Cognitive speed training cut dementia risk by 25% over 20 years. That is the headline finding from the February 2026 update of the ACTIVE study, the largest cognitive training trial ever conducted. Speed training was the only one of three training types that showed the protective effect — not memory training, not reasoning training. The training itself took five weeks. The benefit lasted two decades. (Source: Coe et al., Alzheimer's & Dementia: Translational Research & Clinical Interventions, 2026.)

The 25% Number, in Context

The Advanced Cognitive Training for Independent and Vital Elderly (ACTIVE) trial began in 1998. Nearly 3,000 healthy older adults were randomly assigned to one of three training groups (speed, memory, reasoning) or a no-contact control. Each training group did 10 sessions over 5 weeks, with optional 4-session boosters at year 1 and year 3. Then researchers waited.

In February 2026, Johns Hopkins, the National Institute on Aging, and partner institutions published the 20-year follow-up. They reviewed claims-based dementia diagnoses across the full sample. The result: speed-training participants who completed the boosters were 25% less likely to be diagnosed with dementia than controls.

Memory training and reasoning training showed no equivalent benefit. The protective effect was specific to processing speed.

Why Speed Training, Specifically

The speed training in ACTIVE used a Useful Field of View task — participants had to identify a target in the center of the screen while simultaneously locating a peripheral target, all under tight time pressure. As participants got faster and more accurate, the task got harder. This is adaptive difficulty — the training continuously calibrated to each person's ability.

Researchers point to two possible reasons speed training worked while memory and reasoning training did not:

1. Implicit learning. Speed training trains automatic, unconscious processes. Implicit skills tend to be more durable than skills that depend on conscious recall.

2. Brain network coverage. Processing speed is a global capacity — improving it touches attention, working memory, and decision-making in turn. Memory and reasoning training tend to be more domain-specific.

Either way, the practical implication is concrete: when researchers found something that protects the brain over 20 years, it was speed-pressured, adaptively-difficulty-tuned cognitive training.

What This Means for Daily Brain Training

The ACTIVE result is one study, and it has limits. It studied older adults (mean age 73 at baseline). It used claims data, not in-person diagnostics. The training was structured, not gamified. So it is not a license to claim "play brain games and you will not get dementia." That is the kind of overclaim that got Lumosity fined $2 million by the FTC in 2016.

What the result *does* support, in our reading: regular practice on adaptively-difficult, speed-pressured cognitive tasks is a low-cost, low-risk lifestyle intervention with the strongest long-term evidence of any cognitive intervention to date. For older adults thinking about a daily routine in light of these findings, our [brain training guide for adults 60+](/brain-training/for-seniors) walks through the cadence and which FOKIQ surfaces most directly map to the ACTIVE protocol.

How FOKIQ Maps to the ACTIVE Findings

FOKIQ is not the ACTIVE protocol. But its design directly maps to the factors the study identified as effective:

  • Speed Processing is a full domain in FOKIQ. Quick Math, Change Detect, Rapid Sort, Counting, and Odd One Out all impose time pressure under accuracy constraints — the same shape as Useful Field of View.
  • Adaptive difficulty runs across all 6 domains. Difficulty rises and falls based on the player's rolling accuracy. The target is roughly 70% — challenging enough to drive learning, gentle enough to not discourage.
  • Daily structure. The Daily takes about two minutes. ACTIVE participants did 10 sessions in 5 weeks. The cumulative annual training volume of someone playing The Daily is far higher than ACTIVE's protocol.
  • Booster sessions matter. ACTIVE participants who took the year-1 and year-3 boosters were the ones who showed the 25% reduction. The lesson: long-term consistency outperforms short-term intensity.

What to Train First

If your goal is brain-health-driven training in light of the ACTIVE findings, the highest-leverage starting point is daily Speed Processing practice. In FOKIQ:

  • Play The Daily every day. Speed Processing rotates into the daily set regularly.
  • Use Playground in the Speed domain on the days when Speed is *not* in The Daily.
  • Use Blitz mode (60 seconds, max puzzles) when you want pure speed pressure.
  • Track your Speed Processing score on your MindMap over weeks and months. You are looking for the line to bend upward.

What ACTIVE Does Not Say

A few things the study explicitly does *not* establish:

  • It does not prove a causal mechanism. The 25% reduction is correlational with the assigned training condition; the brain mechanism is plausible but not pinned down.
  • It does not show that brain training reverses existing cognitive decline. ACTIVE participants were healthy adults at baseline.
  • It does not say speed training is the only thing that matters. Physical exercise, sleep, social engagement, and cardiovascular health are all stronger and more multi-causally validated dementia-risk factors than any cognitive training intervention.

Bottom Line

A 25% reduction in 20-year dementia risk from a 5-week cognitive intervention is the strongest result the field has produced. The intervention was specifically speed-pressured, adaptively-difficult cognitive training. FOKIQ's Speed Processing domain is built on those exact principles. Pair it with sleep, exercise, and social engagement, and you have a brain-health stack that is cheap, evidence-grounded, and free.

Try [today's puzzle](/) — two minutes, your Speed Processing score updates with every play.