Long-term follow-up suggests targeted speed training may reduce dementia diagnoses later in life, raising new questions for prevention policy.
For decades, the landscape of dementia prevention has been defined by a wearying catalogue of what does not work – a frustrating cycle of supplements that disappoint, lifestyle advice that struggles to find scale and drug candidates that invariably arrive too late and leave too early. It is a space where the complexity of the pathology consistently outpaces the relative simplicity of our interventions. Now, new long-term data from the ACTIVE trial offers something rarer – a signal that persists; specific, not sweeping, and anchored to a hard endpoint rather than a pleasing cognitive score [1].
ACTIVE began in the late 1990s with more than 2,800 cognitively healthy older adults randomized to memory, reasoning or speed-of-processing training, or a no-contact control. Two decades later, researchers linked participants to Medicare claims to track dementia diagnoses over 20 years; the lower risk appears only in the speed-of-processing arm among those who completed booster sessions, not in memory or reasoning, and not in speed training without boosters [1].
Longevity.Technology: A randomized trial with 20-year follow-up and real-world dementia diagnoses is an unusual commodity in prevention science, which is precisely why these results deserve attention – and restraint. The effect is not “brain training” in the abstract, but a specific speed-of-processing intervention with boosters; a salient reminder that dose and durability often matter more than the branding on the box. It effectively drags dementia prevention out of the realm of general advice and into operational questions that health systems cannot dodge indefinitely: who possesses the agency to stick with a program, how equitable access is managed, what happens outside the tidy constraints of a trial and whether we are looking at delayed diagnosis via cognitive reserve or a more fundamental shift in disease trajectory. Commercial lineage makes independent replication the sensible next step, not a footnote. To find out more about speed-of-processing training, neuroplasticity and cognitive reserve, we spoke with Dr Henry Mahncke, CEO of Posit Science, the company behind the BrainHQ Double Decision platform used in the study.
From trial data to real-world implications
ACTIVE’s value, two decades on, lies in its endpoint. Rather than relying on repeated cognitive testing alone, the investigators used Medicare claims to track dementia diagnoses over 20 years, finding a lower risk only among those randomized to speed-of-processing training who also completed booster sessions. That specificity is the point. It hints at a threshold effect, or at least a requirement for continued cognitive challenge, while discouraging any temptation to treat digital cognitive training as a single, interchangeable category. It also pushes the discussion into less comfortable territory – mechanisms, dose, adherence, equity and the unresolved distinction between delaying diagnosis and altering disease itself.
Why ACTIVE looks different at 20 years
The longevity of the ACTIVE signal is, by Mahncke’s own admission, not something he takes lightly. “Even I am surprised by the 20-year results,” he says, noting that this is the first time an intervention has been shown, in a randomized controlled trial, to significantly reduce the risk of being diagnosed with Alzheimer’s or other dementias. For researchers steeped in brain plasticity, the direction of travel was not unexpected; for much of the dementia field, he acknowledges, it may feel counterintuitive.
What shifts the conversation, he argues, is not just the duration but the endpoint. “Seeing in a gold standard randomized controlled trial that a modest amount of training – 10 to 12.5 hours at the start of the study, and another five hours at the end of the first and third year – would have a significant and substantial impact over the next 20 years is astonishing.” The use of Medicare records, rather than repeated cognitive tests, moves the findings from the abstract to the clinical; diagnosis, after all, is where the personal and economic costs of dementia begin to accumulate.
That time horizon matters because dementia does not arrive suddenly. “We do generally know that the onset of Alzheimer’s and dementia is a long disease process,” Mahncke says, “only becoming diagnosed after the root brain health causes have been insidiously having a slow and gradual effect over decades.” Seen in that light, the implication is uncomfortable but also hopeful. “The new results tell us that it’s never too early to start brain health dementia prevention activities – particularly speed training.”
From cognitive training to disease prevention
Why speed-of-processing training should differ from memory or reasoning interventions is, he suggests, a question of biology as much as behavior. All three arms of ACTIVE improved cognitive performance, but only speed training translated into fewer dementia diagnoses [1]. “What’s different,” he explains, “is that it requires procedural learning – skill building – as opposed to declarative strategy learning.” The training is adaptive, progressively challenging participants at their own threshold, and that matters. “It engages brain plasticity to rewire the brain structurally, functionally and chemically.”
The effects, he says, are durable by design. “Think of when you learned to ride a bike and how long-lasting that is – even if you don’t get on a bike for decades.” In this case, the rewiring targets attention and processing speed, the substrate on which higher cognition depends. Recent neuroimaging work adds a biochemical layer: “We’ve confirmed that this same training uniquely increases the production of acetylcholine – the pay-attention chemical – which is an important part of the explanation.”
For Mahncke, the policy implications now loom as large as the science. Observational studies have long linked healthy behaviors to lower dementia risk, but causality has remained slippery. “With observational studies, it’s never quite clear what is cause and what is effect,” he says. ACTIVE, by contrast, offers a cleaner line. “Speed training with booster sessions is the cause; lowered risk of Alzheimer’s and dementia is the effect.” With that, he argues, “it’s time to update what experts tell people about how to maintain their brain health… It’s time to declare that the age of dementia prevention has begun.”
[1] https://alz-journals.onlinelibrary.wiley.com/doi/10.1002/trc2.70197