Since 2000, Northwestern Medicine has tracked 290 volunteers over age 80 using a 15-word memory test, and after 25 years the ones who still score like adults in their fifties and sixties turn out to share one habit above the rest: close, frequent contact with other people.

In 2000, a research team at Northwestern University started recruiting people over 80 for a study that, at the time, had no name for what it was looking for.

A quarter century later, it has one. The participants who kept testing sharp are known in the research literature as SuperAgers, and the program built around them has spent twenty-five years trying to figure out what actually separates them from everyone else their age.

The test at the center of it is almost comically plain. Read someone a list of 15 unrelated words, wait a while, then ask them to recall as many as they can. Most 80-year-olds get about a third of the list back. A shrinking group gets nine or more, a score that’s typical for someone in their late fifties or early sixties, not their eighties. In all, 290 volunteers have gone through this testing since the program began, and a portion of them have kept clearing that bar year after year.

What the word test actually sorts for

The Northwestern University SuperAging Program, launched with initial funding in 2000 and 2001, uses a delayed word-recall task called the Rey Auditory Verbal Learning Test to draw its line. Researchers reviewing the program’s first 25 years describe the cutoff plainly: recalling at least 9 of 15 words after a delay is average performance for someone in their late fifties or early sixties, well above the average for an 80-year-old, which researchers put at roughly 5 of 15. Anyone 80 or older who clears that 9-word bar qualifies as a SuperAger for the purposes of the study.

That’s a narrow, mechanical definition, chosen for exactly that reason: it measures whether someone can do, cold, what a person twenty-some years younger can typically do, rather than relying on how sharp they seem in conversation. Of the 290 people who have gone through the program’s testing, only a fraction ever clear that bar, and clearing it once isn’t the same as clearing it for years running.

The search started in the brain

For a long stretch of the program’s history, the leading theory was biological. Brain scans turned up a region of the cortex that was, on average, thicker in SuperAgers than in neurotypical adults decades younger, and a follow-up study tracking the same participants over 18 months found their cortex thinning at less than half the rate of their peers’.

Post-mortem tissue from SuperAgers who later donated their brains also showed a considerably higher density of von Economo neurons, a rare cell type concentrated in brain regions tied to social behavior. That work mattered, and it’s still part of how the program describes what makes a SuperAger’s brain distinct on a scan.

But brain scans don’t explain why someone ended up with that anatomy in the first place, and a scan alone doesn’t predict who’s going to keep testing well five or ten years out. So alongside the imaging work, the research team also started asking participants ordinary questions about their daily lives: how satisfied they were, how connected they felt, what their days actually looked like.

The one thing that kept showing up

The answer to those questions turned out to be less about any single lifestyle choice than about who these people spent their time with.

Some SuperAgers exercised religiously and ate carefully. Others skipped exercise, smoked, drank most evenings, and did not sleep particularly well. Researchers reviewing the program’s quarter-century of data found no diet, workout routine, or health regimen that reliably set SuperAgers apart from their same-age peers who tested closer to average.

What did hold up, across a program covering hundreds of participants and multiple separate studies, was a pattern in how these people described their relationships. As neuropsychologist Sandra Weintraub and her Northwestern colleagues put it, reviewing the program’s full 25-year history, “the group was particularly sociable and relished extracurricular activities. […] The one observation that could be generalized was the gregariousness of the participants.”

A separate study within the program, drawing on a widely used well-being questionnaire, found something narrower and just as consistent: SuperAgers and their same-age peers reported similarly high well-being on most measures, but SuperAgers stood out specifically on one.

As the National Institute on Aging summarized that finding, “SuperAgers scored significantly higher on the positive social relationships scale, which measures satisfying, high-quality relationships reported by the individual.”

What this doesn’t prove

It’s worth being careful about what a finding like this can and can’t carry. This is correlational research: people who already tested as SuperAgers were found to report richer social ties, not the other way around. The study can’t rule out the possibility that whatever gives someone resilient memory in old age also happens to make them more inclined toward close relationships, rather than the relationships themselves protecting the memory. None of this is medical advice, and it isn’t a diagnosis or a prediction about any individual’s future; anyone with real concerns about their own memory or a family member’s is better served talking to a doctor than reading a research summary.

What the researchers are willing to say is narrower and more honest than a headline would suggest: that perceived social connection tracked closely enough with cognitive resilience, across enough of the program’s participants, to be worth taking seriously as a factor rather than a footnote. That’s a more modest claim than “friendship prevents memory loss,” and it’s also the one the data can actually support.

A quarter century of the same finding

What makes this particular thread notable is how many separate times it turned up. A program that started by looking for a biological signature kept circling back, study after study, to the same social one. Brain volume differences held up. The rare neuron findings held up. But neither of those turned out to be things a person could act on this week. The finding about close, frequent contact reads more as a description of how these people actually spend their time than as a fact about their brain scans, which makes it a different kind of finding: one that points to a habit rather than an inheritance.

Whether that habit is a cause, a consequence, or just a companion of resilient memory is still an open question after 25 years. What isn’t open to much debate, at this point, is that it’s the one thread the researchers themselves keep pointing back to, out of everything they went looking for.