U.S. flag

An official website of the United States government

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Institute of Medicine (US) and National Research Council (US) Committee on the Science of Adolescence. The Science of Adolescent Risk-Taking: Workshop Report. Washington (DC): National Academies Press (US); 2011.

Cover of The Science of Adolescent Risk-Taking: Workshop Report

The Science of Adolescent Risk-Taking: Workshop Report.

Show details

4The Psychology of Adolescence

The neurobiological processes that define adolescence and influence risk-taking are complex, and the role they play is emerging as a key factor in adolescent behavior. These processes must be understood in the context of psychological development and social influences. B. Bradford Brown provided an overview of psychosocial development and adolescent risk-taking, and Valerie Reyna explored recent research on reasoning and decision making as it applies to adolescent risk-taking.


Brown began with the primary psychosocial tasks adolescents must accomplish. Put simply, there are four key tasks:

  1. to stand out—to develop an identity and pursue autonomy,
  2. to fit in—to find comfortable affiliations and gain acceptance from peers,
  3. to measure up—to develop competence and find ways to achieve, and
  4. to take hold—to make commitments to particular goals, activities, and beliefs.

He identified two ways in which these basic tasks relate to the risks that adolescents take. First, many risk behaviors can either foster or impede the successful accomplishment of these tasks. Second, adolescents may turn to risky behaviors to help themselves cope with the failure to succeed in one of these areas.

Brown looked first at the relationship between risk-taking and the development of identity, which has been viewed by some psychologists as primarily an individual psychological process and by others as more of a social process. In the first view, originally associated with the work of Erik Erikson, the task is understood as a process of distancing oneself from the views of others, particularly parents, to form a clear sense of who one is as a person and how one wishes to behave in the world. When that process is successful, individuals are likely to avoid major risk-taking, but for individuals who have a more diffuse state of identity, there may be an association with drug use and other risks. Those who take the second perspective think that individuals draw their sense of self from the social world and that they have a primary interest in the way they are perceived and in how others respond to them. The result may also be a coordinated, secure sense of self, but for individuals who go through this process in a social context in which risk-taking is the norm, they are likely to be more prone to taking risks.1

Researchers have identified other components that also play a part in identity formation, such as identification by gender, ethnicity, and sexual orientation. A part of the task for adolescents is to discern the criteria for these possible identities, evaluate them, and decide whether and how to incorporate them into their personal sense of self. This process is a particular challenge for immigrant youth, who often must decipher both the culture of their family and ethnic or national group and the culture into which they have immigrated. Research on immigrant youth has suggested that, in this circumstance, many young people choose either to stick closely with their home culture, conforming to traditional customs and styles of dress and gaining the reputation of a good boy or girl, or to reject that option in favor of a more Americanized identity. The Americanized orientation often means association with risk-taking peers.

The development of autonomy is closely linked to identity formation and is also generally conceptualized primarily as either a psychological or an interpersonal process. Some researchers, Brown explained, have suggested that there is a universal process through which individuals develop healthy autonomy (Kagitcibasi, 2005). If individuals develop a high sense of agency (taking responsibility for their own actions) while retaining close connections with significant adults, they are likely to develop a healthy “autonomous, relational self,” which is likely to result in relatively low risk-taking. When this process goes awry, the result is often increased risk-taking. Other researchers have shown that young people who develop autonomy either too early or too late in their development often have poorer outcomes than those who develop it at the same time that their peers do (Dishion et al., 2000; Dornbusch et al., 1990; Feldman and Wood, 1994).

Finally, adolescents spend a lot more time with their peers than younger children do and are more heavily influenced by them than younger children are. The drive for affiliation and acceptance at this stage makes adolescents more open to peer influence and also tends to promote the rapid development of new relationships—with less time spent on negotiation of the basis for the friendship than at other stages of life. Researchers (e.g., Berndt, 1979; Brown et al., 1986) have identified a linear pattern that associates age and openness to peer influence, with a peak of openness to antisocial influences at about 9th grade. Openness to both neutral and prosocial influences is higher at every stage. Less is known about the reasons underlying the trajectory of openness to influence, although Brown noted that it seems likely to be related to the neurobiological developments discussed above.

Some evidence suggests that adolescents are most susceptible to peer influence in the early stages of new relationships or just prior to the development of a new relationship. Risk behaviors are also correlated with several more specific kinds of social situations, including romantic relationships that develop early in adolescence, association with older peers or permissive peer groups, romantic or sexual relationships with older partners, and lax adult supervision. What this suggests, for Brown, is that, apart from the cognitive and biological processes that affect adolescents’ behavior, it is important to understand the meaning that adolescents attach to risky behaviors in the social context in which they encounter them. If teenagers perceive, for example, that risky driving makes them more attractive or that engaging in unprotected sex makes them appear more faithful—those images may be important to their personal identity—within their peer group, they may decide to engage in those behaviors despite awareness of the risks.

This reality highlights the importance of developing social competence, another of the key tasks of adolescence. The capacity to engage effectively in social relationships is very important both for developing identity and for gaining acceptance from desired peer groups. Two skills are of particular importance for developing social competence: impulse control and the regulation of emotions. Individual adolescents who control their impulses very effectively are likely to have a very different pattern of social relationships than individuals who do not—and they are less likely to manifest aggressive behavior (Cairns et al., 1989). Consequently, many other individuals will tend to avoid impulsive adolescents, leaving them with a peer group made up of other aggressive individuals. Such peer groups tend to be less stable than others. The pattern is similar with emotion regulation—adolescents who are not successful at it struggle to form stable peer relationships. The last key task, developing commitments, is a protective factor; for example, religious and civic involvement is associated with low rates of deviance and therefore less involvement in many of the risky behaviors.

Brown highlighted the importance of viewing these developing competencies together. Deficits in social skills or social competence, for example, help to establish an adolescent’s reputation, his or her identity in a peer group as an aggressive and unpredictable person. This reputation directs the adolescent to peers who share some of these deficits, thereby intensifying the social pressures to which the adolescent is exposed. It is this coalescence of forces that have more influence on risk-taking than any of the factors alone. At the same time, however, these same factors can be a positive influence as well. A prosocial identity, healthy autonomy, and relationships with prosocial peers, facilitated by strong social skills are all likely to protect individuals from risk.

Robert Wm. Blum also stressed the importance of individual traits and skills that can help adolescents navigate adolescence and protect them from risk. These include values, goals, and positive orientation and affiliations—factors likely to be built in a positive family and community context—as well as the development of effective skills for coping with anxiety, stress, and adversity. It appears, he suggested, that innate personality traits, such as resilience, play an important role, but that the development of social competence is also learned.

Brown suggested that it is more likely that psychosocial tasks encourage risk than that they discourage it. The key tasks of adolescence challenge an individual to explore possible identities and fashion a comfortable social identity, to try to gain acceptance into groups, and to develop the skills to navigate romantic relationships. These tasks require new skills, and all require some level of risk-taking to reap what are likely to be very positive rewards. The individuals who do not engage in those sorts of risks get left behind. He cautioned against overlooking the value of some kinds of risk-taking and the extent to which adults actually encourage risk-taking.

Brown closed with a look at important questions about psychosocial influences that remain open. How much cultural variability is there in the “normative scripts,” or expected pathways, for the accomplishment of the key psychosocial tasks of adolescence? How do social contexts affect the accomplishment of these psychosocial tasks? What are the connections between these psychosocial tasks and the bio-cognitive-neural developments researchers have identified? Brown suggested that the most fruitful research approaches would coordinate findings related to individual behavior, social processes, and internal processes of development.


Each of the factors already discussed ultimately affects adolescents’ risk-taking by influencing the decisions they make. Researchers have also examined decision making itself. Reyna provided an overview of classic thinking about decision making, some new thinking, and research on the differences between the decision making of adults and adolescents.


In what Reyna described as the classic view of decision making, the mind operates much like a computer. That is, the most successful decision makers process more information more precisely. They “compute” a decision by estimating risks precisely, weighing potential rewards, and then acting based on the balance between them. The capacity to reason in this way improves as individuals mature. This general, classic view encompasses several different specific models, some of which incorporate the processing of such factors as social norms, self-efficacy, and perceived control over outcomes (for an overview of these models, see Reyna and Rivers, 2008). Reyna also mentioned a range of other theories, such as information processing, behavioral decision making, the theory of planned behavior, and prototype willingness. Noting evidence for the classic model, she suggested that although it can account for a significant portion of the variance in real-life risk-taking, it does not adequately account for the increased risk-taking of adolescents.

Surely, many of the decisions adolescents make are not reasoned or intentional, she said. As mentioned earlier, emotion, altered sensitivity to rewards, and increased impulsivity appear to play a role in adolescents’ decisions. The developmental differences in brain processing discussed earlier also need to be accounted for in a conceptual picture of adolescents’ decision making. Impulsivity has been shown to decrease steadily with maturity, when the ability to delay gratification increases. These capacities, however, do not fully account for adolescent risk-taking, either singly or together.

Moreover, some empirical evidence contradicts the classic view, Reyna observed. Studies of responses to what economists call a standard gamble—a situation in which subjects are asked to choose between a guaranteed $100 and the chance to win $200, at the risk of winning nothing—for example, have shown that adults tend to choose the certain $100 (Reyna and Ellis, 1994). What is surprising, she said, is that it is children who respond the most rationally, in a classic economic sense, to this choice by assessing the risk quantitatively. The older people are, the more likely it is that they will respond in a more qualitative way. They use what she called “gist-based” intuition to make their choice, which leads them to avoid the risk (Reyna, 2008; see below).

In addition, studies of the effect that perception of risk has on the likelihood of risk-taking have yielded some counterintuitive results. Most of the time, increased perception of risk decreases the likelihood that an individual will take the risk (a negative relationship), but sometimes the opposite effect is evident (Mills et al., 2008). There is a positive relationship between risk perception and risk-taking when adolescents are cued to remember their actual behavior: risk-takers perceive their risk as high, and nonrisk-takers perceive their risk as low.

Gist-Based Reasoning

The finding that adults tend to rely more on their intuitive reactions—or the gist of the situation—has been incorporated into what Reyna and colleagues have labeled the “fuzzy trace” theory (Reyna and Brainerd, 1995; Reyna, 2004). In this view, decision-making processes change from childhood to adolescence and from adolescence to adulthood. As in the classic theory, knowledge, acquired through both education and experience, is understood to play a part in decision making, but other factors are also explicitly identified. The way in which individuals perceive and understand a situation—how they represent it to themselves—and the way they retrieve relevant information from memory and apply it to their decisions also play an important part. Representation is central because decisions depend on how individuals subjectively perceive reality, not on reality itself. Furthermore, the kind of representation that is used to make a decision changes the nature of the decision process (e.g., from verbatim-based analysis of details to gist-based global thinking; see below). Thus, the individual makes a decision by integrating what he or she perceives with retrieved memories or knowledge, a variable and uncertain process.

Many public health interventions proceed from the premise that if adolescents knew of and understood a risk, they would not take it. Yet, Reyna explained, many studies have shown that not only are adolescents well aware of prevalent risks, but they also actually overestimate the risks of developing HIV or lung cancer or getting into a vehicle crash (see, e.g., Fischhoff et al., 2009; Millstein and Halpern-Felsher, 2002; Reyna and Farley, 2006). Figure 4-1 shows how 12th graders’ changing perception of the risk of smoking marijuana tracks compared with their actual use over the past few decades—demonstrating that the two are related as if decisions were being made rationally (higher risk perceptions, lower risk-taking). It is a myth that teenagers do not understand the risks of prevalent behaviors—or believe they are invulnerable—Reyna observed. Numerous studies have also shown that risk-taking could be predicted based on adolescents’ perceptions of the risks and benefits of the behavior, which suggests that impulsiveness does not easily account for it (Reyna and Farley, 2006). Adolescents act despite awareness of risks. The key is that, although adolescents overestimate many risks, they often also rate the potential benefits as very high—and thus the perceived benefits outweigh the perceived risks.

FIGURE 4-1. Twelfth graders perceived risk and actual use of marijuana over time.


Twelfth graders perceived risk and actual use of marijuana over time. SOURCE: Reyna and Farley, 2006. Reproduced with permission from Psychological Science in the Public Interest, Vol. 7, pp. 44. Copyright © 2006 by Sage Publications.

What complicates this kind of decision making, in Reyna’s view, is that individuals tend to have two, sometimes conflicting, versions of reality in their minds, and this is the essence of fuzzy trace theory. First, there is the gist-based representation of reality—the quick take on or summary of what is essential about something the individual has experienced or observed. That version of events may be in conflict with what she calls the verbatim, or more literal, representation of reality. The gist representation reflects the meaning of the event to the individual and hence also reflects culture, personality, and personal history. Studies of these types of thinking suggest that the more education and experience people acquire, the more likely they are to rely on their sense of the gist of the situation in making decisions (Reyna and Lloyd, 2006). Verbatim thinking is more specific and relates to thinking about and remembering specific facts or situations in detail. Because young people lack experience, they tend to base risky decisions on verbatim details rather than on the gist of the situation.

The knowledge that adults bring to a gist-based decision helps them put new information or situations in context. Because adults focus on gist rather than details, the effects of context can sometimes be paradoxical. For example, Fagerlin and colleagues (2005) showed that women who first estimated their risk of breast cancer (overestimating it as 46 percent), perceived the true value of 13 percent as lower than women who did not make an initial estimate. The perception of 13 percent varied depending on whether it was interpreted in the context of 46 percent; the verbatim representation of risk was identical for both groups (literally 13 percent), but the gist of the risk differed. Women who make initial risk estimates also show reduced interest in screening tests to detect breast cancer. Thus, highlighting the true level of risk (as is typically done in public health messages) can backfire if individuals overestimate that risk, as adolescents often do. In other words, it is the gist of the risk, or the meaning of the information in context, that is critical. This point was also demonstrated in a study of high school students’ ratings of the benefits and risks of sexual intercourse (Reyna, 2008; Reyna and Adam, 2003). The adolescents’ perceptions of risks and benefits varied with their psychosocial context, as reflected in effects of gender, age, and cultural background.

The effect of context can also be seen in the way that memories and impressions are triggered when a particular decision needs to be made. Reyna explained that when one changes the memory cue, or type of question that is asked, people retrieve different sorts of memories and thus answer similar questions in contradictory ways (Brainerd and Reyna, 2005; Mills et al., 2008). In general, when verbatim-type memories (e.g., of lonely Saturday nights, when the risk of pregnancy or sexually transmitted disease was very low) are triggered, risk perceptions reflect those memories. Alternatively, if gist-type memories (e.g., of general knowledge about potentially catastrophic consequences of unprotected sex) are activated, the same individual might have an entirely different perception of personal risk. Values and other principles that people endorse are also stored in long-term memory, but they influence choices only when they are retrieved and applied in the decision context. Research has shown that simple values and principles, what Reyna called gist principles, such as “avoid risk” or “better safe than sorry,” guide decision making, especially for adolescents who avoid unhealthy risks (Reyna, 2008). However, most adolescents engage in what Reyna called dual processing, that is, they engage in both verbatim thinking and gist thinking, making their decisions vulnerable to the sorts of contextual cues they receive when faced with a risky choice.

Survey studies in which adolescents are questioned about their thinking show that they engage in both gist-based and verbatim-based reasoning, but as they age they are less likely to ponder explicit trade-offs and more likely to apply gist-based reasoning, which tends to make them increasingly risk-averse for gains (Reyna and Farley, 2006). She likened the trade-off thinking to a Russian roulette scenario, in which the adolescent may reason that if the reward is high enough the risk would be worthwhile, whereas an adult would be more likely to intuitively recognize that the size of the reward is irrelevant when the risk is catastrophic.

Preschool-age children are the most quantitative when it comes to reasoning about a decision; studies show that they will take greater risks for greater rewards and will also scale back their risk tolerance if the reward is reduced. They also change their risk-taking when the level of risk changes. By adulthood, decision making is predominantly gist-based, which, Reyna suggested, is broadly consistent with the thinning of the gray matter and the pruning of synapses discussed in Chapter 3. That is, the selective pruning has reduced the number of connections in the brain, facilitating quicker, focused processing and allowing the individual to make many decisions with less deliberation and weighing of risks and benefits.


Examination of the psychosocial aspects of adolescent development and insights about adolescent reasoning further filled in the picture of how and why adolescents take risks. Much of the primary work of adolescence—including developing an identity, building competence, and gaining acceptance from peers—requires some degree of risk-taking. These tasks also help to explain why adolescents’ perspective on risky behavior may be very different from that of adults—a point that may provide useful guidance for those crafting messages and developing interventions designed to discourage youth from taking risks. At the same time, adolescents process decisions related to risk quite differently from the way adults do. That is, not only are they attuned to different goals than adults, but they also think differently, transitioning between verbatim-based analyses of risk-reward trade-offs to gist-based intuitions about the essential bottom line of risky decisions. Experience, context, and culture shape the gist representations and the retrieval of values that are central to healthy decision making. Although the thinning of gray matter and the pruning of synapses discussed in Chapter 3 might seem to reduce processing power, theoretical mechanisms emphasizing streamlined gist-based processing suggest that pruning might be important in developing the capacity to make sound decisions.



Brown cited the work of Jay McLeod on inner-city youth as an example of the social process of identity formation (McLeod, 1987).

Copyright © 2011, National Academy of Sciences.
Bookshelf ID: NBK53420


  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (1.2M)

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...