Adolescence is a dangerous time. Some of the most life-threatening risks that people take--drunk driving, binge drinking, smoking, having unprotected sex--are especially common during the teenage years. The following statistics illustrate the enormous toll in human suffering caused by adolescent risk taking:

  • Both males and females between the ages of 16 and 20 are at least twice as likely to be in car accidents than drivers between the ages of 20 and 50 are. Auto accidents are the leading cause of death among 15- to 20-year-olds, and 31 percent of young drivers killed in motor vehicle crashes in 2003 had been drinking.
  • Three million adolescents contract sexually transmitted diseases every year.
  • More than half of all new cases of HIV infection occur in people younger than 25, making AIDS the seventh leading cause of death among 13- to 24-year-olds. Two young people in the U.S. are infected with HIV every hour.
  • Forty percent of adult alcoholics report having their first drinking problems between the ages of 15 and 19.
  • Evidence of pathological or problem gambling is found in 10 to 14 percent of adolescents, and gambling typically begins by age 12.

In addition to the immediate consequences of risk taking--both for adolescents and for those who suffer from their actions--many behaviors that affect adult health begin and become entrenched during adolescence. So risky activities such as heavy drinking and drug use, which begin as voluntary experimentation, can be perpetuated by addiction. And whereas most teen drinkers, for example, do not progress to alcoholism, virtually all alcoholics started drinking in adolescence.

Preventing risky behavior while it is still a matter of deliberate choice is crucially important--not just for protecting troubled teens but also for society. An obvious answer is early intervention, which is both more successful and less costly than efforts to deal with established addictions later.

Strategies that help to postpone sexual activity, binge drinking and other risky behaviors also have the virtue of giving the forebrain and other neurological structures time to mature. As studies are now showing, the immature adolescent brain may be responsible for much of the risky business that young people engage in.

Over the past two decades, studies using magnetic resonance imaging (MRI) and other imaging techniques have shown that the human brain undergoes major remodeling during childhood and throughout the teen years--anatomical changes that may account for the risk taking, novelty seeking and impulsivity that characterize adolescent behavior. Gray matter in the brain, for example, begins thinning early in childhood--a sequential maturation process that begins at the back of the brain. Not until early adulthood does this wave of gray-matter thinning finally reach the forebrain areas where planning, reasoning and impulse control occur.

This growing evidence that risk taking may be hardwired into the adolescent brain has influenced the way that we and other psychologists now view troubled teenagers and the standard intervention programs aimed at preventing their risky behavior.

Why Programs Fail
Traditional intervention programs emphasize the importance of giving teens information about risks and allowing them the freedom to decide for themselves what to do. These programs encourage teens to trade off potentially deadly risks against often transient benefits and assume that they will see the light: just tell them the risks of HIV infection and unwanted pregnancy, these programs assume, and teens will not engage in unprotected sex.

Such programs are based on a collection of theories of decision making with names like "the behavioral decision framework" and "the theory of reasoned action." As their names imply, these theories expect that teenagers will weigh risks against benefits and come to the "rational" conclusion about their actions.

Some programs based on these theories have helped reduce risky actions taken by teens. For the most part, however, they have achieved only limited success. In addition to the modest percentage of teens influenced by these intervention efforts, the positive effects of these programs--many of which involve 10 to 20 hours of instruction--typically fade away in a matter of months.

In our view, intervention programs appealing to teen rationality are inherently flawed--and not because teens fail to weigh risks against benefits; as we will see, most teens do so conscientiously. Part of the problem may be that the "unfinished" architecture of their brains hinders adolescents from thinking like adults. Recent studies, for example, show that teens tend to weight benefits more heavily than risks when making decisions. So, after carefully considering the risks and benefits of a situation, the teenage brain all too often comes down on the side of the benefits--and chooses the risky action.

Just as important, traditional intervention programs are flawed because they are based on the notion that teens consider themselves invulnerable--despite evidence now pointing in exactly the opposite direction.

The Invulnerability Myth
For decades, a seductive explanation for risky teen behavior has reigned supreme among both the public and health professionals alike: teens drive too fast, binge drink and have unprotected sex because they feel they are invulnerable. They must therefore be underestimating their risks, or otherwise they would not take such chances. But studies uniformly dispute the widespread belief that adolescents consider themselves more invulnerable than adults (who, it turns out, are more likely to consider themselves invulnerable when compared with teens). And when it comes to risk, studies over the past five years show that teens actually tend to overestimate rather than underestimate the true risks of potential actions.

For example, a 2002 study by Susan Millstein and Bonnie Halpern-Felsher of the University of California, San Francisco, found that adolescents were more likely than adults to overestimate risks for every outcome that could be evaluated, including low-probability events (earthquakes and HIV transmission from unprotected sex, for instance) as well as higher-probability events (acquiring sexually transmitted diseases such as gonorrhea and chlamydia).

Another study, published in 2000 by Baruch Fischhoff of Carnegie Mellon University and his colleagues, reported on risk predictions assessed in a nationally representative sample of 3,544 adolescents from the 1997 National Longitudinal Study of Youth. Adolescents risk estimates for "die from any cause--crime, illness, accident and so on" in the next year or by age 20 were much higher than shown by statistical data. Recent data collected by one of us (Reyna) underline these differences between perceived and actual risks when it comes to sexually transmitted infections.

Interestingly, teens overestimation of risk appears to decline after early adolescence, and evidence suggests that experience may be responsible: engaging in risk taking without incurring immediate consequences may encourage complacency.

If adolescents often overestimate risks and do not think of themselves as being invulnerable, then why do they engage in risky behaviors? A number of studies indicate that when adolescents are mulling over risk taking, the perceived benefits of the action tend to outweigh and offset the perceived risks. For example, in a 2002 study of young (fifth to ninth grade) adolescents, Julie H. Goldberg of the University of Illinois at Chicago and her colleagues at the University of California, San Francisco, found that the perceived benefits of alcohol outweighed perceived risks in predicting the students drinking behavior six months later.

It now becomes clearer why traditional intervention programs fail to help many teenagers. Although the programs stress the importance of accurate risk perception, young people already feel vulnerable and overestimate their risks. And programs fail to alert teens about the allure of benefits, even though the teenage mind emphasizes the benefits of a potentially dangerous situation over its risks.

Some teens have certainly been "scared straight" by traditional intervention programs. But for the most part, such programs have not done much to deter risky behavior--and, even worse, they may actually be encouraging it.

Consider the adolescent who puts his odds of becoming infected with HIV through a single act of unprotected sex at 5050... and then learns through his intervention program that his true risk is one in 500 at most. The programs emphasis on inundating teens with risk information could well backfire, making them more rather than less likely to have unprotected sex or engage in other risky actions.

To improve the success of intervention efforts, we are testing a strategy fundamentally different from the one that traditional programs are based on: rather than asking teens to rationally balance risks and benefits, we are training them to think less logically and more intuitively--the way mature adults do, in other words.

Accentuate the Intuitive
This new strategy is based on a theory jointly proposed about 20 years ago by one of us (Reyna) and Charles Brainerd, now at Cornell University. Called fuzzy-trace theory, it originally was regarded as quite radical. Today, however, it can be described as an "establishment" theory of cognitive development because research has confirmed so many of its surprising predictions. It offers an explanation for the evolution of behaviors and memories from childhood, through adolescence and on to adulthood based on changes that occur in the way we reason. A decade ago fuzzy-trace theory predicted and discovered the counterintuitive finding that some false memories are more stable over time than true memories, among other novel findings.

Fuzzy trace is a so-called dual-processes theory positing that people rely on two quite different ways of reasoning to reach conclusions about situations confronting them. The first way is a deliberative, analytical approach that relies on details, such as those collected during rote exercises and fact memorization. This verbatim style of reasoning involves the kind of computational processing assumed by risk-intervention programs, when risks are traded off precisely against rewards. Far from being analytical, the second, or "fuzzy," style of reasoning occurs unconsciously and above all involves intuition, allowing people to penetrate quickly to the gist, or bottom line, of a situation. (The word "trace" in fuzzy-trace theory refers to the mental pictures, or traces, that collectively constitute memory.)

Fuzzy-trace theorys different modes of reasoning--verbatim and gist--are by no means mutually exclusive and can actually operate in the same person at the same time. But each predominates at different stages of life in normal human development.

Legendary developmental psychologist Jean Piaget contended that we start off as intuitive children who become analytical adults. Fuzzy-trace theory reverses things, proposing instead that the verbatim mode of reasoning reigns during childhood and adolescence. Then, with maturity, gist thinking takes over as we make decisions that disregard distracting details and instead are filtered through our experience, emotions, worldview, education and other factors.

The intuitive, gist-based approach to decision making tends to yield a "simple" answer--a black-and-white conclusion of good or bad, safe or hazardous, for example. Yet gist appears to be the more advanced form of reasoning, because the tendency to base decisions on gist increases with age, experience and expertise, as shown by research with children and adults.

Fuzzy-Trace Theory and Risk
When it comes to handling risks, fuzzy-trace theory predicts that mature decision makers will not deliberate about the degree of risk and the magnitude of benefits if a nontrivial chance of a catastrophic or health-compromising outcome exists. In contrast, the verbatim-based, analytical approach of adolescents faced with a risky situation would be expected to take longer. And indeed, studies comparing the reaction times in milliseconds for adults and adolescents to questions such as "Is it a good idea to set your hair on fire?" and "Is it a good idea to drink a bottle of Drano?" show that adults respond faster than teens.

In recent years, colleagues have suggested that fuzzy-trace theory could be applied to the vexing problem of adolescent risk taking. We have taken up the challenge, and our research suggests that adding a gist-based component to intervention programs serves a useful purpose. We believe that emphasizing intuitive rather than "logical" reasoning in potentially risky situations could help many--but not all--adolescents avoid engaging in risky behavior.

We propose that there are two kinds of teens who make similarly risky choices but do so through very different routes. We have dubbed these two groups the risky deliberators and the risky reactors.

The risky deliberators encompass the vast majority of teenagers--those who are in the normal developmental stage of adolescence. Before doing something potentially dangerous, risky deliberators rationally trade off risks against benefits, just as risk-intervention programs encourage them to do. And all too often, the risky deliberators come to a conclusion that, for them, is entirely logical: they conclude that the benefits of a risky action outweigh its risks--and intentionally go ahead and do it.

Consider the extreme example of Russian roulette, which was featured so prominently in the movie The Deer Hunter. Nick, played by Christopher Walken, has made a considerable amount of money gambling on Russian roulette. We last see him in a gambling den in Saigon sitting opposite his old friend Michael (Robert De Niro) and holding a gun to his head.

Nick clearly was mentally unstable, traumatized by his ordeal in the Vietnam War and addicted to heroin. But for risky deliberators and for the standard intervention programs aimed at helping them (and for economists of a certain stripe), the decision to play Russian roulette could be considered rational if the payoff in dollars were large enough. After all, the benefit could be a fortune that lasts a lifetime... and the risk of dying is only one in six.

The young risky deliberator has relied on verbatim reasoning that is age-appropriate and logical but that could result in a tragic outcome. Most adults, on the other hand, will look at this scenario--money to win and a gun with a single bullet in the chamber--and ask, "Are you crazy? No amount of money you could offer would get me to put that gun to my head. This is not about the number of dollars or the number of bullets--were talking about a significant risk of dying here." Adults, of course, are using gist-based thinking to cut quickly through the distractions, grasp the bottom-line meaning and arrive at a simple answer: absolutely not.

Impulsive Reactors
Risky reactors, on the other hand, are not thinking deeply or analytically. Instead they act impulsively because of some temptation in their environment. Risky reactors do not intend to do something dangerous. But for any number of reasons--including peer-group pressure or the excitement of the moment--they are pulled into risky situations, often against their better judgment.

Fortunately, most risky reactors grow out of their impulsiveness once they reach adulthood. But in the meantime, efforts to influence cognitive development by encouraging intuitive thinking probably will not help these teens, who are responders rather than thinkers. Instead measures for protecting unintentional risk takers should focus on adult supervision or monitoring to minimize opportunities for reacting to temptation.

Risky deliberators--the much larger group of at-risk adolescents--stand a far better chance of benefiting from exposure to intuitive, gist-based thinking. These teens do engage in reasoning--flawed though the outcomes may be--so we may be able to influence how they reason. To that end, we are now testing a gist-enhanced intervention program in a clinical trial involving more than 800 adolescents. Results should be available by the end of the year [see box on page 65 for comments of one at-risk teenager who seems to be benefiting from this gist-based intervention effort].

We are optimistic that gist-based thinking will one day be widely incorporated into risk-intervention programs, where it could help young people pass unscathed through their dangerous teenage years. For now, we offer the following empirically supported recommendations for guiding adolescents and helping them avoid taking unhealthy risks:

  • Offer risky deliberators well-reasoned arguments for resisting risky behaviors as well as factual information about social norms (The notion that everyone your age is having sex just isnt true). Focus on reducing the perceived benefits of risky behaviors--and on increasing the perceived benefits of safer, alternative behaviors.
  • Teens may not grasp the concept of harmful consequences because of their lack of relevant experience (which can also make them prone to repeated risk taking, if they have so far managed to dodge the bullets of negative consequences). Help them to understand the meaning of risk-related truths (the fact that HIV is not treatable with antibiotics means that AIDS cannot be cured) and to derive the gist, or bottom line, of messages that will endure in memory longer than verbatim facts.
  • Reduce risk by retaining or implementing higher drinking ages, eliminating or lowering the number of peers who can accompany young drivers, and reducing exposure to potentially addictive substances (rather than trying to teach minors to drink responsibly, for example).
  • Monitor and supervise younger adolescents rather than relying on them to make reasoned choices or to learn from the school of hard knocks; remove opportunities for them to engage in risky behavior.
  • Encourage teens to develop positive gists or images of healthy behaviors and negative images of unhealthy behaviors by exposing them to films, novels, serial dramas or other emotionally evocative media.
  • Identify and encourage teenagers to adopt so-called self-binding strategies (I will not attend unsupervised parties) and help them to practice recognizing cues that signal danger before it is too late to act (I will not ride with a drinking driver).

(The Authors)
VALERIE F. REYNA and FRANK FARLEY have studied risk for a quarter of a century. Reyna is co-director of the Center for Behavioral Economics and Decision Research and professor of human development and of psychology at Cornell University. Farley is L. H. Carnell Professor at Temple University and former president of the American Psychological Association.

(Further Reading)

  • How People Make Decisions That Involve Risk: A Dual-Processes Approach. Valerie F. Reyna in Current Directions in Psychological Science, Vol. 13, No. 2, pages 6066; 2004.
  • The Development of Judgment and Decision Making in Children and Adolescents. Edited by Janis E. Jacobs and Paul A. Klaczynski. Lawrence Erlbaum Associates, 2005.