The Anxious Generation: How the Great Rewiring of Childhood Is Causing an Epidemic of Mental Illness

by Jonathan Haidt

Even small mistakes can bring heavy costs in a viral world where content can live forever and everyone can see it. Mistakes can be met with intense criticism by multiple individuals with whom one has no underlying bond. Apologies are often mocked, and any signal of re-acceptance can be mixed or vague. Instead of gaining an experience of social mastery, a child is often left with a sense of social incompetence, loss of status, and anxiety about future social interactions.

This is why there is no contradiction when I say that parents should supervise less in the real world but more in the virtual—primarily by delaying immersion.

A related factor is a declining sense of social cohesion throughout the late 20th century, which had many causes. When people no longer knew their neighbors, they no longer had “eyes on the street” from adults who could look out for kids.[38] But perhaps the most important change in the 1980s was the rising fear among parents that everyone and everything was a threat to their children.

He analyzes changing parental behavior as a response to social, economic, and technological changes in the 1980s and 1990s: for example, the rise of cable TV (and 24/7 news cycles) and its ability to spread stories that frighten parents; the rising number of women working and the corresponding increase in day care and after-school programs; and the increasing influence of parenting “experts,” whose advice was often a better reflection of their social and political views than of any scientific consensus.

Children were taught to fear unknown adults, particularly men.

At the same time, adults internalized the reciprocal message: Stay away from other people’s children. Don’t talk to them; don’t discipline them if they are misbehaving; don’t get involved.

But when adults step away and stop helping each other to raise children, parents find themselves on their own. Parenting becomes harder, more fear-ridden, and more time consuming, especially for women,

But in a psychologically safe group, members can disagree with each other and criticize each other’s ideas respectfully. That’s how ideas get vetted. What emerged on campus as emotional safety, in contrast, was a much broader concept that came to mean this: I should not have to experience negative emotions because of what someone else said or did. I have a right not to be “triggered.”

We used the term “safetyism” to refer to “a culture or belief system in which safety has become a sacred value, which means that people become unwilling to make trade-offs demanded by other practical and moral concerns. ‘Safety’ trumps everything else, no matter how unlikely or trivial the potential danger.”

parents (on average) say children should be at least 10 years old to play unsupervised in their own front yard.[54] They say that kids should be at least 12 years old before being allowed to stay alone in their own home unsupervised for one hour. They say that kids should be 14 before being allowed to go, unsupervised, to a public park. And these respondents include the same Gen X and baby boom parents who say, gleefully and gratefully, that they were let out, in a much more dangerous era, at ages 6, 7, or 8.

The young child’s brain has enormous potential (it can develop in many ways) but lower ability (it doesn’t do most things as well as an adult brain).

developmental psychologist Laurence Steinberg notes that adolescence is not necessarily an especially stressful time. Rather, it is a time when the brain is more vulnerable to the effects of sustained stressors, which can tilt the adolescent into mental disorders such as generalized anxiety disorder, depression, eating disorders, and substance abuse.

Puberty is therefore a period when we should be particularly concerned about what our children are experiencing.

Because children are antifragile, it is essential that those experiences involve some fear, conflict, and exclusion (though not too much). Safetyism is an experience blocker. It prevents children from getting the quantity and variety of real-world experiences and challenges that they need.

“there is an inverted U-shaped pattern in the relationship between stress and well-being. A little stress is beneficial to development, but a lot of stress, acute or chronic, is detrimental.”

In fact, smartphones and other digital devices bring so many interesting experiences to children and adolescents that they cause a serious problem: They reduce interest in all non-screen-based forms of experience.

Are screen-based experiences less valuable than real-life flesh-and-blood experiences? When we’re talking about children whose brains evolved to expect certain kinds of experiences at certain ages, yes.

The fact that most societies used to have such rites suggests to me that our extremely new secular societies may be losing something important as we abandon public and communally marked rites of passage. A human child doesn’t morph into a culturally functional adult solely through biological maturation. Children benefit from role models (for cultural learning), challenges (to stimulate antifragility), public recognition of each new status (to change their social identity), and mentors who are not their parents as they mature into competent, flourishing adults.

This suggests that there may be a deep need among many adolescents for belonging and for the rites and rituals that create and express that belonging. Can we use that knowledge to improve adolescents’ transitions to adulthood?

Pornography sites also welcome children, as long as they click a box to say that they are 18 or older. Porn sites will show them how to have anal sex long before they’ve had their first kiss.

Once a child gets online, there is never a threshold age at which she is granted more autonomy or more rights. On the internet, everyone is the same age, which is no particular age.

Importantly, these would mostly be norms, not laws, which any parent could choose to follow or ignore. Engaging in commonly held norm-based rituals and sharing milestones might be more effective than practices that each family invents for itself.

As an initial proposal, to start a conversation, I suggest that we focus on even-year birthdays from ages 6 to 18. We might make a big deal out of those birthdays by linking them to new freedoms, new responsibilities, and significant increases in allowance. We want children to feel that they are climbing a ladder with clearly labeled rungs, rather than just having an annual party with games, cake, and presents.

The U.S. Congress should undo the mistake it made in 1998 when it made 13 the age at which children can sign contracts with corporations to open accounts and give away their data without their parents’ knowledge or consent. I believe the age should be raised to 16 and enforced. The 16th birthday would become a major milestone at which we say to teens, “You can now get a driver’s license, and you can now sign certain kinds of contracts without any legal requirement for parental consent. You can now open social media accounts as well.” (There are good arguments for waiting until 18, but I think 16 would be the right minimum age to be established by law.)

She called out to me: “Daddy, can you take the iPad away from me? I’m trying to take my eyes off it but I can’t.” My daughter was in the grip of a variable-ratio reinforcement schedule administered by the game designers, which is the most powerful way to take control of an animal’s behavior short of implanting electrodes in its brain.

“The one impulse, out of many accidental ones, which leads to pleasure, becomes strengthened and stamped in.” He said that animal learning is “the wearing smooth of a path in the brain, not the decisions of a rational consciousness.”[1]

Social media has evolved over time,[7] but there are at least four major features common to the platforms we generally think of as being clear examples of social media: user profiles (users can create individual profiles where they can share personal information and interests); user-generated content (users create and share a variety of content to a broad audience, including text posts, photos, videos, and links); networking (users can connect with other users by following their profiles, becoming friends, or joining the same groups); and interactivity (users interact with each other and with the content they share; interactions may include liking, commenting, sharing, or direct messaging).

First and foremost, in 2009, Facebook introduced the “like” button and Twitter introduced the “retweet” button. Both of these innovations were then widely copied by other platforms, making viral content dissemination possible. These innovations quantified the success of every post and incentivized users to craft each post for maximum spread, which sometimes meant making more extreme statements or expressing more anger and disgust.[8] At the same time, Facebook began using algorithmically curated news feeds, which motivated other platforms to join the race and curate content that would most successfully hook users. Push notifications were released in 2009, pinging users with notifications throughout the day. The app store brought new advertising-driven platforms to smartphones. Front-facing cameras (2010) made it easier to take photos and videos of oneself, and the rapid spread of high-speed internet (reaching 61% of American homes by January 2010[9]) made it easier for everyone to consume everything quickly.

Even users who don’t actively post are affected by the incentive structures these apps have designed.[10]

As Jean Twenge has shown, teens who spend more time using social media are more likely to suffer from depression, anxiety, and other disorders, while teens who spend more time with groups of young people (such as playing team sports or participating in religious communities) have better mental health.[20]

A 2014 survey of children ages 6–12, conducted by Highlights magazine, found that 62% of children reported that their parents were “often distracted” when the child tried to talk with them.[23] When they were asked the reasons why their parents were distracted, cell phones were the top response.

A review of 36 correlational studies found significant associations between high social media use and poor sleep, and also between high social media use and poor mental health outcomes.[34]

One experiment found that adolescents who restricted their use of screen devices after 9 p.m. on school nights for two weeks showed increased total sleep time, earlier sleep onset times, and improved performance on a task that required focused attention and quick reactions.[35]

“You know, think about it, when you watch a show from Netflix and you get addicted to it, you stay up late at night. We’re competing with sleep, on the margin.”

For example, a 2020 study found that greater sleep disturbance and shorter total sleep time were associated with greater internalizing scores (which include depression), as well as greater externalizing scores (which include aggression and other antisocial actions associated with a lack of impulse control).[41]

When you add it all up, the average number of notifications on young people’s phones from the top social and communication apps amounts to 192 alerts per day, according to one study.[42] The average teen, who now gets only seven hours of sleep per night, therefore gets about 11 notifications per waking hour, or one every five minutes. And that’s just for the apps that are about communication.

But even when students don’t check their phones, the mere presence of a phone damages their ability to think. In one study, researchers brought college students into the lab and randomly assigned them to (1) leave their bag and phone out in the entry room of the lab, (2) keep their phone with them in their pocket or bag, or (3) put their phone on their desk next to them. They then had the students complete tasks that tested their fluid intelligence and working memory capacity, such as by solving math problems while also remembering a string of letters. They found that performance was best when phones were left in the other room, and worst when phones were visible, with pocketed phones in between. The effect was bigger for heavy users. The article was titled “Brain Drain: The Mere Presence of One’s Own Smartphone Reduces Available Cognitive Capacity.” [48]

A Dutch longitudinal study found that young people who engaged in more problematic (addictive) social media use at one measurement time had stronger ADHD symptoms at the next measurement time.[50]

The loop starts with an external trigger, such as a notification that someone commented on one of her posts. That’s step 1, the off-ramp inviting her to leave the path she was on. It appears on her phone and automatically triggers a desire to perform an action (step 2) that had previously been rewarded: touching the notification to bring up the Instagram app. The action then leads to a pleasurable event, but only sometimes, and this is step 3: a variable reward. Maybe she’ll find some expression of praise or friendship, maybe not. This is a key discovery of behaviorist psychology: It’s best not to reward a behavior every time the animal does what you want. If you reward an animal on a variable-ratio schedule (such as one time out of every 10 times, on average, but sometimes fewer, sometimes more), you create the strongest and most persistent behavior.

At this point, after investment, the trigger for the next round of behavior may become internal. The girl no longer needs a push notification to call her over to Instagram. As she is rereading a difficult passage in her textbook, the thought pops up in her mind: “I wonder if anyone has liked the photo I posted 20 minutes ago?” An attractive off-ramp appears in consciousness (step 1). She tries to resist temptation and stick with her homework, but the mere thought of a possible reward triggers the release of a bit of dopamine, which makes her want to go to Instagram immediately. She feels a craving. She goes (step 2) and finds that nobody liked or commented on her post. She feels disappointment, but her dopamine-primed brain still craves a reward, so she starts looking through her other posts, or her direct messages, or anything that shows that she matters to someone else, or anything that provides easy entertainment, which she finds (step 3). She wanders down her feed, leaving comments for her friends along the way. Sure enough, a friend reciprocates by liking her last post. An hour later, she returns to her study of photosynthesis, depleted and less able to focus.

Unfortunately, when an addicted person’s brain adapts by counteracting the effect of the drug, the brain then enters a state of deficit when the user is not taking the drug. If dopamine release is pleasurable, dopamine deficit is unpleasant.

Lembke says that “the universal symptoms of withdrawal from any addictive substance are anxiety, irritability, insomnia, and dysphoria.”[57] Dysphoria is the opposite of euphoria; it refers to a generalized feeling of discomfort or unease. This is basically what many teens say they feel—and what parents and clinicians observe—when kids who are heavy users of social media or video games are separated from their phones and game consoles involuntarily.

When we gave children and adolescents smartphones in the early 2010s, we gave companies the ability to apply variable-ratio reinforcement schedules all day long, training them like rats during their most sensitive years of brain rewiring. Those companies developed addictive apps that sculpted some very deep pathways in our children’s brains.[61]

They also confer some unique benefits for specific groups such as sexual minority youth and those with autism—where some virtual communities can help soften the pain of social exclusion in the real world.[65]

Dopamine release is pleasurable, but it does not trigger a feeling of satisfaction. Rather, it makes you want more of whatever you did to trigger the release.

On the most prototypical platforms, like Instagram, users post content—often about themselves—and then wait for the judgments and comments of others. Such posting and waiting, along with social comparison, is having larger and more harmful effects on girls and young women than on boys and young men, and this difference shows up consistently in many correlational studies.

There is a clear, consistent, and sizable link[7] between heavy social media use and mental illness for girls,[8] but that relationship gets buried or minimized in studies and literature reviews that look at all digital activities for all teens.[9]

The more time a girl spends on social media, the more likely she is to be depressed. Girls who say that they spend five or more hours each weekday on social media are three times as likely to be depressed as those who report no social media time.

Another study randomly assigned teen girls to be exposed to selfies taken from Instagram, either in their original state or after modification by the experimenters to be extra attractive. “Results showed that exposure to manipulated Instagram photos directly led to lower body image.”[14]

This meant that they made eye contact less frequently, laughed together less, and lost practice making conversation. Social media therefore harmed the social lives even of students who stayed away from it. These group-level effects may be much larger than the individual-level effects, and they are likely to suppress the true size of the individual-level effects.[18]