Preschoolers who nap after learning a memory game remember roughly ten percent more of what they learned the next day than classmates who were kept awake, according to a 2013 study led by Rebecca Spencer at the University of Massachusetts Amherst and published in the Proceedings of the National Academy of Sciences. When the same children were later monitored in a sleep lab, their EEG traces showed dense bursts of activity called sleep spindles — the same electrical signature adults produce during the deep, memory-stitching phases of sleep.
The finding is small, specific, and quietly radical. Ten percent is the difference between a child who can point to seven picture locations on a grid and a child who can point to eight. Multiply that across a school year of songs, letters, routines, and social rules, and the arithmetic gets loud.
The Memory game, the darkened room, the foot rub
The experiment itself was almost domestic in its simplicity. Spencer’s team taught preschoolers a version of the board game Memory. On a computer screen, children saw a grid of nine to twelve pictures — an umbrella, a hammer, a rubber duck — and learned where each one lived. Then, one image at a time, the children were asked to point to the correct square.
The Artful Age
A weekly letter on aging well, family across generations, and the creative life after the kids leave home.
In the afternoon, the children were split. Half were led to individual cots in a darkened, quiet room and soothed toward sleep with back rubs and foot rubs. Half were kept awake in the same darkened room with quiet activities. Both groups were tested shortly after nap time, and again the following morning after a full night’s sleep at home.
The nappers won on both tests. A summary of the results in the Los Angeles Times noted that the nap advantage held up the next day, even after the awake group had also slept overnight. The sleep that mattered was the sleep that came soonest after learning.
What a sleep spindle actually looks like
In a follow-up laboratory run, the researchers wired the children with EEG caps and recorded what their brains were doing during the naps that produced the memory gain. The pattern they were watching for was the sleep spindle: a brief, tight burst of oscillation that flickers across the cortex during non-REM sleep. On a printed trace it looks like a small tuft of dense scribble against the slower background waves, lasting about half a second to a second and a half.
Each night during sleep, these spindles appear throughout the brain, and neuroscientists have linked them to the transfer of new memories out of the hippocampus and into longer-term cortical storage. What Spencer’s team found was that the children who napped after learning produced more spindles — and the density of those spindles predicted how much of the game they remembered later. The Voice of America account of the study described the effect as strongest in children who habitually napped, suggesting the nap habit itself may keep the machinery well-oiled.
Why the ten percent doesn’t wash out overnight
The obvious objection is boredom. Perhaps the awake kids were just cranky by test time. Spencer anticipated it. If restlessness were the culprit, the gap would close after a good night’s sleep. It didn’t. The next-morning test — when both groups were rested — still showed the nappers ahead.
The mechanism the researchers propose is consolidation. New learning arrives fragile, dependent on the hippocampus to hold it in place. During slow-wave sleep, the brain appears to replay the day’s patterns and hand them off to more permanent cortical addresses. Miss that window, and some of the material never makes the transfer. A separate paper in Nature’s Scientific Reports on sleep-dependent emotional memory in early childhood found the same broad pattern for feelings attached to pictures: sleep, especially prompt sleep, protected the memory.
What this changes about the nursery school afternoon
Across American preschools and kindergartens, nap time has been under quiet pressure for two decades, squeezed by academic standards, longer instructional days, and the reasonable-sounding idea that time on cots is time not spent learning. Spencer has pushed back on that idea directly: in a UMass Amherst profile of her research, she pointed out that Massachusetts law mandates just a 45-minute rest period in preschools, a limit she says was set with little evidence behind it, even as academic pressure pushes nap time further toward the margins.
The arithmetic the Amherst study proposes is simple. If a child learns X in the morning, they will remember roughly 90 percent of X by the next day if they napped, and roughly 80 percent if they didn’t. Over a hundred school days, the nappers are effectively getting ten extra days of retained instruction.
The developmental window is real, and it closes
Not every child benefits equally. Younger preschoolers — the ones who still nap habitually at home — get the strongest boost. Older children who have already dropped their nap show a smaller effect, or none. A HealthDay report on toddler sleep transitions traced the developmental logic: children stop napping when their hippocampus matures enough to hold the day’s memories without an afternoon offload. The nap, in other words, is a scaffold for a brain still under construction.
That brain, at age three or four, is running hot. A three-year-old’s brain burns glucose at roughly twice the adult rate, gram for gram, during one of the most energy-intensive periods in human development. Sleep is when that expensive tissue does the filing.
The questions cluster near the pillow
There is a small, charming detail in the developmental literature that fits neatly beside the Amherst finding. Four-year-olds ask roughly 300 questions a day, and the questions cluster most tightly in the hour before nap and the hour before bed — as if the child senses, in the pre-sleep window, that this is the moment to load in whatever they most need to keep.
The Montessori tradition noticed something related more than a century ago. In her field notes on preschool concentration, Maria Montessori described a three-year-old repeating a cylinder-fitting exercise forty-four times in a single sitting, undisturbed even when Montessori tried to distract her, before stopping on her own, calm and seemingly refreshed, as though she had just woken from a nap. It is a portrait of rehearsal, then rest, recorded a century before the sleep-lab data above existed to explain it.
What the spindles do while the room is quiet
Picture the classroom at 1 p.m. Twenty-two cots, shades drawn, the hum of a filtered air unit, the muffled scrape of a teacher’s chair. On the EEG monitor down the hall in Spencer’s lab, one child’s trace shows the long, rolling slow waves of stage-three sleep, and every few seconds, riding on top, a burst of spindle activity — the brief scribble, then back to the roll.
Each of those scribbles is, on the leading theory, a hippocampal replay. The umbrella in the top-left square. The hammer in the middle. The rubber duck two rows down. Filed, indexed, cross-referenced with what the child already knew about umbrellas and hammers and ducks. When the child wakes up ninety minutes later and points at the screen, the memory is in a different place than it was before sleep, and it is stickier.
Naps aren’t the whole story, but the timing is
The findings do not argue that sleep-deprived children can be rescued by an afternoon lie-down, and they do not argue that a nap can replace a full night. Different nap durations produce different effects — a twenty-minute nap sharpens alertness, a ninety-minute nap allows a full sleep cycle including REM, and the two do different things to memory.
What the Amherst work adds is the developmental piece: for a preschooler, the afternoon nap is not a leftover from infancy that a well-run classroom should phase out. It is an active piece of learning machinery. Cutting it to make room for more phonics may cost some of the phonics.
The quiet room, revisited
Back in the Amherst experiment, the awake children were not doing anything strenuous. They were in the same darkened room, on the same cots, listening to the same white noise. They simply weren’t allowed to fall asleep. And still, the next morning, they remembered less.
The variable that mattered was not effort or attention or engagement. It was the passage of an hour of stage-two and stage-three sleep, densely populated with spindles, close enough in time to the morning’s learning that the hippocampus had not yet let the traces fade. Move the nap to the evening, and the effect weakens. Skip it entirely, and it disappears.
Which is why, in a preschool classroom at 1 p.m., the most academically consequential thing happening in the room may be the child on the third cot from the window, breathing slowly, mouth slightly open, brain quietly writing the morning down.