Perceptions of Risk and Social Judgments: Biases and Motivational Factors
Janis E. Jacobs
Adolescents are often characterized as poor decision makers by teachers, parents, and policy makers who point to underage drinking, teen pregnancy, or delinquency as evidence of poor judgment. Indeed, there has been great interest in understanding adolescent risk taking and decision making over the past two decades. More than 10 years ago, Elliott and Feldman (1990) suggested that our understanding of adolescent decision making was critical because of the large number of roles and choices available to teenagers in our culture, combined with “social conventions [that] have granted adolescents greater self-determination at steadily younger ages, even as the diversity of socially acceptable life options has expanded” (p. 4). The same authors noted that “adolescents are left with the task of, but not necessarily the requisite intellectual and emotional tools for, reconciling [such decision making tasks]” (p. 5).
Despite widespread interest in how children and youth make decisions and numerous programs to improve their decision-making skills, few of the educational programs have been based on our current knowledge about the development of biases in reasoning and motivational processes that may contribute to judgments related to risk taking. In this chapter, after briefly reviewing previous research on adolescent cognition related to decision making, I discuss a number of reasoning, motivational, and developmental issues that are critical to our understanding of adolescent risk taking.
EARLIER RESEARCH ON ADOLESCENT REASONING RELATED TO DECISION MAKING
Much of the research related to adolescent risk taking has relied on the assumption that decision making is based on reasoning and information-processing abilities, and it typically has been rooted in traditional developmental models. The result of the exclusive emphasis on reasoning competence as the basis for decision making is reflected in research from the 1980s and early 1990s that compared adolescents’ and adults’ decision making on unfamiliar, hypothetical scenarios. These studies indicated that by early adolescence (typically about age 12), teens are able to use basic adult concepts and reasoning (Dunkle, 1992; Moshman and Franks, 1986; Overton, 1990). Thus, many studies suggested that by early to middle adolescence, reasoning abilities were similar to those of adults (for reviews, see Byrnes and Beilin, 1991; Kuhn, 1989; Overton, 1990). Moshman (1993) concluded that “there is no evidence of any important component of rationality that is lacking in adolescents and found in most adults” (p. 37). In addition to general reasoning abilities, a few studies investigating specific decision making skills were conducted during the same period. A number of these supported the findings of the reasoning literature, indicating few differences between the decision making of adolescents and adults. For example, studies looking at information use in everyday decision making found that by age 12, adolescents systematically searched for relevant information (Klayman, 1985), and that early adolescents used almost the same sources of information as did their parents when making everyday decisions about certain topics, such as which bike to buy or which camp to attend (Jacobs, Bennett, and Flanagan, 1993).
Other research, however, did not support the adolescent-as-adult perspective. These studies reported age-related differences between early and later adolescence in decisions that involved making inferences, perceiving risk (Lewis, 1981; Shtarkshail, 1987), considering consequences (Beyth-Marom, Austin, Fischhoff, Palmgren, and Jacobs-Quadrel, 1993; Gouze, Strauss, and Keating, 1986; Lewis, 1981), planning (Rowe, 1984; Urberg and Rosen; 1987), and rates of perceived invulnerability (Quadrel, Fischhoff, and Davis, 1993). A big difference between this second set of studies and the first was that they focused on real-world issues, social situations, and judgments based on prior experiences, whereas the other studies were concerned with cognitive performance under ideal conditions. Thus, the earlier studies indicate that decision making is not a single cognitive competence and that there are many complexities inherent in any particular decision related to risk taking behavior. Fischhoff and Quadrel (1991) made a similar point, suggesting that formal decision analysis be used to understand the many variants that affect each adolescent decision regarding alco-
hol because it allows the estimation of the many options, consequences, and uncertainties that an adolescent might face when making such a decision.
It is clear from the previous research that decision making about underage drinking (or any other risky behavior) is a product of interactions among a number of different competencies, situations, and emotional states. Thus, studies that focus exclusively on discerning age differences in cognitive competence may tell us only what adolescents can do under ideal conditions, rather than what they will do under more realistic conditions when personal goals, beliefs, prior experience, values, and emotions are added to the decision-making equation (Jacobs and Klaczynski, 2002). Although important, this narrow definition of competence may miss the part of decision making that is most closely related to the risk taking behaviors that are of interest to policy makers and parents and that may distinguish adolescents from adults. Several components of decision making have not typically been captured when the focus has been on cognitive competence, including biased social judgments, motivation, and self-perceptions. The remainder of this chapter reviews our current knowledge about those three topics, then concludes by making some observations about the potential implications of this research for our understanding of adolescent risk taking behaviors.
BIASES IN SOCIAL JUDGMENTS
The overriding theme in the adult judgment and decision-making literature has been that adults commonly fall prey to judgment biases, ignore important information, rely on seemingly inappropriate decision making shortcuts, and make nonoptimal decisions across a wide array of situations when they are making social judgments (for reviews see Dawes, 1988; Nisbett, 1993; Plous, 1993). Thus, research on adult judgment and decision making has focused on when and how decision making deviates from normative models. If adults, who have more experience and knowledge about the world, can fall prey to judgment biases under certain conditions, it is not surprising that adolescents might be at even greater risk of succumbing to such biases when making social judgments.
Nonoptimal Decisions Due to the Task
The task of making sound decisions about social situations, including those that involve opportunities for drinking alcohol, may be inherently more difficult than making decisions about nonsocial topics. We know from previous research with adult populations that the use of nonnormative decision strategies is directly related to the social context and content surrounding the decision situation. Adults are more likely to use nonnormative
decision strategies in particular situations, including unfamiliar tasks, choices with uncertain outcomes, group presentation of information, and ambiguous situations (Fischhoff, Slovic, and Lichtenstein, 1979; Bernbaum and Mellars, 1983; Manis, Dovalina, Avis, and Cardoze, 1980; Christensen-Szalanski and Beach, 1982).
Adolescents may be routinely making decisions under exactly the same conditions in which most adults turn to faulty decision rules. They must consider new options that have never been available to them before, and their valuing of those options begins to depend on their weighing of family versus peer versus their own values. They must attempt to consider the potential consequences for their decisions from an often inconsistent pattern of restrictions and privileges (e.g., childlike restrictions about bedtime that don’t match the adult privilege of driving the family car). Furthermore, while they must determine the likelihood of various consequences, their estimates must be based on limited knowledge of a more diverse group of people than they have encountered before.
We have found evidence for greater biases in social judgments when the topics are social rather than nonsocial (Jacobs and Potenza, 1991); when the outcomes of a risk taking decision are probabilistic (Finken, Jacobs, and Laguna, 1998); when adolescents are reasoning about populations with greater variability (Jacobs and Narloch, 2001); and when they are reasoning about unfamiliar others (Jacobs, Greenwald, and Osgood, 1995). These findings suggest that, when faced with decision tasks that have these characteristics, adolescents (and adults) may be more likely to exhibit less-thanoptimal reasoning. Underage drinking and other forms of risk taking are likely to occur in social situations and when adolescents find themselves with large groups of unfamiliar peers. In these situations, they are left to estimate how others typically behave and what they think. This may lead to overestimations of others’ drinking and acceptance of such behavior.
Nonoptimal Decisions Due to Judgment Biases
In social psychology, the terms “intuitive statisticians,” “intuitive psychometricians,” and “intuitive social scientists” (Nisbett and Ross, 1980) have been used to label adults’ use of statistical rules to make everyday social judgments. A large body of literature on this topic in social psychology suggests that adults are reasonably good psychometricians in many everyday social judgments, but that they sometimes over- or under-apply particular rules or use shortcut “heuristics” instead of relying on the normative rules. Although some researchers have suggested that the decision biases that adults exhibit are overgeneralizations of rules that are usually beneficial (e.g., Arkes and Ayton, 1999; Baron, 1990; Gigerenzer, Hoffrage, and Kleinboelting, 1991), much of the research on adult judgment and
decision making has focused on deviations from normative models, demonstrating judgment biases and reliance on heuristic shortcuts (for reviews see Dawes, 1988; Nisbett, 1993; Plous, 1993). This research is important for our understanding of adolescents’ risk taking judgments related to alcohol use because biased judgment rules may influence their decision making as they encounter opportunities for underage drinking.
To examine the developmental antecedents of judgment shortcuts and biases reported for adults, procedures from adult decision making research have been adapted for research with children and adolescents (e.g., Baron, Granato, Spranca, and Teubal, 1993; Davidson, 1995; Jacobs and Potenza, 1991; Jacobs et al., 1995; Reyna and Ellis, 1994). These studies have revealed three trends. First, we know that even very young children use many of the same “rules of thumb” or heuristics that adults use in their decisions and appear susceptible to the same judgment biases observed in adults. Second, reliance on heuristic and judgment biases, at least under some task conditions, increases with age. Third, research illustrating age-related increases in heuristics and biases has typically found that these developments are accompanied by developments in the reasoning competencies that enable children and adults to make normatively prescribed judgments and decisions. Two areas of research on judgment biases that are directly relevant for our understanding of underage drinking are reviewed.
Judgments Based on Social Knowledge
The topic of overreliance on information that is representative of a social category (e.g., age, gender, or what “popular” kids do) versus the real base rates of behaviors or attitudes is one area that has been studied by several developmental researchers. In one of the first studies on the development of judgment heuristics (Jacobs and Potenza, 1991), we found that, although even first graders ignored base rate information to make social judgments, this bias increased across the elementary school years and the use of social categories as an explanation for social judgments increased between sixth grade and college. Paradoxically, over this same age range, on parallel problems that did not involve social content, the use of the actual base rate information increased.
Despite increased competence to use base rate information with age, several studies (Agnoli, 1991; Barrouillet, Markovits, and Quinn, 2002; Davidson, 1995; Jacobs and Potenza, 1991) have shown that in situations in which social concepts are activated (e.g., stereotypes), competence is less frequently displayed with age and, in turn, reliance on biased heuristics is more prominent as children make the transition into adolescence. Similar unexpected developmental trajectories have been found for other biases when judgments involve personal beliefs or social content (Klaczynski,
2001; Klaczynski and Narasimham, 1998; Markovits and Dumas, 1999). In these cases, the majority of older adolescents and young adults exhibited biased reasoning, based their judgments on outcomes rather than a priori probabilities, and relied on small samples of vivid evidence rather than large samples of relatively pallid evidence.
These findings have serious implications for many real-world situations when the base rates are not clear (e.g., social gatherings, parties). In such situations, adolescents are likely to make judgments based on their beliefs about how others behave. For example, at a party when 20 teens are drinking and 15 are not, the individual is likely to make a decision about whether or not to drink based on his or her prior social stereotype that “people who attend such parties all drink,” rather than on the actual numbers. A number of studies have shown that adolescents believe that underage alcohol use is normative (Aas and Klepp, 1992; Barnes, Farrell, and Banerjee, 1995; Beck and Treiman, 1996; Olds and Thombs, 2001) and that their estimates of the norms for drinking and for drinking and driving are high (Basch, DeCicco, and Malfetti, 1989; Jaccard and Turrisi, 1987). One teen summed it up by saying, “Everybody’s going to drink and get behind the wheel some time” (Basch et al., 1989:392).
Estimation and Perceptions of Risk
A different kind of problem that may arise when individuals are making social judgments occurs when errors are made in the course of estimating risk. This happens because the actual base rates of most behaviors and risks are not available when we make everyday decisions. In order to judge whether three or six drinks of alcohol are “reasonable,” the adolescent needs to have some kind of “data” about how often others engage in behaviors and how often particular consequences of behaviors occur. Such base rate information in the real world seldom is presented to us as we make judgments, however, leaving us to “collect the data” for ourselves. Previous research indicates that even adults have a tendency to overestimate small and underestimate large base rates (Lichtenstein, Slovic, Fischhoff, Layman, and Combs, 1978) and to assume that others’ attitudes and behaviors would resemble their own (Nisbett and Kunda, 1985). This general dilemma is compounded by the fact that adolescents must make judgments based on a limited amount of previous experience (Jacobs et al., 1995). Thus, if adolescents hear stories of others drinking a lot or happen to attend parties at which alcohol is served, they are likely to conclude from their limited sample that the norms for underage drinking are much higher than they really are.
We have conducted a series of studies to determine the developmental patterns of estimation about day-to-day behaviors, as well as estimates of
some low-probability, but high-risk, behaviors. In one study, we measured the accuracy of children’s base rate judgments with “accuracy” defined as the correspondence between their estimates of classmate’s behaviors and the criterion of children’s self-reports of the same behaviors. We found that children at all grade levels were generally overestimating the base rates of behaviors, while slightly underestimating attitudinal base rates; however, even first-graders’ estimates only deviated from the criterion by an average of 1.11 points on an 11-point scale. The most important finding was that accuracy increased for both behavioral and attitudinal estimates throughout the elementary school years (Jacobs et al., 1995). We know, however, that children and adolescents are not well calibrated to the size of the sample when they are collecting social data (Jacobs and Narloch, 2001). They are as likely to make similar generalizations from small and from large samples, although they are sensitive to the expected variability of the particular behaviors (e.g., the number of alcoholic drinks consumed is expected to be more variable than the number of burgers eaten).
Others have examined the accuracy of adolescents’ perceptions of risk regarding alcohol and they have found that adolescents rate the risks of drinking alcohol as greater for others than for themselves (Hansen, Raynor, and Wolkenstein, 1991). Interestingly, normative beliefs and beliefs about consequences are among a small set of factors that have the strongest relationships to later alcohol and substance abuse for adolescents (Hansen and Rose, 1997). In addition, McNeal and Hansen (1999) have shown that the deterioration of normative beliefs mediates the transition from nonuse to use of alcohol during middle adolescence. Finally, adolescents who are involved in heavy drinking typically perceive greater benefits relative to risks and fewer long-term consequences, as well as a higher rate of participation by others (Benthin, Slovic, and Severson, 1993; Hansen et al., 1991). Similarly, in a recent study with adolescents between the ages of 13 and 15 (Johnston and Jacobs, 2003), we examined the accuracy of estimates of classmates’ involvement in problem behaviors (ranging from drinking alcohol to shoplifting). We found that although all adolescents in the study overestimated the rates of peer behaviors, great variation was seen in the size of the estimation bias, enabling us to distinguish three groups. Participation in deviant activities was related to size of estimation bias. Extreme overestimators were engaged in significantly more mild and severe deviant behaviors than either the moderate or on-target group, and they reported poorer self-esteem and lower grade point averages (GPAs).
This pattern of results supports previous studies (e.g., Nisbett and Kunda, 1985) indicating that adults and adolescents have a tendency to overestimate the population base rates for activities in which they, themselves, are engaged (risky or not). Indeed, adolescents, as well as adults, who participate in high-risk activities generally believe that the rate of
participation by others is higher than do nonparticipants (Benthin et al., 1993); thus, beliefs about normative practices may be related to older adolescents’ decisions to engage in risky behaviors (Basch et al., 1989). In addition, people are only moderately successful at assessing how much they know, resulting in overconfidence about their judgments (Quadrel, Fischhoff, Fischhoff, and Halpern, 1992). We know that even college students tend to have a number of misperceptions about the effects of use (e.g., Jaccard and Turrisi, 1987). Feelings of overconfidence about one’s knowledge combined with misperceptions about the facts may lead to greater risk taking among adolescents. This may be particularly pronounced among some teens; Quadrel et al. (1992) found that at-risk adolescents exhibited greater overconfidence about their knowledge than other adolescents or adults.
One of the most intriguing findings from the studies just reviewed is that there are individual differences in judgment biases—we have a group of teens out there who perceive fewer long-term consequences of alcohol use, who are “extreme overestimators” of others’ behaviors, and who exhibit greater overconfidence and perceptions of immunity. The same group of adolescents is also reporting more alcohol use, engaging in more deviant behaviors in general, doing more poorly in school, and exhibiting lower self-esteem. This is the group that we would expect to make the worst decisions. It is difficult to determine whether these judgment biases are precursors to decisions to begin using alcohol (e.g., everyone else is doing it, so I want to try) or if they are the result of engaging in particular behaviors, so that it is used as a post-hoc justification for behavior (e.g., if I’m doing it, everyone else must be doing it more).
SELF-MOTIVATION AND RISK TAKING
What do adolescents do when faced with decisions that matter to them personally? Constructs that are intimately tied to the sense of self, such as personal goals, attitudes, values, emotional states, and self-beliefs, quite likely affect adolescents’ choices. Recent research has explored motivational influences on children’s and adolescents’ reasoning and judgments. A few of these topics will be discussed.
Cochran (1991) describes some decisions as self-invested, as compared to self-divested, choices. Self-divested decisions are those that are devoid of self-considerations, and based exclusively on objective data. In adolescence, it is probable that most (if not all) decisions have a self-invested quality, even if on the surface they appear to be self-divested (selecting clothes, for
example). This is supported by the findings from the adolescent risk taking literature that suggest that judgments about hypothetical scenarios do not predict actual behaviors that are self-invested for risk-taking behaviors, such as marijuana smoking (Bauman, 1980), cigarette smoking (Bauman, Fisher, Bryan, and Chenoweth, 1984), sexual activity (Gilbert, Bauman, and Udry, 1986), or contraceptive use (Paikoff, 1990). More support is found in studies indicating that, during middle and late adolescence, teens believe that choices regarding alcohol and drug use are personal decisions rather than ones related to social convention or morality (Nucci, Guerra, and Lee, 1991). These authors also report that high-drug-use teens view themselves as the sole authority on such decisions rather than their parents or the law.
As with many motivational factors, beliefs about oneself may play a dual role in decision making as both outcome and predictor. Self-beliefs will play a role in individuals’ choices by guiding the selection and use of information used to evaluate oneself, and at the same time, decision outcomes contribute to the beliefs about the self. Self-evaluation is one area that has been well studied. The adult social psychology literature has frequently contrasted the concept of “self-serving biases” that drive self-evaluation (e.g., Greenwald, 1980; Kruglanski and Ajzen, 1983) with the view that people are really trying to evaluate themselves accurately (e.g., Darley and Goethals, 1980), but fail to get sufficient information or use faulty judgment rules (e.g., Ajzen and Fishbein, 1975; Nisbett and Ross, 1980). Trope (1986) describes the first view as “self-enhancement” because it suggests that people select and use the information that will enhance their self-evaluation; he describes the second view as “self-assessment” because it assumes that people are choosing information and strategies that they believe will yield the most accurate self-evaluation.
Self-assessment goals may be particularly strong during adolescence because identity formation is critically important during this period (Erikson, 1968; Marcia, 1980). If adolescents are “trying on” various identities or attempting to construct coherent self-beliefs, self-evaluation may play a large role in decision making, and the particular role may vary depending on the adolescent’s point in development. For example, while constructing identity, the adolescent’s motivation may be for new experiences that will allow him or her to evaluate what fits and what does not. At this point, adolescents may knowingly make choices that they know they may later regret “just for kicks” or “just to see what it is like.”
Self-enhancement or even self-preservation goals may also play a large role during this period. Due to increased concerns about fitting in and conforming to the peer group, adolescents may want to select information that will enhance their self-perceptions. Of course, the information that adolescents find to be self-enhancing may differ dramatically from that
used by adults. For example, Harter, Waters, and Whitesell (1998) and others have reported increased focus on physical appearance during adolescence that declines with age. Therefore, adolescents may be motivated to make decisions that will enhance their self-evaluations of their appearance, whereas adults may be more motivated to select information that will enhance their evaluations of their parenting skills or stature in careers. For some adolescents, self-enhancing information may involve risk taking so that they build their identities around selected memories of risky behaviors. In addition to the immediate goal of self-enhancement, each small choice will influence the adolescent’s self-beliefs. For example, a negative self-evaluation of ability is met with the decision options of cheating, setting lower standards, lowering the perceived value of success, or giving up (so that ability can be discounted if no effort was expended). Once one of these self-protecting decisions is made, self-evaluation is more positive and other decisions will be motivated by the same need to maintain positive beliefs about the self.
Some support for adolescents’ self-enhancement or self-protection bias can be found in a series of studies done by Klaczynski and his colleagues to examine the use of biased judgments that are personally relevant. By presenting “evidence” that is either consistent with or threatening to their goals and beliefs, they found that adolescents exhibit more biased reasoning and less statistical reasoning on scenario problems that threaten their personal goals (Klaczynski and Fauth, 1997; Klaczynksi and Gordon, 1996) and those that are discrepant with their personal beliefs (Klaczynski, 1997; 2000; Klaczynski and Aneja, 2002). These investigations have illustrated that the motivation to protect favored beliefs directly affects adolescents’ use of reasoning. Additional evidence of self-protection related to alcohol use by adolescents may be found in the studies reviewed earlier indicating that at-risk adolescents perceive the consequences of alcohol use and other risk taking as lower than other teens. In addition, adolescents who are heavy drinkers are more likely to believe that they are “immune” from the negative consequences of alcohol than those who drink moderately or not at all (Hansen et al., 1991).
One’s history of previous choices (based on values and goals) may constrain one’s options over time. This may occur because the value of some options decreases if they are infrequently selected, until their value is so low that they are no longer considered. The result is a more constrained set of options due to the cumulative restrictions of previous social and economic choices. For most people, options become more limited in scope because of the funneling effect that occurs as life choices are made. For
example, a 16-year-old and a 26-year-old are each faced with deciding whether to stay at a party and continue to drink heavily. Each has to go to work the next morning. The most negative consequence for either of them would be missing work due to a hangover, and ultimately being fired. This outcome is valued differently by the two people because the 16-year-old knows that even if he loses this job, that other similar part-time jobs are available, and his parents will continue to support him (although he may have less spending money). The 26-year-old is more concerned about the negative outcome of losing the job because of previous choices concerning college major, geographic location, and financial obligations for a car and home that have preceded this job. The older man is likely to have lower values for the same positive outcomes (e.g., impressing friends) due to the constraints imposed by his previous choices.
Gardner and Herman (1991) have discussed a similar concept with regard to adolescents’ AIDS risk taking behavior. They suggest that one reason adolescents are more likely than adults to behave as if opportunities for gains are more valuable than protection from losses is that adolescents have less to lose. They may discount the value of protection from risk because they believe that what they might gain is worth more than what they have to lose. Gardner and Herman (1991) suggest that this is a rational response to uncertainty, and adolescents have greater uncertainty concerning their ultimate social and economic niches in life. Whether it is rational or not, social and experiential constraints clearly limit decision making, and these may occur with greater frequency as adolescents get older and the expectations for them change.
In addition, options may be constrained by objective changes in the child’s life (Fischhoff, 1992). Clearly, decision opportunities, as well as the circumstances under which they are made, vary considerably from toddlerhood to adulthood. Toddlers have little autonomy and make choices primarily by refusing to do certain things; as children mature they are allowed to make more choices and are more often stuck with the consequences. Our theories tend to focus on change within the individual rather than change in the environment and in how it responds to the individual; thus, the ways in which changes in decision-making skills are responses to changes in the environment rather than cognitive changes is not clear.
The Role of Context
It is important to remember that decisions are not content or context free. They are always focused on a topic and made within a particular context. The adult literature clearly indicates that information selected for decision making is specifically guided by goals, and that the structuring of information depends on its intended use (e.g., Kahneman and Tversky,
1984; Zukier and Pepitone, 1984; Zukier, 1986). Most recent developmental research acknowledges the importance of context; thus, we should expect task dependence and unevenness in the development of judgment and decision-making skills. There is ample evidence of domain specificity during development from areas such as problem solving (Bransford, Sherwood, Vye, and Rieser, 1986; Sternberg and Martin, 1988) and decision consultation during adolescence (Finken and Jacobs, 1996). As Kuhn (1989) suggests, it appears that both domain-specific and domain-general processes need to be considered because there is likely to be some degree of generality within any particular area.
Another important part of domain specificity is the social context. Adolescents’ decisions and decision processes may vary, depending on whether they are made in the context of school, extracurricular activities, home, or the mall; they may have different goals when making decisions with the family and with friends; their motivations may even vary depending on which group of friends they are with at a given time. One study found that older adolescent females consult different individuals, depending on the type of decision they are making (Finken and Jacobs, 1996). In addition to the influence of social context on the process and outcomes of decisions, social interactions are the arenas in which adolescents are most likely to learn to make decisions. Although no one really believes that decisions are made in a vacuum, the important roles of family and friends as decision making models and teachers have often been ignored. Families provide an arena in which fledgling decision makers try their new skills, and where more experienced decision makers model appropriate behavior or even provide instruction on how to make decisions (Jacobs and Ganzel, 1993).
The Role of Consequences
The feedback an adolescent receives in the form of consequences may play a large role in future decisions. Developmental models have been particularly poor at including the role of feedback or practice; however, practice may allow adolescents to gradually replace decision strategies that don’t work with new ones (Kuhn, 1989). Experience is especially important for actions such as identifying alternatives or estimating consequences (Fischhoff, 1992), but may mislead the developing decision maker if outcome alone is used as the hallmark of a good decision or one that should be repeated. This is because the outcomes of many decisions have an element of probability (e.g., not getting caught when speeding), so that poor decisions can sometimes result in good outcomes.
The problem is compounded when misattributions occur for the probabilistic outcomes because an adolescent could experience a negative out-
come even if good decision making skills are used, while the same adolescent could experience a positive outcome even if using poor decision making skills. This is important because if one learns to attribute outcomes to ability as a decision maker when such an attribution is unwarranted, future decisions could be affected. Because outcomes are probabilistic, adolescents may interpret the fact that they previously drank too much and drove home without an accident as evidence that they can drink and drive safely (Jacobs and Ganzel, 1993). For example, if an adolescent decides to drive after having six drinks of alcohol, and gets to her destination without mishap, she may attribute the outcome to good judgment ability (e.g., “I know when I’ve had too much to drink”) rather than to good luck. Because the actual probability of an accident is quite low, this event may happen several times and the self-evaluation will be reinforced, leading to more risk taking in the future. Basch et al. (1989) report comments from focus groups with adolescents that support this point; one participant in their study said, “I do not think I’ve ever had so much to drink that I (couldn’t) drive. I’m always driving and I’ve never had an accident” (p. 393). Another said, “Even if I’m way over the alcohol limit, I just pay extra close attention … do a lot of driving … driving is second nature to me … and even drunk driving, I do that a lot. One should just know how to do it” (p. 393).
The effect of such misattributions may be particularly profound during adolescence, when individuals are just beginning to make decisions with consequences, and have little experience to guide their attributions. We have found some support for this idea in two studies related to older adolescents’ risk taking behaviors. In one study of adolescent women who were sexually active, those who had taken risks (not used birth control), but experienced no consequences (pregnancy), gave the lowest estimates for the probability of getting pregnant in the future (Jacobs and Potenza, 1992). In a second study, we found that older adolescent drinkers who had not encountered negative consequences believed their risks of being in an accident when drinking and driving were lower than those who did not drink or had less experience drinking and driving (Finken et al., 1998). These findings highlight the importance of perceptions of consequences (e.g., McNeal and Hansen, 1999) and make it clear that experience with underage drinking without negative consequences may lead to misattributions and a heightened sense of invulnerability.
A related issue is the serial nature of everyday decision making. Most current models that have been used to study adolescents or adults assume that one major decision is made, on the basis of lots of information, and without feedback. In reality, few if any decisions are made this way. An adolescent decides to “say no” to alcohol and does so at the first opportunity. However, after getting feedback from peers, indicating that the choice was not popular, the adolescent may change that decision or put off imple-
menting it. In addition, a decision to drink heavily at a party may be followed by decisions to abstain or to drink responsibly. When pregnant adolescents are asked about their decision not to use contraception, they are not typically asked about the many times they did use it (and how or when their decision changed). Instead of a reasoned, one-time decision, arrived at after considering the value of various options and their related consequences, we may need serial models of decision making (see Beyth-Marom et al., 1993, for evidence related to one-time versus regular decisions). According to this conceptualization, a “decision” to become involved in underage drinking would be reached by making a series of choices over time that appear to be made quite independently rather than by considering options and outcomes at the same time. Each particular choice then contributes to the next choice.
CONCLUSIONS AND IMPLICATIONS
The work from the 1980s suggested that, although some cognitive skills that are relevant for judgments about risk are in place by early adolescence (e.g., deductive and logical reasoning, information search, conceptual understanding of risk/benefit information), other skills, such as drawing relevant inferences, perceiving risk, considering potential consequences, generating more alternatives, and developing contingency plans, appear to develop during the adolescent period. This chapter focused on more recent work that suggests some of the ways in which adolescent decision making about risk taking may be impacted by motivational and reasoning biases in an attempt to point out the many ways in which decision making in adolescence may be a product of more than just cognitive capabilities.
The chapter presents an image of the development of judgment and decision making skills that runs contrary to that painted by traditional developmental theories because the evidence suggests that judgment heuristics and other biases appear to be linked to increases in knowledge and to the construction of personal theories about the social world. As different social concepts are acquired, and as these become increasingly consolidated, adolescents are likely to base their judgments on social category information, increasing rather than reducing the potential for biased judgments. In addition, the number of situations in which shortcut reasoning strategies are likely to be applied increases with age. Interestingly, the ability to reason logically increases during the same time period, so that by middle adolescence, individuals possess the ability to think rationally and logically, but they may not invoke those skills in everyday judgment situations. At these times, they may rely instead on what has been termed “experiential processing” that may result in biased decision making.
Adding to the potential for biased social judgment processes are the many motivational factors that play a role during adolescence, including the development of self-concept and identity, values and goals, contextual factors, and overreliance on the consequences of events as a guide to future decision making. An important message from the research is that motivations have a substantial impact on whether heuristics or more analytic reasoning abilities are used to make judgments and decisions.
What are the implications for understanding underage drinking and making an impact on adolescents? Several generalizations can be made from the research reviewed:
In a variety of situations, adolescents, like adults, fall prey to judgment biases and decision heuristics that are likely to impact their perceptions of risk. Adolescents use judgment shortcuts and heuristics instead of the available base rates, particularly in social situations in which it is cognitively “easier” to use prior beliefs, social stereotypes, and highly visible information. These heuristics are based on their understanding of the social world, which may in itself be biased.
Adolescents overestimate the base rates of peers’ risk taking behaviors, especially when they are engaged in similar risk taking. Whether their perceptions are motivated by self-protection or faulty estimation strategies, these inaccurate estimates are likely to form the basis for their views of their own behavior as normative.
Self-preservation plays a large role in adolescents’ selection of information to use for social comparison and for estimating risk. This is especially likely to be important during the period of identity formation.
Decisions are content and context specific—choosing to drink alcohol at a party of close friends may be very different than choosing to drink at a large tailgate event with hundreds of people in attendance.
There is evidence that people can be trained to overcome biased judgment strategies by getting them to focus on the relevant information; some of this work has been done with adolescents.
As the developmental and adult studies reviewed here illustrate, logical deductive reasoning strategies are not frequently used when social judgments are being made about events with uncertain outcomes. Therefore, teaching rational decision making skills alone is not likely to make an impact on everyday, seat-of-the-pants choices about underage drinking. Factors related to judgment biases, self-perceptions, and motivation need to be included in our thinking about adolescent alcohol use. In addition, differences in the effectiveness of various approaches may depend on the developmental level, context, and prior experiences of the adolescents them-
selves; a one-size-fits-all approach is unlikely to work. Finally, adolescents’ “readiness” to drink, despite the fact that they are underage in our country, is related to a confluence of cognitive, social, emotional, and community factors. The positive trends in development that lead to greater autonomy and a focus on the self and peers are the same trends that lead to great opportunities for risk taking. The key to program development related to underage drinking is to understand and acknowledge adolescents’ judgment and decision making abilities, while being aware of the limitations and biases inherent in the same processes.
Aas, H., and Klepp, K.I. (1992). Adolescents’ alcohol use related to perceived norms. Scandinavian Journal of Psychology, 33(4), 315-325.
Agnoli, F. (1991). Development of judgmental heuristics and logical reasoning: Training counteracts the representativeness heuristic. Cognitive Development, 6, 195-217.
Ajzen, I., and Fishbein, M. (1975). A Bayesian analysis of attribution processes. Psychological Bulletin, 82, 261-272.
Arkes, H.R., and Ayton, P. (1999). The sunk cost and Concorde effects: Are humans less rational than lower animals? Psychological Bulletin, 125(5), 591-600.
Barnes, G.M., Farrell, M.P., and Banerjee, S. (1995). Family influences on alcohol abuse and other problem behaviors among Black and white adolescents in a general population sample. In G.M. Boyd, J. Howard, and R.A. Zucker (Eds.), Alcohol problems among adolescents: Current directions in prevention research (pp. 13-31). Hillsdale, NJ: Lawrence Erlbaum Associates.
Baron, J. (1990). Harmful heuristics and the improvement of thinking. In D. Kuhn (Ed.), Developmental perspectives on teaching and learning skills (pp. 28-47). Basel: Karger.
Baron, J., Granato, L., Spranca, M., and Teubal, E. (1993). Decision making biases in children and early adolescents: Exploratory studies. Merrill-Palmer Quarterly, 39, 22-46.
Barrouillet, P., Markovits, H., and Quinn, S. (2002). Developmental and content effects in reasoning with causal conditionals. Journal of Experimental Child Psychology, 81(3), 235-248.
Basch, C.E., DeCicco, I.M., and Malfetti, J.L. (1989). A focus group study on decision processes of young drivers: Reasons that may support a decision to drink and drive. Health Education Quarterly, 16, 389-396.
Bauman, K.E. (1980). Predicting adolescent drug use: Utility, structure, and marijuana. New York: Praeger.
Bauman, K.E., Fisher, L.A., Bryan, E.S., and Chenoweth, R.L. (1984). Antecedents, subjective expected utility, and behavior: A panel study of adolescent cigarette smoking. Addictive Behaviors, 9, 121-136.
Beck, K.H., and Treiman, K.A. (1996). The relationship of social context of drinking, perceived social norms, and parental influence to various drinking patterns of adolescents. Addictive Behaviors, 21(5), 633-644.
Benthin, A., Slovic, P., and Severson, H. (1993). A psychometric study of adolescent risk perception. Journal of Adolescence, 16, 153-168.
Bernbaum, M.H., and Mellars, B.A. (1983). Bayesian inference: Combining base rates with opinions of sources who vary in credibility. Journal of Personality and Social Psychology, 45, 792-804.
Beyth-Marom, R., Austin, L., Fischhoff, B., Palmgren, C., and Jacobs-Quadrel, M. (1993). Perceived consequences of risky behaviors: Adults and adolescents. Developmental Psychology, 29, 549-563.
Bransford, J., Sherwood, R., Vye, N., and Rieser, J. (1986). Teaching thinking and problem solving: Research foundations. American Psychologist, 41(10), 1078-1089.
Byrnes, J.P., and Beilin, H. (1991). The cognitive basis of uncertainty. Human Development, 34, 189-203.
Christensen-Szalanski, J.J., and Beach, L.R. (1982). Experience and the base rate effect. Organizational Behavior and Human Performance, 29, 270-278.
Cochran, L. (1991). Life-shaping decisions. New York: Peter Lang.
Darley, J.M., and Goethals, G.R. (1980). People’s analyses of the causes of ability-linked performances. In L. Berkowitz (Ed.), Advances in experimental and social psychology (Vol. 13, pp. 2-39). New York: Academic Press.
Davidson, D. (1995). The representativeness heuristic and conjunction fallacy effect in children’s decision making. Merrill-Palmer Quarterly, 41, 328-346.
Dawes, R.M. (1988). Rational choice in an uncertain world. San Diego: Harcourt Brace Jovanovich.
Dunkle, M. (1992). Equal access and the development of students’ understanding of a limited open forum. Unpublished manuscript, University of Nebraska.
Elliott, G.R., and Feldman, S.S. (1990). Capturing the adolescent experience. In S.S. Feldman and G.R. Elliott (Eds.), At the threshold: The developing adolescent (pp. 1-14). Cambridge, MA: Harvard University Press.
Erikson, E. (1968). Identity, youth, and crisis. New York: Norton.
Finken, L.L., and Jacobs, J.E. (1996). Consultant choice: A comparison between abortion and other decisions. Journal of Adolescent Research, 11, 235-260.
Finken, L.L., Jacobs, J.E., and Laguna, K. (1998). The role of age, experience, and situational factors in the drinking and driving decisions of college students. Journal of Youth and Adolescence, 27, 493-511.
Fischhoff, B. (1992). Decisions about alcohol: Prevention, intervention, and policy. Alcohol Health and Research World, 16, 257-266.
Fischhoff, B., and Quadrel, M.J. (1991). Adolescent alcohol decisions. Alcohol Health and Research World, 15(1), 43-51.
Fischhoff, P., Slovic, B., and Lichtenstein, S. (1979). Subjective sensitivity analysis. Organizational Behavior and Human Performance, 23, 339-359.
Gardner, W., and Herman, J. (1991). Adolescent AIDS risk taking: A rational choice perspective. In W. Gardner, S. Millstein, and B. Wilcox (Eds.), Adolescents in the AIDS epidemic (pp. 17-34). San Francisco: Jossey-Bass.
Gigerenzer, G., Hoffrage, U., and Kleinboelting, H. (1991). Probabilistic mental models: A Brunswickian theory of confidence. Psychological Review, 98(4), 506-528.
Gilbert, M.A., Bauman, K.E., and Udry, J.R. (1986). A panel study of subjective expected utility for adolescent sexual behavior. Journal of Applied Social Psychology, 16, 745-756.
Gouze, K.R., Strauss, D., and Keating, D.P. (1986, March). Adolescents’ conceptions of stress and coping. Paper presented at the biennial meeting of the Society for Research in Adolescence, Madison, WI.
Greenwald, A.G. (1980). The totalitarian ego. American Psychologist, 35, 603-618.
Hansen, W.B., Raynor, A.E., and Wolkenstein, B.H. (1991). Perceived personal immunity to the consequences of drinking alcohol: The relationship between behavior and perception. Journal of Behavioral Medicine, 14, 205-224.
Hansen, W.B., and Rose, L. (1997). Issues in classification in metaanalysis in substance abuse prevention research. NIDA Research Monograph #170. Washington, DC: National Institute of Drug Abuse.
Harter, S., Waters, P., and Whitesell, N. R. (1998). Relational self-worth: Differences in perceived worth as a person across interpersonal contexts among adolescents. Child Development, 69, 756-766.
Jaccard, J., and Turrisi, R. (1987). Cognitive processes and individual differences in judgments relevant to drunk driving. Journal of Personality and Social Psychology, 53, 135-145.
Jacobs, J.E., Bennett, M.A., and Flanagan, C. (1993). Decision making in one-parent and two-parent families: Influence and information selection. Journal of Early Adolescence, 13, 245-266.
Jacobs, J.E., and Ganzel, A.K. (1993). Decision making in adolescence: Are we asking the wrong question? In M.L. Maehr and P.R. Pintrich (Eds.), Advances in achievement and motivation (Vol. 8, pp. 1-31). Greenwich, CT: JAI Press.
Jacobs, J.E., Greenwald, J.P., and Osgood, D.W. (1995). Developmental differences in base rate estimates of social behaviors and attitudes. Social Development, 4, 165-181.
Jacobs, J.E., and Klaczynski, P.A. (2002). The development of judgment and decision making during childhood and adolescence. Current Directions in Psychological Science, 11, 145-149.
Jacobs, J.E., and Narloch, R.H. (2001). Children’s use of sample size and variability to make social inferences. Journal of Applied Developmental Psychology, 22, 1-21.
Jacobs, J.E., and Potenza, M.T. (1991). The use of judgment heuristics to make social and object decisions: A developmental perspective. Child Development, 62, 166-178.
Jacobs, J.E., and Potenza, M.T. (1992). The effects of risk on adolescent contraceptive behavior. Paper presented at the annual meeting of the Society for Judgment and Decision Making, St. Louis, MO, November.
Johnston, K., and Jacobs, J.E. (2003). Children’s illusory correlations: The role of attentional bias in group impression formation. Journal of Cognition and Development, 4(2), 129-160.
Kahneman, D., and Tversky, A. (1984). Choices, values, and frames. American Psychologist, 39, 341-350.
Klaczynski, P.A. (1997). Bias in adolescents’ everyday reasoning and its relationship with intellectual ability, personal theories, and self-serving motivation. Developmental Psychology, 3, 273-283.
Klaczynski, P.A. (2000). Motivated scientific reasoning biases, epistemological beliefs, and theory polarization: A two-process approach to adolescent cognition. Child Development, 71, 1347-1366.
Klaczynski, P.A. (2001). Framing effects on adolescent task representations, analytic and heuristic processing, and decision making: Implications for the normative-descriptive gap. Journal of Applied Developmental Psychology, 22, 289-309.
Klaczynski, P.A., and Aneja, A. (2002). Development of quantitative reasoning and gender biases. Developmental Psychology, 38(2), 208-222.
Klaczynski, P.A., and Fauth, J. (1997). Developmental differences in memory-based intrusions and self-serving statistical reasoning biases. Merrill-Palmer Quarterly, 43, 539-566.
Klaczynski, P.A., and Gordon, D.H. (1996). Everyday statistical reasoning during adolescence and young adulthood: Motivational, general ability, and developmental influences. Child Development, 67(6), 2873-2892.
Klaczynski, P.A., and Narasimham, G. (1998). Development of scientific reasoning biases: Cognitive versus ego-protective explanations. Developmental Psychology, 34, 175-187.
Klayman, J. (1985). Children’s decision strategies and their adaptation to task characteristics. Organizational Behavior and Human Decision Processes, 35, 179-201.
Kruglanski, A.W., and Ajzen, I. (1983). Bias and error in human judgment. European Journal of Social Psychology, 13, 1-44.
Kuhn, D. (1989). Children and adults as intuitive scientists. Psychological Review, 96, 674-689.
Lewis, C.C. (1981). How adolescents approach decisions: Changes over grades seven to twelve and policy implications. Child Development, 52, 538-544.
Lichtenstein, S., Slovic, P., Fischhoff, B., Layman, M., and Combs, B. (1978). Judged frequency of lethal events. Journal of Experimental Psychology: Human Learning and Memory, 4, 551-578.
Manis, M., Dovalina, I., Avis, N.E., and Cardoze, S. (1980). Base rates can affect individual predictions. Journal of Personality and Social Psychology, 38, 231-248.
Marcia, J.E. (1980). Identity in adolescence. In J. Adelson (Ed.), Handbook of adolescent psychology (pp. 159-177). New York: Wiley.
Markovits, H., and Dumas, C. (1999). Developmental patterns in the understanding of social and physical transivity. Journal of Experimental Child Psychology, 73(2), 95-114.
McNeal, R.B., and Hansen, W.B. (1999). Developmental patterns associated with the onset of drug use: Changes in postulated mediators during adolescence. Journal of Drug Issues, 29, 381-400.
Moshman, D. (1993). Adolescent reasoning and adolescent rights. Human Development, 36(1), 27-40.
Moshman, D., and Franks, B.A. (1986). Development of the concept of inferential validity. Child Development, 57, 153-165.
Nisbett, R.E. (1993). Rules, reasoning, and choice behavior. In G. Harman (Ed.), Conceptions of the human mind: Essays in honor of George A. Miller (pp. 99-110). Hillsdale, NJ: Lawrence Erlbaum Associates.
Nisbett, R.E., and Kunda, Z. (1985). Perception of social distributions. Journal of Personality and Social Psychology, 48, 297-311.
Nisbett, R., and Ross, L. (1980). Human inference: Strategies and shortcomings of social judgments. Englewood Cliffs, NJ: Prentice-Hall.
Nucci, L., Guerra, N., and Lee, J. (1991). Adolescent judgments of the personal, prudential, and normative aspects of drug usage. Developmental Psychology, 27, 841-848.
Olds, R.S., and Thombs, D.L. (2001). The relationship of adolescent perceptions of peer norms and parent involvement to cigarette and alcohol use. Journal of School Health, 71(6), 223-228.
Overton, W.F. (1990). Competence and procedures: Constraints on the development of logical reasoning. In W.F. Overton (Ed.), Reasoning, necessity, and logic: Developmental perspectives (pp. 1-32). Hillsdale, NJ: Lawrence Erlbaum Associates.
Paikoff, R.L. (1990). Attitudes toward consequences of pregnancy in young women attending a family planning clinic. Journal of Adolescent Research, 5, 467-484.
Plous, S. (1993). The psychology of judgment and decision making. New York: McGraw-Hill.
Quadrel, M., Fischhoff, B., Fischhoff, M., and Halpern, S. (1992). Calibration of adolescents’ and adults’ confidence judgments. Unpublished manuscript, Carnegie Mellon University, Pittsburgh, PA.
Quadrel, M. J., Fischhoff, B., and Davis, W. (1993). Adolescent (in)vulnerability. American Psychologist, 48(2), 102-116.
Reyna, V.F., and Ellis, S.C. (1994). Fuzzy-trace theory and framing effects in children’s risky decision making. Psychological Science, 5, 275-279.
Rowe, K.L. (1984, July). Adolescent contraceptive use: The role of cognitive factors. Paper presented at the annual meeting of the American Psychological Association, Toronto, Canada.
Shtarkshail, R.A. (1987). An examination of the concept of “risk taking” and its application to the contraceptive behavior of youth. International Journal of Adolescent Medicine and Health, 3(2), 121-144.
Sternberg, R., and Martin, M. (1988). When teaching thinking does not work, what goes wrong? Teachers College Record, 89(4), 555-578
Trope, Y. (1986). Self-enhancement and self-assessment in achievement behavior. In R.M. Sorrentino and E.T. Higgins (Eds.), Handbook of motivation and cognition: Foundations of social behavior (pp. 350-378). New York: Guilford.
Urberg, K.A., and Rosen, R.A. (1987). Age differences in adolescent decision making. Journal of Adolescent Research, 2, 447-454.
Zukier, H. (1986). The paradigmatic and narrative modes in goal-guided inference. In R.M. Sorrentino and E.T. Higgins (Eds.), Handbook of motivation and cognition: Foundations of social behavior (pp. 465-502). New York: Guilford.
Zukier, H., and Pepitone, A. (1984). Social roles and strategies in prediction: Some determinants of the use of base rate information. Journal of Personality and Social Psychology, 47, 349-360.