CATEGORY: HEALTHY AGING & LONGEVITY
Older adults assigned by chance to a speed-of-processing brain-training program—and who returned for at least one refresher round—had a 25% lower rate of dementia diagnoses over the following 20 years, according to a new analysis of the landmark ACTIVE trial published in Alzheimer’s & Dementia: Translational Research & Clinical Interventions.1 Among the boosted speed-training group, 40% (105 of 264 people) were diagnosed with dementia by 2019, versus 49% (239 of 491) of those assigned to no training. Speed training without the refreshers showed no effect at all, and neither did two other kinds of training.
A Game of Glancing, Played Against the Clock
The training itself looks almost like an arcade game. In speed-of-processing training—also called Useful field of view training—a participant watches a screen, identifies an object flashed at the center of gaze—a car or a truck—while simultaneously spotting a road sign flashed somewhere out toward the edge of vision. Get it right and the next round comes faster, with more distractions; the program adjusts its difficulty to each player, every session. The initial course was ten sessions of 60 to 75 minutes over five to six weeks. Random subsets of participants were later offered booster rounds—up to four extra sessions—about one year and three years in.
Randomized in 1998, Answered by Medicare in 2019
The ACTIVE study enrolled 2,802 healthy adults aged 65 and older at six U.S. sites in 1998 and 1999 and assigned each by chance to memory training, reasoning training, speed-of-processing training, or no training—the design that makes the groups alike in everything except the training itself. The new analysis linked the 2,021 participants who were enrolled in traditional Medicare to their claims records from 1999 through 2019, flagging dementia with a standard federal claims algorithm. That approach is what makes a 20-year answer possible, but it counts only dementia that was actually diagnosed and billed for—it involves no in-person testing, and dementia that never reached a medical record would be missed.
One Arm Out of Three Moved the Needle, and Only With Refreshers
Participants who completed speed training plus at least one booster round had a hazard ratio of 0.75—a 25% lower risk of a dementia diagnosis across the follow-up period, with a confidence range of 5% to 41%. Those who got the initial speed course but no boosters fared no better than the untrained (a hazard ratio of 1.01, meaning no difference). Memory training and reasoning training—strategy-based programs teaching mnemonics and problem-solving patterns—showed no effect on dementia risk in any form. The researchers note that speed training differs from the other two in kind: it builds an automatic skill through fast repetition, the way one learns to ride a bicycle, rather than teaching a strategy one must remember to use.
An Earlier Look at the Same People Pointed the Same Way
This is not the first signal from ACTIVE. A 2017 analysis of the same trial’s first ten years, using in-study assessments rather than Medicare claims, found a 29% lower dementia risk with speed training, and roughly 10% lower risk for each additional session completed.2 The new work extends that pattern another decade with an independent way of counting cases. The boundary worth keeping in view: both analyses come from the same 2,802 people, no other randomized trial has yet tested this kind of training against dementia over decades, and because the benefit appeared only among those who took refresher sessions, the result reads as much like a dose effect as a verdict on the five-week course alone.
Key Takeaways
- In the randomized ACTIVE trial, older adults given speed-of-processing training plus booster sessions were diagnosed with dementia at a 25% lower rate over 20 years—40% versus 49% among the untrained.
- The initial course alone was not enough: without refresher sessions, dementia rates matched the no-training group, and memory and reasoning training showed no effect either way.
- Dementia was identified through Medicare billing records rather than in-person exams, so the finding is best read as a delay in diagnosed dementia rather than proof of prevention.
- A 2017 analysis of the trial’s first decade found the same pattern, including lower risk with each additional training session.
References
- 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(1):e70197. link
- Edwards JD, Xu H, Clark DO, et al. Speed of processing training results in lower risk of dementia. Alzheimer’s & Dementia: Translational Research & Clinical Interventions. 2017;3(4):603–611. link