Are Phones to Blame for Teen Anxiety? Gray Challenges Haidt
Peter Gray's Restoring Childhood: How to Set Kids Free in the Age of Anxiety will be published on September 15, 2026, as a direct reply to Jonathan Haidt's account of teen mental health.
Haidt, a social psychologist, published Anxious Generation: How the Great Rewiring of Childhood is Causing an Epidemic of Mental Illness in 2024. The book was a New York Times bestseller. It argues that social media and smartphone use is largely to blame for the decline in teen mental health since 2010 The Verge.
Haidt leads The Anxious Generation Movement, described as research and action to reverse the phone-based childhood and support healthier development. The project states that constant notifications, endless scrolling and online comparison wire kids for anxiety, depression and attention fragmentation. Haidt has also written a consensus statement on potential negative effects of smartphones and social media on teen mental health. The thesis has reached policy, and Haidt has been credited with helping inspire Australia's nationwide teen social media ban. The dispute is about evidence, and it matters for platforms, schools and parents deciding where to act.
Gray's answer, to be published by Penguin Random House as a 352-page hardcover listed at $32.00 Penguin Random House, points elsewhere. He argues the decline came from decades of reduced autonomy and rigid schooling linked to Common Core, not phones and social media The Verge.
Gray flips the usual internet timeline. He argues home internet use in the 1990s and 2000s improved teen mental health by restoring freedom and connection. In that account, computers, video games and the internet helped restore childhood in the 1990s by giving children a space for self-directed play and peer contact outside adult supervision.
The methods fight is about correlation, or whether two trends move together. Gray told The Vergecast that Haidt's book ignored dozens of meta-analyses, studies that combine many prior studies into one estimate, which found no meaningful link between time on social media or smartphones and mental health. Adolescent researcher Candice Odgers wrote a critical review of Haidt's analysis in Nature, a point Gray and his supporters cite when questioning the causal claim.
Gray cites different data. A University of South Florida study found 11-to-13-year-olds who owned smartphones were more likely to spend in-person time with friends. Gray says studies consistently find children name school pressure, not social media, as their top source of anxiety.
Federal data give context without settling cause. The Centers for Disease Control and Prevention reports 20% of adolescents received mental health therapy and 20% of those ages 12-17 had unmet mental health needs CDC. Teenagers with higher non-schoolwork screen use were more likely to have adverse health outcomes CDC. High screen time has been linked with poor sleep, fatigue and anxiety symptoms CDC. From 2013 to 2023, 10-year trends in the CDC YRBS Data Summary and Trends Report were similar to data shown in 2021 CDC. Depression in the past two weeks was 13.1% in people age 12 and older CDC.
The broader context here will be familiar to readers who have watched earlier debates about youth and technology. Each shift, from home PCs to the web to smartphones, first brought confident claims about harm or benefit based on time-use correlations, followed by slower work that separated access from specific factors like sleep disruption, harassment, social comparison or loss of unsupervised play.
In my view, Gray's challenge helps because it demands detail about mechanisms. If the drivers are notification design, infinite scroll, algorithmic comparison and nighttime use that harms sleep, those are design problems that can be measured, tested and fixed. If the drivers are loss of autonomy, homework load, testing and less in-person time, then device bans alone will miss the target. I watched my own two children move from a shared family PC to personal phones, and what mattered was less the hours than whether the tool helped them make plans with friends. For technologists, the next step is better measurement. Totals for screen time are blunt, like counting hours outdoors without asking what a child did there. Telemetry on session length, interruption rate and nighttime use, joined to validated mental health surveys and school climate data, would support stronger conclusions than self-reports recalled later. That work points to what this enables, tools that keep connection and autonomy while reducing fragmentation.
