When the world tried to stop a virus, it also interrupted the ordinary experiences through which children learn to become human together. New research suggests some of those disruptions may have left biological traces. But the evidence is more complicated—and more revealing—than a headline about teenagers aging overnight.
In 2020, an ordinary childhood became an extraordinary public-health experiment. Classrooms went dark. Birthday parties became squares on a screen. Grandparents were suddenly people to protect by staying away from them. For months, and in some places much longer, smiles disappeared behind masks and friendships were squeezed into video calls, distanced playgrounds or silence.
The measures were designed to slow a dangerous infection, and evidence indicates that combinations of precautions reduced transmission. Yet children were never simply smaller adults waiting patiently for the emergency to end. Their brains, bodies, language skills and social instincts were still under construction.
Now a November 2026 study in Proceedings of the National Academy of Sciences, examined in a VICE report, suggests the pandemic period was associated with changes in molecular markers of biological aging. Among teenagers, the extent of perceived pandemic burden was linked to one measure of faster aging.
That raises a disturbing question. If the consequences of social separation can register in our biology, what happened to children who were still learning to read a smile, trust a stranger, recognize a worried face or navigate friendship?
What lockdowns changed—and why governments imposed them
There was no single worldwide “lockdown.” Governments and communities adopted different combinations, at different times, of stay-at-home orders, school and daycare closures, limits on gatherings, isolation and quarantine requirements, travel controls, physical distancing, face coverings, testing and later vaccination programs. Some children returned to classrooms quickly; others spent extended periods learning remotely. Some families had safe outdoor space and dependable internet; others did not.
In the early months, before vaccines and effective treatment options were widely available, limiting close contact was one way to reduce exposure and prevent hospitals from being overwhelmed. A 2023 Royal Society evidence review concluded that packages of nonpharmaceutical measures reduced SARS-CoV-2 transmission, although the evidence for individual components varied and school-closure effects depended on timing and circumstances.
The legitimate reason for a policy, however, does not make every consequence harmless. A measure can reduce infections while imposing costs on education, development, caregiving, employment and mental health. Evaluating those trade-offs is not the same as pretending the virus was harmless. Nor does documenting a developmental concern prove every restriction was unjustified.
The new evidence: pandemic stress and the body’s molecular clocks
Researchers led by Alicia Schowe at Germany’s Max Planck Institute of Psychiatry studied 1,028 people: 359 children, 385 adolescents and 284 young adults. The participants had provided saliva samples before Germany’s first major containment measures in March 2020 and again during the pandemic period, allowing a rare before-and-after comparison.
Scientists analyzed patterns of DNA methylation—chemical marks associated with gene regulation—using four epigenetic clocks: DunedinPACE, GrimAge2, Horvath and PedBE. These are statistical tools built to estimate different aspects of aging. They do not measure how old someone looks, and they do not show that their DNA sequence has been damaged or rewritten by loneliness.
Three measures showed significant changes in age acceleration or estimated pace of aging across the studied population. But the especially interesting finding was narrower: among adolescents, greater self-reported pandemic burden was associated with a greater increase in DunedinPACE, one measure of aging pace. The same subjective-stress association was not established in the younger children or adult group.
Pre-existing physical health and parental socioeconomic conditions were also associated with changes. Crucially, objective measures such as local infection rates or policy stringency did not establish a simple dose-response relationship: more restrictions did not automatically predict faster molecular aging. The Max Planck research summary highlights loss of peer interaction as a potentially important part of adolescents’ experience.
What the study does not show: that any teenager suddenly became biologically several years older, that masks caused the measured changes, that the changes are permanent, or that participants will develop cancer, diabetes or another disease sooner. There was no comparable unexposed population living through 2020 in the same circumstances. Changes in sleep, physical activity, illness, family finances and broader stress are difficult to disentangle.
It is evidence of a possible biological cost associated with a major societal crisis—not a verdict on one isolated policy.
Teen brains appeared to mature faster, too. Is that aging?
The molecular findings are not the first such warning. In 2024, University of Washington researchers published MRI research in PNAS comparing adolescent brain structure before and after pandemic disruptions. They modeled the expected age-related thinning of the cerebral cortex using scans collected before the pandemic, then compared the model with scans collected in 2021–2022.
The cortex normally changes throughout adolescence. Thinning is part of maturation, not evidence by itself of dementia or brain destruction. The researchers nevertheless found thinner-than-expected cortex in the post-lockdown sample, particularly among girls. Their modeled brain-age difference averaged 4.2 years for girls and 1.4 years for boys. The whole-brain estimate for boys was statistically uncertain, with a confidence interval that included zero; girls’ differences were broader and more robust.
Those numbers are striking, but they are not literal years stolen from a person’s lifespan. A brain-scan estimate and a DNA-methylation estimate are different biological measurements; they cannot be added together, and neither demonstrates a shortened life expectancy.
The study also could not isolate which pandemic experiences mattered most. Lost friendships, parental anxiety, altered sleep, diet, exercise, screen time and SARS-CoV-2 infection itself were potential contributors. Still, the finding is consistent with a larger literature linking adversity and chronic stress to altered developmental trajectories.
Adolescence is a time when peer relationships are not decorative extras. They provide identity, belonging, emotional feedback, conflict resolution and a rehearsal space for adult intimacy. Remove that environment, and the developing nervous system may notice.
The hidden-face question: what happens when children cannot see a smile?
This is one of the pandemic’s most emotionally resonant questions—and one requiring careful scientific language.
Children learn from faces. A raised eyebrow can signal warning. A smile may offer reassurance. A downturned mouth can reveal sadness, and visible lip movement contributes to understanding speech, particularly in noisy rooms or for people with hearing differences. The voice, eyes, posture, gesture and context also convey meaning.
A 2022 study of 131 children ages 7–12 found that recognizing some emotions in photographs was harder when the faces were masked, notably happiness, sadness and fear. That effect was particularly relevant for children reporting greater social disruption. The researchers were measuring a specific recognition task, not diagnosing permanent emotional impairment.
A more revealing 2025 Dutch study of 935 infants and young children compared brain responses recorded before and during pandemic restrictions. The children still distinguished faces from non-face objects: basic face categorization appeared resilient. But their neural responses to happy and fearful expressions differed, with less differentiation between those emotions in the pandemic-era group. That could reflect altered exposure, attention or familiarity with emotional faces.
There is the important distinction: research has detected changes in emotional-face processing, but it has not proved widespread, irreversible developmental stunting caused by face masks. The Dutch researchers did not directly measure later social functioning, and reduced contact with a variety of people—not masks alone—could account for the findings.
Moreover, children generally continued to see uncovered faces at home, and masked interactions did not remove tone of voice, eye expressions or body language. Some experiments found people adapted by relying more on information around the eyes. In a 2023 experimental study, masks impaired recognition of happiness and sadness but did not significantly impair basic language processing.
For infants and toddlers with limited hearing, communication difficulties or reduced access to responsive caregivers, obscured lips and quieter interactions could be an additional barrier. Yet it would be a mistake to equate a documented momentary difficulty reading a masked face with proof that millions of children lost the capacity for empathy.
The social distance that was not measured in feet
Physical distancing was described in meters and feet. Its psychological meaning was harder to quantify.
Young children learn when they approach another child, share toys, disagree, reconcile and discover that a stranger can be kind. Teenagers learn how to enter a room, read whether a friend is uncomfortable, and survive the awkwardness of being rejected or misunderstood. Adults rely on routine proximity too: the casual conversation after work, the relative who notices something is wrong, the handshake, the ordinary warmth of not having to arrange every encounter in advance.
During periods of restrictions, many of these small interactions disappeared or became constrained. People could be simultaneously protective of one another and fearful of one another. A cough could change the mood of a room. An invitation could feel risky. A hug could require a calculation.
Such experiences might plausibly reinforce vigilance, avoidance or social uncertainty in some individuals, particularly those already prone to anxiety. But researchers cannot yet assign a universal, lasting psychological trait to an entire pandemic generation. Children experienced the period differently: some felt lonely or unsafe; others found relief from school bullying, social pressure or exhausting commutes. Family relationships could deteriorate or deepen.
The World Health Organization’s 2025 report on social connection treats loneliness and isolation as serious health concerns in their own right. That is a reason to study pandemic-era separations carefully, not proof that every modern difficulty with friendship began in 2020.
Stress does not remain inside the mind
Stress is a biological process, not merely an unpleasant thought. Perceived threat can activate the sympathetic nervous system and the body’s stress-hormone pathways. When a stressful situation persists, sleep, appetite, attention, immunity and emotional regulation can all be affected. A child who feels unsafe may withdraw or become irritable; an adolescent who loses important relationships may experience low mood, restlessness or difficulty concentrating.
Large studies show that these concerns were not isolated anecdotes. A 2021 meta-analysis covering 80,879 young people estimated clinically elevated depression symptoms in 25.2% and anxiety symptoms in 20.5% during the pandemic’s first year. These are symptom-screening results, not confirmed diagnoses in every participant.
More importantly, a 2023 meta-analysis of 53 longitudinal studies, including more than 40,000 children and adolescents across 12 countries, found an increase in depressive symptoms from before to during the pandemic and a smaller increase in anxiety. The stronger design matters because it compares changes rather than only describing how teenagers felt at one point in time.
And yet the mental-health story did not freeze in 2021. The CDC’s September 2026 release reported improvement among U.S. high-school students: persistent sadness or hopelessness fell from approximately 40% in 2023 to 33% in 2025, while serious consideration of suicide also declined. A third reporting persistent sadness remains deeply concerning, but improvement matters. Youth mental-health difficulties also predated COVID, so current distress cannot all be assigned to lockdowns.
Physical costs: less movement, altered sleep, changing eyesight
The body remembers routines. When sports were cancelled, playgrounds closed or outdoor access became more difficult, some children lost daily movement that remote lessons could not replace.
A 2022 systematic review and meta-analysis of 22 longitudinal studies involving 14,216 children estimated a roughly 20% decline in total daily physical activity during the pandemic, including an average reduction of about 17 minutes per day in moderate-to-vigorous activity. It was a substantial change in exercise, play and time spent moving.
A CDC analysis of 432,302 U.S. children and teenagers found that the monthly rate of body-mass-index increase nearly doubled in the early pandemic compared with the preceding period. Those with higher weight before the pandemic often experienced larger changes. The result describes a studied health-care cohort, not every child in the country—and it is not a reason to stigmatize children over weight.
Eyes may also have been affected. A 2022 review of research on myopia found a faster shift toward nearsightedness in the pandemic period among school-aged children, alongside more near work and reduced outdoor time. Again, a change in lifestyle is a plausible contributor; the exact share attributable to individual restrictions remains uncertain.
Sleep and digital routines are part of the same story. The CDC’s 2025 youth survey found that only about one in four U.S. high-school students reported at least eight hours of sleep on an average school night. The survey does not prove the pandemic caused today’s sleep deficits, but it shows why restoring regular sleep, activity and face-to-face life remains important.
The classroom was also a social laboratory
School is not only where children learn fractions and spelling. It is where they read expressions, wait their turn, speak to adults outside their household, manage disappointment and find a community beyond the family home.
Remote learning kept instruction going under extraordinary conditions. But it often could not reproduce recess, hands-on work, special-education support, school meals, counseling, speech services or the informal help a teacher provides by noticing a struggling student.
A 2025 international analysis of student-assessment data associated longer school closures with larger learning losses and estimated average mathematics declines equivalent to about seven months of learning across the studied settings. Such conversions are model-based estimates—not a literal claim that every child lost seven months of knowledge.
In the United States, 2024 National Assessment of Educational Progress results showed fourth- and eighth-grade reading performance below prepandemic levels, with uneven math recovery. Some declines had already begun before COVID, and household conditions and instruction quality mattered alongside closure duration.
Children from families with quiet rooms, reliable broadband and adult help were not playing the same game as children sharing devices, navigating poverty or coping with unstable housing. Students with disabilities could face additional losses when therapies and tailored instruction were interrupted. The burden was not distributed evenly.
The trauma that came from the virus itself
There is an essential missing piece in any account that blames all harm on public-health rules: the disease itself harmed children and families. Some children became seriously ill or developed prolonged symptoms after infection. The NIH-funded RECOVER pediatric study documented symptom patterns associated with long COVID, including fatigue and problems spanning several body systems. Its research-index percentages should not be treated as population-wide prevalence estimates.
Others experienced the death of a parent or caregiver. A CDC-supported analysis released in October 2021 estimated that more than 140,000 U.S. children had lost a parent or primary or secondary caregiver through COVID-associated deaths by June of that year. For those children, the pandemic’s cost was not an abstract loss of social routine. It was the loss of someone who made home feel safe.
The harm from infection and bereavement must be considered alongside the harms from restrictions. Otherwise, the story becomes politically convenient and scientifically incomplete.
Did the pandemic stunt a generation?
The phrase “stunted development” suggests a single, permanent injury. The evidence supports a more cautious—and still serious—assessment.
Established or well supported: pandemic-era disruptions coincided with worse average emotional well-being, reduced physical activity, significant educational setbacks and measurable differences in some tests of face-emotion processing. Some biological measures of adolescent development and aging also deviated from expected patterns.
Plausible but unresolved: reduced face-to-face contact, limited social practice, household stress and altered routines contributed to those changes to differing degrees. The importance of the developing adolescent brain makes that hypothesis worth continued investigation.
Not established: that universal masking permanently damaged empathy, that lockdown rules alone explain faster epigenetic aging, that every child suffered developmental delay, or that molecular-clock changes will translate into earlier disease or death.
Neurodevelopment is also plastic. Children can learn social skills, rebuild friendships and receive effective educational and mental-health support. A difference observed during a crisis is not necessarily a permanent forecast for the rest of a child’s life.
What a future public-health emergency should learn from this one
Future decisions should treat preventing infection and protecting child development as simultaneous responsibilities, not opposing slogans. That means assessing the costs and benefits of each precaution as conditions change, investing in indoor air quality and ventilation, protecting safe in-person schooling where feasible, preserving outdoor play and accessible services, and avoiding unnecessary interruption of early intervention, disability support and mental-health care.
It also means paying attention to children who do not communicate distress in obvious ways. Teachers and caregivers can use clear verbal descriptions of emotions, visible gestures, accessible communication supports and regular opportunities for unpressured social interaction. When young children show persistent speech, hearing, learning or social difficulties, appropriate developmental assessment is more useful than an automatic assumption about their pandemic experience.
For teenagers, recovery may look less like a dramatic medical intervention than reliable relationships, counseling when needed, sleep, physical activity, stable routines, time away from relentless online comparison and places where friendship can unfold without constant supervision.
The most uncomfortable lesson may be that an emergency can have more than one set of victims. Public health measures helped limit transmission, while the pandemic and its restrictions also altered the conditions in which childhood unfolded. The precise contribution of each remains a question for research—not a question that should be shut down by politics.
We counted infections, hospitalizations and deaths because those were urgent and measurable. But who was counting the missed friendships, the muffled reassurances, the lost recesses, the nervous systems on alert—and the years needed to rebuild a sense of belonging?
Question. Investigate. Reconsider.
Primary studies and further reading
- Schowe et al., PNAS (2026): Epigenetic aging during the COVID-19 pandemic
- Corrigan et al., PNAS (2024): Adolescent cortical thinning after pandemic lockdowns
- Van den Boomen et al. (2025): Face and emotion processing in infants and young children
- Waller and colleagues (2022): Masking and emotion recognition in late childhood
- JAMA Pediatrics (2023): Longitudinal meta-analysis of child depression and anxiety
- JAMA Pediatrics (2022): Changes in child and adolescent physical activity
- Royal Society (2023): Evidence review of nonpharmaceutical interventions
- CDC (2026): Youth Health in Focus, based on 2025 survey data
