In the early 1960s, a graduate student named John Horn was working in a psychologist’s laboratory at the University of Illinois, putting to the test something the field had believed for decades. The prevailing model was tidy: intelligence follows a single curve. It rises through adolescence and young adulthood, peaks somewhere around the late twenties, and then begins its long gradual descent. The challenge wasn’t Horn’s idea. Raymond Cattell, whose laboratory he was working in, had proposed the alternative back in 1943, and Horn was there to see whether the data bore it out.
The data didn’t cooperate.
What Horn found, and what he and Cattell spent the next two decades proving out, was that the single curve was actually two. One type of intelligence, which they called fluid intelligence, does peak in young adulthood and then decline. It’s the ability to reason through novel problems with no prior knowledge, to hold and manipulate unfamiliar information under time pressure, to respond quickly to things you’ve never encountered before. It’s what most cognitive tests measure, largely because those tests were designed in an era when these capacities were considered the whole substance of intelligence. Fluid intelligence goes down, starting in your thirties.
The other type, crystallized intelligence, goes up. Vocabulary, general knowledge, the capacity to apply accumulated experience to new situations, comprehension of complex ideas in context. It rises, in most people, through the sixties. In many, through the seventies and beyond. Long-term studies that follow individuals over decades show it continuing to climb well past the point where every other headline has declared the brain in terminal decline.
Two curves. Running in opposite directions. For sixty years, we’ve watched the descending one and barely mentioned the ascending one.
I want to be honest about the descending curve first, because the case for what goes up is weakened if you skip past what genuinely goes down.
Processing speed declines with age. This is real and measurably consistent across populations and research traditions. Reaction time slows beginning in the mid-thirties, and the effect accumulates. By sixty, a person genuinely processes information more slowly than they did at twenty-five. Working memory, the mental scratch pad where you hold several pieces of information simultaneously while using them, also shrinks somewhat. If you’ve noticed that following a complex spoken explanation is harder than it used to be, or that you can’t quite juggle seven things at once the way you once could, that’s not imagination. It’s documented, reproducible, and nobody serious in the field disputes it.
The standard story stops there, at the losses, and calls it a reckoning.
What it leaves out is what the brain does in response.
Neuroimaging research from the past two decades, using tools that didn’t exist when Horn was running his studies, has revealed something that older models couldn’t see. The aging brain doesn’t simply slow down. It reorganizes.
Young adults performing a difficult memory task typically show neural activation concentrated in one brain hemisphere. Older adults performing the same task show activation in both. This pattern is usually read as compensation: the brain recruiting more of itself to keep pace. That reading has real weight behind it. In the work that established the pattern, the older adults showing bilateral activation were the high performers, while the low performers stayed unilateral. It also has serious opposition. A 2018 analysis found the extra prefrontal recruitment carried no additional information about how well people actually performed, which looks less like compensation than like noise. The answer takes longer to arrive. Whether it is more reliable when it gets there is genuinely unsettled.
This is not a consolation prize. It’s a different instrument, doing different things.
Then there’s the thing that doesn’t look like a feature but turns out to be one.
Older adults, as a group, show something researchers call the positivity effect: they attend to and remember positive and emotionally meaningful information more readily than neutral or negative information. For a while, this was classified in the literature as selective inattention, a form of mild cognitive slippage where the older brain fails to catch things it once would have caught.
Laura Carstensen, a psychologist who has spent three decades studying how priorities shift with age, has argued for a different reading. The positivity effect isn’t a failure of attention. It’s attention working better, not worse. Guided by a shortening sense of how much time is left, which pushes emotional meaning to the front of the queue, the older brain weights what it takes in differently. Younger adults show the same shift when their own time horizons are experimentally narrowed. What it holds onto is the better selection.
You can call that decline if you want to. It’s also a description of wisdom functioning as a cognitive mechanism rather than a greeting-card sentiment.
Which brings me to the question people actually ask when they find themselves reading late at night about brain health: what can you do about any of this?
The research is quite clear here. The answer is considerably less interesting than the supplement industry would prefer.
Aerobic exercise is one of three interventions the National Academies identified as having encouraging, though inconclusive, evidence for protecting cognitive function over time (Source). Though in the largest randomized trial to test it, a year of moderate-to-high-intensity aerobic training left cognition no better off than low-intensity stretching, balance, and range-of-motion work (Source). Sustained aerobic activity that gets the cardiovascular system working appears to increase blood flow to the brain, stimulate the growth of new connections in the hippocampus (the region most closely associated with memory formation), and slow age-related volume loss in areas governing executive function. The effect sizes in the best studies are modest. They’ve held up unevenly across independent replications: a 2024 meta-analysis of eight randomized trials found no effect of aerobic training on hippocampal volume (Source).
Sleep is the other one. Not rest. Actual sleep, with architecture. Slow-wave sleep in particular has been thought to play a critical role in the brain’s waste-clearance system, a network of channels that flushes metabolic byproducts that accumulate during waking hours, including the amyloid proteins associated with Alzheimer’s disease. Chronically short or fragmented sleep correlates consistently with faster cognitive decline in long-term studies. That epidemiological finding has held up. The mechanism has not settled: a 2024 study in mice found brain clearance was actually reduced during sleep rather than increased, and the argument over it is still live.
Social connection has decent evidence behind it, though this research is harder to read cleanly. Isolation consistently associates with faster cognitive decline; rich social engagement appears protective. Whether engagement causes the protection, or whether cognitively healthy people simply stay more engaged, is difficult to untangle. The honest answer is probably some of both, which is the case with a lot of things.
And supplements? The largest, most carefully designed clinical trials on ginkgo biloba, vitamin E, B vitamins, and most other cognitive health products have produced disappointing results. A large, eight-year study published in 2008 that followed more than three thousand older adults found ginkgo biloba had no effect on the rate of cognitive decline or dementia development. Smaller studies find various things, which is exactly what you’d expect when publication bias tilts the record toward surprising results. If something were clearly working at scale, we’d know by now. We’d have known for years.
I turn sixty in three weeks. September ninth. I have a personal stake in this question that I’ve tried to acknowledge honestly, because a man making the case for the aging brain, at the age when you start needing the case made, is inherently suspect. The conflict of interest is real. You should hold it against me appropriately, and I won’t argue.
But here’s what I keep returning to.
The brain I’ll have at sixty isn’t the one I had at thirty-five. It’s slower at certain tasks. It knows considerably more. It’s better at ignoring things that don’t matter and more likely to notice connections between things that seem unrelated but aren’t. It recruits more of itself to solve hard problems, which takes longer and may produce more reliable answers. It carries sixty years of crystallized experience that no standard cognitive test was designed to measure, because the tests were built by researchers in their thirties studying subjects in their twenties, and what those researchers had in mind as “intelligence” was a perfect reflection of what they themselves could do at the time.
Horn’s two curves suggest that the question “is the brain declining?” is the wrong question. The right question is: which part, declining toward what, and what else is rising while that’s happening?
The hill has two sides. You can be going down one while climbing the other. Which side matters more depends entirely on what you’re trying to do with the brain you actually have.
