Have you ever looked back on your teenage years and thought, “Why did I ever think that was a good idea?” Just about everyone regrets some risk that they took in their adolescence. When we’re lucky, such regrettable risks are mild – a particularly creative outfit, perhaps, or hitting on an object of affection that was out of our league. When we’re not so lucky, adolescent risk-taking can have more dire consequences – teen pregnancies, drug problems, car accidents, and so on.
Scientists who study how teens make decisions often paint adolescents as irresponsible risk-seekers who take risks for the thrill of it all and end up getting into lots of trouble. This vision of adolescence helps you get grants. If you can argue that understanding how teens make decisions can help prevent teenagers from engaging in risky sexual behaviors, drinking and driving, and experimenting with drugs, then you’re more likely to get funding for your studies.
A recent study from the Proceedings of the National Academy of Sciences, however, suggests that this view of adolescence is inaccurate. Based on their data, Tymula and colleagues argue that teens actually take fewer risks than adults do when those risks are made explicit. When those risks are uncertain and ambiguous, however, adults shy away from risk while teens are unaffected.
The paper: Tymula, A., Rosenberg Belmaker, L. A., Roy, A. K., Ruderman, L., Manson, K., Glimcher, P. W., & Levy, I. (2012). Adolescents’ risk-taking behavior is driven by tolerance to ambiguity. PNAS, 109(42), 17135-17140.
The abstract (emphasis mine): “Adolescents engage in a wide range of risky behaviors that their older peers shun, and at an enormous cost. Despite being older, stronger, and healthier than children, adolescents face twice the risk of mortality and morbidity faced by their younger peers. Are adolescents really risk-seekers or does some richer underlying preference drive their love of the uncertain? To answer that question, we used standard experimental economic methods to assess the attitudes of 65 individuals ranging in age from 12 to 50 toward risk and ambiguity. Perhaps surprisingly, we found that adolescents were, if anything, more averse to clearly stated risks than their older peers. What distinguished adolescents was their willingness to accept ambiguous conditions—situations in which the likelihood of winning and losing is unknown. Though adults find ambiguous monetary lotteries undesirable, adolescents find them tolerable. This finding suggests that the higher level of risk-taking observed among adolescents may reflect a higher tolerance for the unknown. Biologically, such a tolerance may make sense, because it would allow young organisms to take better advantage of learning opportunities; it also suggests that policies that seek to inform adolescents of the risks, costs, and benefits of unexperienced dangerous behaviors may be effective and, when appropriate, could be used to complement policies that limit their experiences.”
The key to understanding this study is understanding the difference between risk and ambiguity. A risky situation is one in which you’re not sure which outcome will occur, but you know each outcome’s probability of occuring. Rolling a regular die is a risky situation. You don’t know what number is going to come up, but you do know that there is a 1/6 chance of each number being rolled.
Above is an example of a risky trial from the study. Participants could choose to take $5 for sure, or they could a take a gamble with a 50% chance of winning nothing (red box) and a 50% chance of winning $50 (blue box). Riskiness on each trial was varied by changing the gamble’s win amount (making the blue box have a value between $5 and $125) and/or the probability of winning (making the blue box represent a 13% to 75% chance of winning).
An ambiguous situation is one in which you’re not sure which outcome will occur, and you don’t know each outcome’s probability. Rolling a trick die that you’ve never seen before is an ambiguous situation. You don’t know what number is going to come up, and you don’t know how the die is tricked (weighted towards a particular number), so you don’t know what each number’s probabilities are either.
Above is an example of an ambiguous trial in this study. Participants could choose to take $5 for sure, or they could take a gamble with an unknown probability. In this case, they know that they have at least a 25% chance of winning $20 (red box) and at least a 25% chance of winning nothing (blue box), but the grey box covers up the rest. The true probability of winning $20 could range from 25% to 75%, so this trial is ambiguously risky.
The experimenters varied the degree of ambiguity by changing the size of that covering gray bar. The bigger the bar, the more ambiguous that trial was.
In this study, adolescent participants between the ages of 12 and 17 and adult participants between the ages of 30 and 50 were given different risky and ambiguous bets. The experimenters then used modeling techniques on participants’ choices to determine each participant’s attitude towards risk and ambiguity. They calculated each participant’s risk aversion, a measure of how much they preferred the sure $5 to the risky bets, and each participant’s ambiguity aversion, a measure of how much they preferred the sure $5 to the ambiguously risky bets.
What they found was surprising. Contrary to popular belief, adolescent participants were not more risk-seeking than adult participants. In fact, adolescents were, on average, more averse to gambling on the trials in which risk was made explicit.
The ambiguous trials had even more surprising results. Adolescents were less ambiguity averse than adults. That is, adolescents weren’t scared off by those gray bars that covered up a gamble’s true probability as much as adults were. Furthermore, while adults increasingly avoided the ambiguous gambles as the size of those gray bars increased, adolescents weren’t as affected by the size of the bars.
The below figure shows one example adult (top) and one example adolescent (bottom). The orangey curved lines on the left graphs show each participant’s probability of choosing to gamble instead of taking the sure $5 when the risks were explicit, with each colored line representing a different probability of winning that gamble. You can easily see that both the adult (top) and the teen (bottom) change their gambling likelihood as explicit probability of winning changes. Lower probabilities of winning have to be paired with higher potential win amounts in order for participants to want to gamble.
The blue curved lines on the right graphs show each participant’s probability of gambling, with each colored line representing a different level of ambiguity. You can see that the top adult adjusts his/her gambling preference based on the amount of ambiguity, so that more ambiguous gambles need to be paired with higher potential win amounts in order to make the gamble more attractive than the sure $5.
The bottom adolescent, however, has his/her blue lines all overlapping. This means that the amount of ambiguity had no effect on his/her gambling. Just by eyeballing this particular adolescent, it looks like he/she treated any ambiguous gamble about the same as a 50-50 gamble (compare the blue lines to the orange lines).
The study’s authors suggest that what we’ve all assumed to be risk-taking behavior on the part of teenagers may actually be not-scared-of-the-unknown behavior. It’s not that teens aren’t averse to risk – the risk trials of this study show that they are! It’s that teens aren’t averse to unknown risks.
Let’s take this into a real-world example that will bring us oodles of grant funding. Teenagers know that it’s possible to get pregnant from unprotected sex, but they also know that they’re not going to get pregnant every time that they have unprotected sex. For many adults, a risk of pregnancy that’s somewhere between 0 and 100% is too big a risk to take; for teens, 0-100% may be an acceptable amount of unknown risk.
Perhaps this means that the best preventative education for adolescents involves being as honest as possible with them. This BBC article references a study that suggests the true risk of getting pregnant from a single act of unprotected sex is somewhere around 5%. For adolescents, that explicit 5% may be scarier than an ambiguous 0-100%.
Now that the authors have found this lack of ambiguity aversion in teens, the next science step is to figure out why. Are teenagers just overly optimistic and assume that the best possible outcome will happen to them? Are adults overly pessimistic at always assuming the worst? Do adults see the unknown as something to be feared, while teenagers view it as something to be explored? I’m sure plenty of future grant applications will want to find out.
Tymula A, Rosenberg Belmaker LA, Roy AK, Ruderman L, Manson K, Glimcher PW, & Levy I (2012). Adolescents’ risk-taking behavior is driven by tolerance to ambiguity. Proceedings of the National Academy of Sciences of the United States of America, 109 (42), 17135-40 PMID: 23027965