🧠 The ACTIVE Study: A 20-Year Experiment
The Advanced Cognitive Training for Independent and Vital Elderly (ACTIVE) study began between 1998 and 1999, when researchers from Johns Hopkins Medicine, the University of Pennsylvania, and other top U.S. institutions recruited 2,802 adults aged 65 and older to participate in a pioneering experiment. Participants were randomly assigned to four groups: three training groups and one control group with no program.
The three forms of training were: memory exercises (mnemonic techniques for memorizing lists and details), reasoning exercises (pattern recognition and solving logic problems), and processing speed exercises (quickly identifying visual elements on a computer screen while difficulty gradually increased). Each group completed up to 10 sessions of 60-75 minutes, over a period of five to six weeks. Some participants also received up to four additional “booster” sessions one to three years later.
The results, published in February 2026 in the journal Alzheimer's & Dementia: Translational Research and Clinical Interventions, represent the first randomized clinical trial to track dementia outcomes for two full decades in elderly individuals who underwent cognitive training.
📊 The Results: Only One Type of Training Worked
Here lies the big surprise of the study: of the three forms of training, only processing speed led to a statistically significant reduction in dementia risk. Memory and reasoning exercises had no protective effect compared to the control group.
Specifically, among participants who completed speed training along with booster sessions, 105 out of 264 (40%) developed dementia over 20 years. In the control group, the corresponding rate was 239 out of 491 (49%). This translates to a 25% lower risk of dementia — a remarkable result for an intervention that lasted just a few weeks.
"The fact that boosted speed training was associated with lower dementia risk two decades later is remarkable, because it shows that a relatively modest non-pharmacological intervention can have long-term effects."
— Marilyn Albert, Ph.D., Johns Hopkins Medicine
🔬 Why Processing Speed Stands Out
Researchers believe processing speed was particularly effective for two reasons. First, the program was adaptive: difficulty was automatically adjusted based on each participant's performance. Those who performed well advanced to more complex exercises, while those who struggled worked at a slower pace. The memory and reasoning programs, by contrast, taught the same strategies to everyone.
Second, speed training is based on implicit learning — it resembles building a skill or habit rather than memorizing facts. Memory and reasoning, by contrast, depend on explicit learning, meaning the conscious acquisition of techniques. Scientists know that these two forms of learning activate different brain systems — and this distinction may explain the difference in results.
Dr. Marilyn Albert from Johns Hopkins noted that speed training isn't particularly fun: “It's hard,” she admitted. But that very difficulty — the fact that it pushes the brain to continuously adapt — appears to be the key to its effectiveness.
🧩 Key Findings in Numbers
- 2,802 elderly participants (65+)
- 5-6 weeks initial training (10 sessions of 60-75 min)
- 25% lower dementia risk in the speed + boosters group
- 20 years of follow-up — the longest randomized trial of its kind
- 0% effect from memory/reasoning — no difference from the control group
🎮 Can Video Games Achieve the Same Results?
The findings raise an interesting question: could brain-training apps and video games replicate this effect? There are hundreds of such apps on the market claiming to enhance cognitive function. However, the researchers caution: "When these things get commercialized, sometimes claims get made that go beyond the data," noted neuroscientist Art Kramer.
Albert was even more cautious. She emphasized that the most important factor in speed exercises — adaptiveness — may be missing from commercial games. The training in the study involved identifying objects at the center and edges of a screen, with continuously increasing speed and complexity based on performance. This dynamic adjustment is absent from many popular apps.
💡 Why This Matters for All of Us
Dementia is one of the greatest health challenges of the 21st century. It is estimated to affect approximately 42% of adults over 55 at some point in their lives, with costs exceeding $600 billion annually in the U.S. alone. Alzheimer's disease accounts for 60%-80% of cases, followed by vascular dementia, Lewy body dementia, and frontotemporal dementia.
In this context, the idea that a relatively brief, non-pharmacological intervention can delay or reduce risk is extremely attractive. "Even small delays in the onset of dementia can have an enormous impact on public health and help reduce rising healthcare costs," said Albert.
Researchers emphasize that cognitive training should not be considered a standalone solution. Other factors associated with lower cognitive decline risk include managing blood pressure, physical exercise, controlling blood sugar and cholesterol, and maintaining a healthy body weight. The next step, according to the researchers, is investigating the biological mechanisms behind these results — possibly through MRI scans that would show how cognitive training changes brain structure.