I want to tell you something I’ve noticed in my years as an instructional coach that I’ve never quite figured out how to say in a faculty meeting without sounding like I’m blaming everyone in the room.
The kids are struggling. Not academically — though that too — but in the more fundamental sense of not knowing how to handle conflict, setbacks, boredom, or problems that don’t have an adult available to solve them. I watch students freeze at the first sign of ambiguity. I watch them escalate minor peer disputes immediately to adults rather than working them out themselves. I watch them sit in unstructured moments — the rare ones — with visible discomfort, as if the absence of instructions were a failure of the environment rather than an invitation.
I don’t think this is a character problem. I think it’s a development problem. Specifically, I think we have systematically removed the conditions under which children develop the capacity to manage themselves — and we’ve done it with the best possible intentions.
This is, in short, exactly what research psychologist Peter Gray has been saying for over a decade. And the evidence behind it is more substantial than most people realize.
The Research Nobody Wants to Sit With
Gray is a research professor of psychology at Boston College and one of the foremost scholars on childhood play. His work takes an evolutionary approach: what do children’s play behaviors look like across cultures, across species, and across history — and what does that tell us about what play is actually for?
His central finding, documented across decades of research, is this: rates of anxiety, depression, and psychopathology in children and adolescents have been rising continuously since roughly 1955. Not spiking — continuously rising, decade after decade. And over that same period, children’s opportunities for free, independent, self-directed play have been continuously declining.
In 2023, Gray published a paper in the Journal of Pediatrics with co-authors David Lancy and David Bjorklund, laying out multiple lines of evidence for a causal relationship between these two trends. The paper summarizes decades of data on what independent activity does for children’s psychological development — and what its absence does to it.
The suicide rate among school-age children is now roughly six times what it was in the 1950s. The rate of mental health hospital admissions for children is double during the school year compared to summer, when school is not in session. These are not subtle correlations.
Gray also published a piece in The Lancet Child and Adolescent Health in July 2025, continuing this line of argument, cementing a body of work that is difficult to dismiss as ideological and increasingly hard to ignore.
What “Play” Actually Means (And What It Doesn’t)
The word gets muddied, so it’s worth being precise about what the research actually refers to.
Play, in Gray’s definition, is “an activity that is self-chosen and self-directed — something you do because you want to do it.” By this definition, organized sports are not play. Adult-supervised classroom activities are not play. Structured recess with rules enforced by a teacher is not play. Even a child following a parent’s suggested activity is not, in any meaningful developmental sense, playing.
This is a harder truth than it sounds. Many of the things we offer children as “play” aren’t. A soccer league is exercise, social development, and competition — all valuable — but the child isn’t deciding the rules, resolving disputes without a referee, or navigating the social dynamics of a group without adult mediation. Those are different skills, and they require genuinely unstructured time to develop.
As Gray explained in an NPR interview: “That’s how children develop the kinds of character traits that allow them to ultimately become independent adults. They learn how to deal with peers without an adult intervening. They learn how to deal with minor bullying… But if you’re always protected from bullies by some adult, you’re not learning how to deal with that yourself.”
This will make some readers uncomfortable. It made me uncomfortable the first time I really sat with it. Because it means that a significant amount of what we consider protective parenting and good schooling is actually developmental interference — removing children from the exact experiences through which they learn resilience, problem-solving, and self-regulation.
How We Got Here
Gray argues that the shift happened in overlapping waves.
Television first — it brought children indoors and somewhat isolated them. Then came the gradual cultural shift toward the belief that children develop best when guided and controlled by adults, which led to increased schooling, organized activities, and adult-directed time outside school. And then, beginning especially in the 1980s, a fear of allowing children outdoors unguarded.
That last shift is worth pausing on. What was considered ordinary parenting in 1970 — sending your kids outside in the morning and expecting them back for dinner — began to be coded as negligence by the 1990s. Not because the world became dramatically more dangerous (by most measures, it became safer), but because the perception of danger increased and because cultural expectations of parental supervision tightened.
The result: a childhood increasingly spent inside, in structured activities, under adult oversight, with every problem either solved by an adult or at least supervised by one.
As an educator, I see the downstream consequences of this every day. Students who’ve never had to navigate a disagreement without a teacher moderating it. Students who don’t know how to occupy themselves when there’s no assigned task. Students who experience normal adolescent difficulties — which is to say, all of them — as catastrophic rather than manageable, because they’ve never had low-stakes practice at managing things.
A Note on Smartphones — and Why I’m Skeptical of the Simple Story
There’s a more popular explanation for the youth mental health crisis right now, and I want to address it directly because I think it’s incomplete in ways that matter.
Jonathan Haidt’s The Anxious Generation has gotten enormous traction with its argument that smartphones and social media are the primary drivers of rising anxiety and depression in young people. I understand why it lands — the timing seems to correlate, the mechanism makes intuitive sense, and it offers a satisfying villain and a concrete solution (take away the phones).
But I find myself skeptical. Not because Haidt is wrong that screens can be harmful — they can be, particularly for young adolescents. But because the “phones did this” narrative has serious holes, and because I’ve watched every generation get this treatment.
I grew up in the 80s and 90s. For my generation, it was television and video games. We were told we were rotting our brains, becoming isolated, losing the capacity for real connection. For the generation before mine, it was rock music and comic books. Before that, novels — yes, actually, novels were once considered a moral hazard for young people. Every generation has its technological panic, and the panic is always most credible to the generation that didn’t grow up with the thing in question.
More importantly, the data on declining children’s mental health predates smartphones by decades. Gray’s 2011 paper in the American Journal of Play documented the trend using research from before the iPhone existed. The continuous rise in anxiety and depression among young people stretches back to the 1950s — long before social media, long before mobile devices, long before any technology Haidt is pointing at. If smartphones caused this crisis, what caused the preceding sixty years of the same trend?
Gray’s answer — which I find more compelling — is that the roots are structural, not technological. Children turned to screens partly because the alternative, outdoor independent play with peers, had already been progressively eliminated. The screen fills a vacuum. Restricting devices without restoring unstructured time is treating a symptom while leaving the underlying condition untouched.
I’m not saying screens are irrelevant. I’m saying the conversation we need to have is harder than “take away the phones,” and I worry that Haidt’s framing lets adults off the hook for decisions about schooling, scheduling, and supervision that are much more directly within our control — and much more directly implicated in the data.
What This Means in Schools
I work with teachers and students in a district that, like most, has systematically reduced recess time, eliminated free periods, increased structured supervision, and added academic programming into spaces that used to be unscheduled. We’ve done this in the name of academic achievement and safety. These are not bad intentions.
But Gray’s research suggests we are making a significant developmental trade-off that we rarely acknowledge. Every minute of structured, adult-supervised activity that replaces unstructured, child-directed time is a minute of developmental practice that doesn’t happen.
Some concrete implications for schools:
Recess should be genuine and protected. Not a supervised activity period. Not an extension of PE. Unstructured time outside where children decide what they’re doing and work out their own problems.
Conflict is developmental, not just a management problem. When children argue on the playground, our instinct is to intervene immediately. Sometimes that’s necessary. Often it’s counterproductive — we’re preventing them from learning that they can resolve conflict themselves.
The impulse to fill every moment is worth examining. Not every unstructured moment is lost instructional time. Some of it is essential developmental time that looks unproductive from the outside.
The school year vs. summer mental health data is important. The rate of mental health admissions being double during the school year compared to summer suggests that something about how we structure the school day contributes to the problem, not just external factors. This should prompt uncomfortable conversations.
What Parents Can Do
If you’re a parent reading this, the research points in a fairly clear direction — and it’s one that runs counter to most dominant cultural pressures around parenting.
Let them be bored. Boredom is not a problem to solve. It’s the precondition for self-directed activity. A child who says “I’m bored” and hears “go figure something out” is getting something valuable.
Let them go outside without you. Age-appropriate independence varies, and you know your child and neighborhood. But the research is clear that children need time in the world without adult oversight, and the age at which we tend to grant this has been steadily rising in ways that don’t match the developmental data.
Let them work it out. When your child has a conflict with a friend, resist the impulse to mediate immediately. Ask what they’ve tried. Ask what they think they could do. Be available but not the first resort.
Reduce the schedule. Organized activities have real value. But the balance most children currently have — heavily scheduled with structured activities from after school through evening — leaves almost no time for the self-directed play that Gray identifies as essential.
The Books Worth Reading
Free to Learn by Peter Gray — The essential text. Gray’s full argument was written for a general audience. Accessible, well-researched, and genuinely challenging to conventional assumptions about schooling and child development. If you read one book off this list, make it this one.
Last Child in the Woods by Richard Louv — An earlier and influential book on what Louv calls “nature-deficit disorder” — the developmental consequences of children spending less time in unstructured outdoor environments. Published in 2005, it was ahead of its time and holds up well.
The Anxious Generation by Jonathan Haidt — I’d recommend reading this one critically rather than taking it as a settled truth. Haidt’s data on smartphones and social media is real and worth engaging with, but his causal claims are contested, and the deeper structural issues that Gray identifies predate the technology Haidt focuses on. Read it, but read it alongside Gray, not instead of him.
How Children Learn by John Holt — A classic, first published in 1967, that still challenges almost everything about how we structure formal education. Holt observed children learning and drew conclusions that the education system has largely ignored ever since. Gray cites him approvingly, and for good reason.
The Honest Reckoning
I’m writing this as someone who works in a system — public K-12 education — that is structurally oriented toward exactly the kind of adult-oversight-heavy, risk-managed, achievement-focused environment that Gray’s research implicates. I’m not exempt from this critique. Nobody in education is.
The honest reckoning is that we’ve built schools that optimize for the things adults can measure and control, and in doing so we’ve progressively eliminated the developmental experiences that children need most and that are hardest to schedule, supervise, or assess.
The children are trying to tell us something. The data is trying to tell us something. The question is whether we’re willing to hear it.
Related on this site: Why aren’t people reading books anymore? — a parallel argument about what we lose when children’s time becomes entirely structured and output-focused.
The Eclectic Educator is a free resource for everyone passionate about education and creativity. If you enjoy the content and want to support the newsletter, consider becoming a paid subscriber. Your support helps keep the insights and inspiration coming!