Why Teenagers Hit Puberty and Take Dumb Risks

In adolescence, the brain’s reward centers light up when acting recklessly in front of peers.

Issei Kato / Reuters

Why do teenagers do so many stupid things when other teens are around?

This week, I’m sharing a variety of responses to the question, “What insight or idea has thrilled or excited you?” Professor Laurence Steinberg, an expert on the adolescent brain, teen risk-taking, and psychological development, recalls two times when research findings particularly excited him. The first happened early in his career:

I had done my dissertation on the ways in which family relationships change as kids go through puberty. The main finding is that as kids go through puberty, conflict between them and their parents (especially mothers) intensifies. It wasn’t just getting older that did this––it was actually going through puberty.  

A few years after I had published this, I was visiting a friend who was teaching at a different university, and he asked how I knew that going through puberty was causing the increase in conflict, and not the reverse? I thought this seemed like a crazy idea, but when I returned home, I pulled out the data and reanalyzed it. Sure enough, it looked like fighting with parents actually accelerated the onset of puberty. No one had ever suggested such a thing before. I was out at Stanford giving a talk on the subject, and an eminent psychologist came up to me and said, “That was a lovely talk, but I don’t believe any of it.” That was in the late 1980s.

The finding has since been replicated dozens of times.

The second example that he related happened more recently, during a series of studies he was doing on the influence of an adolescent’s peers on the risks he or she takes:

Adolescents many more risks than either children or adults. It’s a huge public-health problem, since the things that teenagers are hurt by (or die from) generally aren’t diseases or illnesses, but reckless behaviors, like risky driving.

Most of us know, from our own experiences growing up or from watching our kids grow up, that adolescents do a lot of risky things when they are with their friends that they would never do if they were alone. (Crash rates among teen drivers increase exponentially as a function of how many other teenagers are passengers in the car.) We ran a series of experiments in which we had people of different ages play a series of risk-taking games, including a video driving game, on the computer, after randomly assigning them to be alone or in the presence of two friends.

We found that having friends in the same room doubled the number of risks that teenagers took but had no effect on adults. We then repeated this experiment using brain imaging: we scanned people while playing the same games either with or without peers able to see their performance on a monitor in another room. Not only did we once again find that the presence of peers increased risk taking among adolescents but not adults––we also found that when peers were watching, this lit up reward centers in the adolescents’ brains but not in the adults’ brains, and that the more these centers were activated, the more risks teenagers took.

This was interesting because the peers weren’t communicating with the game players—the players simply knew that they were being watched by their peers––so this ruled out the possibility that the reason kids take risks when they are with their friends is that their friends explicitly encourage them to. This study was reported on in the New York Times. Still, the study left open the possibility that, even if adolescents couldn’t talk to their friends, they still knew what their friends expected them to do, and they behaved more recklessly to impress their pals.

We struggled to think of how we might get around this. Where could we find a group of adolescents who weren’t capable of imagining what their friends wanted them to do? I thought for weeks about this and then asked a colleague of mine, who studies brain development in mice, if it was possible to replicate this study with rodents. All mammals go through puberty, so all of them go through a stage that resembles human adolescence, at least biologically. We designed a study in which we raised mice from birth in groups of three, creating “peer groups.”

Then we tested them by putting them in a cage where they had unlimited access to alcohol to see how much they would drink (this was a strain of mice that likes to drink). Half were tested as “adolescents” (shortly after puberty) and half were tested as adults. Within each age group, half were tested alone and half with their “friends” in the cage with them. When with their peers, adolescent mice drink more alcohol than when alone. But being in front of other mice has no such effect on adult mice.

In essence, we had discovered that the effect of peers on adolescent risk taking may be hard-wired in the adolescent brain because of the impact of peers on the brain’s reward centers. Just being around your friends is so rewarding that it makes you do crazy things. Now that was pretty exciting.

Email conor@theatlantic.com to share an idea or insight that has thrilled or excited you.