to a friend.

Many traditions have built this in for centuries. Quilting circles, choirs, apprenticeships and the work of passing a language to grandchildren combine challenge with company. No trial has tested those traditions directly, but they share the features that did well in studies.

The same healthy skepticism you bring to drug ads belongs here. A brain-training company’s research deserves the scrutiny you’d give a drug maker’s. So does any program that sells protection from dementia.

Questions readers ask

The bottom line

“Use it or lose it” is half right. Some kinds of mental effort show real promise: adaptive speed training with refresher sessions, learning demanding new skills and regularly helping other people. Most brain games mainly improve the games themselves, and no product has been shown to prevent dementia.

The most practical plan is also the cheapest. Keep learning things that are hard for you, keep doing it with other people, and treat any app as one small item on a much bigger menu.

Related reading

• Testimonials can be honest and still prove nothing, and federal law agrees
• What eases loneliness, which depends less on how many people you see
• Purpose tracks with a longer life, while the Blue Zones story is less certain
• What held up from a famous study linking dancing to less dementia
• The dementia risk factors most people never guess

Sources

1. Johns Hopkins Medicine. “Cognitive Speed Training Linked to Lower Dementia Incidence Up To 20 Years Later” (on Coe NB, Albert MS, et al., Alzheimer’s & Dementia: Translational Research & Clinical Interventions, 2026, doi:10.1002/trc2.70197). News release, February 2026. hopkinsmedicine.org
2. Coe NB, Miller KEM, Sun C, et al. “Impact of cognitive training on claims-based diagnosed dementia over 20 years: evidence from the ACTIVE study.” Alzheimer’s & Dementia: Translational Research & Clinical Interventions, 2026;12:e70197, doi:10.1002/trc2.70197 (full text; copy posted by BrainHQ, which sells the speed task). brainhq.com
3. Being Patient. “Can Brain Training Reduce Dementia Risk 20 Years Later?” 2026. beingpatient.com
4. BrainHQ (Posit Science). “Brain Train Game May Help Protect Against Dementia for Up to 20 Years.” Company page, February 9, 2026 (cited to show the company’s claims). brainhq.com
5. Stanford Center on Longevity and Max Planck Institute for Human Development. “A Consensus on the Brain Training Industry from the Scientific Community.” October 20, 2014. longevity.stanford.edu
6. Simons DJ, Boot WR, Charness N, et al. “Do ‘Brain-Training’ Programs Work?” Psychological Science in the Public Interest, 2016;17(3) (abstract). pubmed.ncbi.nlm.nih.gov
7. Federal Trade Commission. “Lumosity to Pay $2 Million to Settle FTC Deceptive Advertising Charges for Its ‘Brain Training’ Program.” Press release, January 5, 2016. ftc.gov
8. Park DC, Lodi-Smith J, Drew L, et al. “The Impact of Sustained Engagement on Cognitive Function in Older Adults: The Synapse Project.” Psychological Science, 2014 (abstract, Association for Psychological Science). psychologicalscience.org
9. University of Texas at Dallas. “Mentally Challenging Activities Improve Memory as Baby Boomers Age.” UT Dallas News Center, October 22, 2013. news.utdallas.edu
10. Berggren R, Nilsson J, Brehmer Y, Schmiedek F, Lövdén M. “Foreign language learning in older age does not improve memory or intelligence: evidence from a randomized controlled study.” Psychology and Aging, 2020;35(2):212-219 (Tilburg University record). research.tilburguniversity.edu
11. Klimova B. “Are There Any Cognitive Benefits of Computer-Based Foreign Language Training for Healthy Elderly People? A Mini-Review” (summarizing Wong et al., 2019). Frontiers in Psychology, 2021;11:573287. frontiersin.org
12. Edwin TH, Håberg AK, Zotcheva E, et al. “Trajectories of Occupational Cognitive Demands and Risk of Mild Cognitive Impairment and Dementia in Later Life: The HUNT4 70+ Study.” Neurology, 2024;102(9):e209353 (Health and Retirement Study bibliography record). pubmed.ncbi.nlm.nih.gov
13. Fisher Center for Alzheimer’s Research Foundation. “A Mentally Challenging Job Can Bolster the Brain.” ALZinfo.org, May 1, 2024. alzinfo.org
14. National Institute on Aging. “Memory Problems, Forgetfulness, and Aging.” NIA, content reviewed November 22, 2023. nia.nih.gov
15. University of Texas at Austin, College of Natural Sciences. “Helping Others Shown To Slow Cognitive Decline” (on Han SH, Zhang S, Burr J, Social Science & Medicine, 2025). August 14, 2025. cns.utexas.edu