A Delicate Balance: Risks, Rewards, and the Adolescent Brain



by Carl Sherman

July 10, 2019

Among adolescents, unintentional injury (primarily motor vehicle accidents) is the leading cause of death, suicide and homicide follow, and excessive drinking, unprotected sex, and assorted misadventures leave a trail of turmoil. Why are otherwise smart, savvy young people notoriously prone to unwise, often impulsive actions that put themselves and others at risk? The question has engaged a broad range of research from which a complex, nuanced picture is emerging.

Beyond the biological and environmental interactions that characterize adolescent brain development, researchers are teasing apart individual differences: why only some teens are risk takers, only some of the time. But there’s also increasing respect for the necessity of risk-taking for adolescents to become well-functioning adults. Understanding the process more fully, many hope, could lead to strategies that buffer this vital and vulnerable stage of development against serious harm.

A Question of Connection

“Over a decade ago, you’d read articles suggesting that adolescents engage in high-risk behavior because the prefrontal cortex [a key brain area for judgment and self-control] was not fully developed,” says B.J. Casey, Ph.D., director of the Fundamentals of the Adolescent Brain (FAB) lab at Yale University, and a Dana Alliance for Brain Initiatives member. “But [that area] is even less developed in children who don’t engage in such behavior. We now think more in terms of neural circuitry; how regions of the brain talk to one another.”

Research at Casey’s lab and elsewhere tells a story in which the evolution of physical and functional connections across the brain can help us understand the perils and successes on the road from adolescence to adulthood.

“We came up with a model of imbalance within a widespread circuit: various regions are activated and the one that screams the loudest wins,” Casey says. Prominent within this decision-making circuit, in her view, are the emotionally reactive ventral striatum/nucleus accumbens, which responds to reward and the anticipation of reward, and cortical areas that inhibit impulses and regulate behavior.

The reward system, as Casey and others have shown, shows marked changes by adolescence and is highly reactive to cues of value in that period. “It’s really banging away, while the prefrontal cortex is not quite fully developed,” she says. Problems arise primarily in emotionally charged situations. “Adolescents are quite capable of making rational decisions… they just have more difficulty in the heat of the moment.” A measured appreciation of severe long-term consequences is no match for the immediate gratification promised by fast driving, excessive drinking, or unprotected sex, when the brain is in this mode, she suggests.

Beatriz Luna, Ph.D., professor of psychiatry and director of the Laboratory of Neurocognitive Development at the University of Pittsburgh, agrees that cortical brain structures that tell us how to act have pretty much matured to adult levels, and adolescents are capable of engaging these systems to regulate behavior. “But they’re not doing so in a controlled, sustained, reliable fashion,” she says.

In particular, she says, the cortical “alarm system” that warns the individual of errant behavior is not entirely up to speed, promoting the sensation-seeking and need for novelty that characterize adolescence. But what may seem like a defect is essential for development, she emphasizes. Learning about the world through such forays—which may involve risk—is “adaptive; it’s what needs to happen. Having a really effective alarm system would not be good.”

Variability in teen behavior—sometimes mature decision-making, sometimes irrational impulse—reflects the same developmental imperative. “The brain is experimenting, turning the volume of various regions up and down, till it finds its optimal performance range,” Luna says.

In terms of overall brain structure and function, her research points to differences in integration between parts of the brain. Individual systems—the network that regulates response to reward, the one that governs senses and movement, the decision-making system—are already organized on adult levels. But communication among these systems, which enables collaborative, effective judgment and action, is still a work in progress.

An fMRI study in Luna’s lab linked coordinated brain activity and self-control. On a test of the ability to inhibit a response—looking away from a flash of light, versus the reflexive tendency to look toward it—children performed much worse than adults, with adolescents in between. The fMRI data showed an age-associated increase in how smoothly frontal regions and lower sensorimotor brain areas worked together to perform the task. What’s more, the analysis linked stronger signals from the higher brain to superior inhibition.

Anatomically, such changes reflect the process of myelination—the development of an insulating sheath that allows neurons to carry messages faster and more efficiently across the brain. While myelination generally increases from adolescence to adulthood, Luna and her colleagues found that some of these white matter tracts shrink. “We’ve found dramatic decreases in connectivity from the limbic system [the network that regulates emotion and reward] to the prefrontal cortex,” she said, which may account in part for the diminished influence of emotion on judgment and decision-making that comes with maturity.

Dynamic changes in white matter, progressive integration between brain systems, an emerging ability to match response to situation—all these suggest extraordinary neuroplasticity during adolescence, which allows the brain to sculpt circuits equal to the demands of adult life. Luna calls this a “critical period” similar to the early years when sensory input organizes the brain for acute vision and hearing.

“Sensation seeking and risk-taking peak during adolescence across cultures, even across species,” she says. “Why? The brain needs to get the greatest amount of information in order to specialize properly.”

Who Takes Risks When

“A lot of risk-taking occurs during adolescence, but not all adolescents are risk-takers,” observes Adriana Galván, Ph.D., director of the Galván Laboratory for Developmental Neuroscience at UCLA. “It’s important not to lump all teenagers together.”

While a doctoral candidate at B.J. Casey’s lab, Galván was part of a team that analyzed child, adolescent, and adult brain activity and the propensity for risk. “When we looked at the data, there was a lot of variability within groups,” she says. People who reported more risk-taking behavior—at any age—showed a neurobiological correlate in the reward response.” In particular, nucleus accumbens-frontal cortex activity rose more, during a money-winning game, in teens who were more prone to sexual escapades, heavy drinking, high-impact sports, and the like.

Galván has been exploring how individual differences might play out in the real world. There’s abundant evidence that stress can disrupt decision-making in general and, in adolescents in particular, to amplify the tendency toward risk-taking. “But there are great individual differences in stress response and perception,” she says.

An fMRI study in Galván’s lab confirmed that scores on a risk-taking exercise go up in male, but not female adolescents under high stress, and associated the increase with prefrontal cortex deactivation.

Another study found that poor sleep amplified the risk-raising effect of stress, while extra sleep softened it.

In dissecting the complexities of risk-taking, Laurence Steinberg, Ph.D., Distinguished University Professor of psychology at Temple University, is taking a closer look at a factor of well-recognized importance in teen life: peer influence.

In one series of experiments, adolescents and adults performed similarly in a simulated driving test. In the presence of two friends, however, teen risk-taking and its consequences—they ran more lights and had more crashes—rose dramatically, while adult performance was unaffected.

The difference that peers make, according to fMRI data collected during a risk-taking exercise, appears linked to increased activity in striatal areas that heighten the rewards of sensation-seeking. The presence of friends did not, on the other hand, weaken cortical circuits responsible for inhibition, or make teens more impulsive.

Peer influence on risk-taking has its subtleties. “We’ve tried to identify circumstances under which it’s more or less powerful,” Steinberg says. In various studies, same-sex peers seemed to increase risky behavior, but the presence of a long-term romantic partner led male adolescents to take fewer risks than when by themselves. In the company of a new, attractive female acquaintance, on the other hand, they took more risks.

Other studies have looked at the influence of a slightly older person. Put four college kids together, they’ll take more risks than when alone. If one of the group is in his mid-20s, however, the peer effect is nullified.

It may be possible to reduce peer-linked dangerous risk-taking through training, Steinberg says. One study showed that teens who went through a four-week program of memory exercises designed to strengthen self-control were more resistant to the influence of others than a control group.

Another approach might be to reduce reckless behavior while encouraging the upside of the adolescent need for novelty and stimulation, Steinberg says. “The same sensation-seeking drive powers positive risk-taking—trying out for a part in a school play, taking more challenging classes than usual. A difference is that negative risk-taking is linked to impulsivity, while positive is not.”

“If we were to find ways to engage teens in more positive risk-taking, would this lead to a decline in negative risk-taking?” he asks. “It’s an interesting and important question.”

Experience Matters

Overall, complexity is a key theme in risk research. “There’s nothing in human behavior that’s simply biological or environmental,” says Dana Alliance member Abigail Baird, Ph.D., professor of psychological science at Vassar College, in a lecture on YouTube. “Puberty is a biological phenomenon… Adolescence is the social, cultural, emotional construction of it.” To understand risk-taking one must appreciate all these factors, she suggests.

A survey of adolescents across the world by Steinberg and colleagues found that the propensity for risk was similar in societies as diverse as Kenya, Thailand, and the US. But real-life patterns of behavior differed. “Juvenile crime rates, drug and alcohol use vary,” says Steinberg. “Adolescence may be a time of risky behavior, but how it manifests varies from culture to culture.”

“We know that experience matters,” he says. “What we’re just beginning to study as a field is how it plays out in the brain.”

An experiment in Baird’s lab suggests one answer. Researchers compared brain activity in adults and adolescents when asked to rate whether various scenarios were a good or bad idea. Both groups wisely rejected such notions as “swimming with sharks,” “biting a light bulb” or “jumping off a roof”—although adults did so significantly faster. The difference was in the underlying mental processes: adults showed greater activation in the visual cortex and the insula (a brain area that translates thoughts into visceral sensations), while the prefrontal cortex worked harder in teens. Put simply, adults could visualize the prospect and respond immediately, while teens had to mull it over, Baird suggests.

“What I think happens is that teens just haven’t had enough experience to develop the gut feelings that grown-ups can use to make decisions without thinking.” An adult who has cut her hand on a glass might be able to visualize and physically respond to the idea of “biting a light bulb” in a way a teenager can’t.

“The adolescent brain is hungry for experience—it’s a biological imperative,” and teens can’t develop adult capacities without taking risks, says Baird. “The trick is to help them train up their system with informative, not lethal experience. One of my greatest concerns is that many parents are inadvertently harming their children by trying to protect them too much. I’d much rather a kid fall off her bike than crash the car.”  

Social changes compound the problem, she suggests. Close-knit communities once enabled adolescents to take risks in a controlled fashion: “free-ranging kids used to be a viable option, when there was a local community that supported and protected them.”

In today’s world, “[i]t’s tricky for parents to find the right balance between being overprotective and ensuring their kids will get the experience that will help them thrive,” she concludes. Biological differences in temperament—some kids are born inhibited and need to be encouraged to take risks, others need to be roped in—add to the parenting challenge.

Baird’s own observations affirm the need for more balance. “Having been a college professor and house fellow—we live in an apartment attached to a dorm—I’ve seen first year students get younger and younger socially, increasingly scared of ideas, of disappointment, failure, breakups.” Although they’ve largely caught up by their senior year, “I still see them call their parents when leaving the classroom.”

Do differences in upbringing and experience alter adolescent brain structure and function? That’s a key question for future researchers to answer, she says.

Updated July 2019; Originally published October 2012