VALERIE REYNA: This is kind of an overview of the argument I'm going to give. To begin with, of course, adolescent risk taking has ramifications for a wide variety of outcomes. We're going to talk about some of those.
But it also, in particular, has ramifications for educational attainment and educational achievement. Now that's education speak for do people graduate and do they learn anything? Right?
And we're talking about a range of behaviors, things like crime, drug taking and drug abuse, alcohol use, unintended pregnancy, all the different kinds of risk taking. Driving fast while on the highway while drunk, for example, that would be risk taking behavior, and so on.
Many of these behaviors that ultimately become lifetime patterns, they debut in adolescence or young adulthood. They certainly often reach their peak, according to many studies, in that time period. And then sometimes those habits are sustained across a lifetime. And we'll talk about that.
Obviously, if we can understand this better, if we can do something about it on behalf of young people to improve their health and help them take healthy risks, for example, like going to college, that would be a healthy risk for people from many, many backgrounds. It's a risk, because it may not be what you're used to. But it's a healthy, positive risk.
So if we could have an impact, even a slight impact on this vast range of unhealthy behavior, we can make a difference. At least, we hope so. And that's what the data suggests.
Now what do we mean by risk taking? Well, a picture is worth a thousand words, isn't it? This was actually at the beginning of a Scientific American Mind article that was about my theory. So those are the kinds of behaviors we're talking about.
Well, let's talk a little bit about statistics. And I'm going to go through several slides, probably pretty quickly. I think you know this already, but just sort of to give us a sense of the prevalence of the problems that we're talking about in the real world.
So we're talking about things like auto accidents that both males and females who are 16 to 20 years of age have twice the chance of being in an auto accident than people 20 to 50. That auto accidents are a leading cause of death among young people. And many of them who were killed had been drinking.
40% of alcoholics report having problem drinking between the ages of 15 and 19. Now some of those studies suggest that this early drinking is a marker of a genetic risk factor. Other studies suggest that there's an independent effect of exposure to alcohol.
So if you start drinking very, very early and you're already having a problem, that signals a lifetime issue. And it's both a dose response effect, the more you're exposed, the more likely you are to head down a certain path and then lead to alcoholism. But it's also an indication of a problem. So in some of the latest research which I've seen, which isn't quite published yet, suggests that both of these factors are at work in this very, very young age group.
Pathological gambling, this is something that years ago, we didn't even think of among college student populations, it turns out the prevalence rate of that is enormously high. And of course, I could go on. And I will.
There's risk taking in adolescence that has to do with things like sexually transmitted diseases and sexually transmitted infections. This is one of the topics that concerns folks in my lab. And I see some of the team members right here.
Nine million cases of sexually transmitted diseases among 15 to 24-year-olds. One in four adolescent girls has an STD, one in four. OK? Between 1994 and 2003, there were decreases in some of these troubling statistics. But the decreases didn't happen for young people, they happened for a little bit older, young adults.
So those 25 to 34 years of age, new HIV diagnosis did go down. Hallelujah. They went down. But 13 to 24-year-olds during that same time period, new HIV diagnoses remained stable. So we're not getting the message across to the adolescents the way, perhaps, we may be getting the message across to some of the younger adults. There's a different issue there.
So there's obviously a mental, physical, and economic toll from this sort of thing. And then pregnancy. Now many of us thought that the teen pregnancy rate was becoming a receding issue, right? You've heard this on the news. And this is not a problem anymore.
Well, that's not quite true, it turns out, if you look at the most recent statistics. For example, 51% of Latina teens become pregnant. 51% become pregnant, teens. OK?
There's some evidence that the progress we've had recently has begun to slow, if you look at the curve. 3 in 10 teams become pregnant by age 20. So we're talking 750,000 teens annually, and so on.
So what about the effect on educational outcomes? And, again, I'm just going to give you a sampling, just to thoroughly depress you. No, actually, it's to give you a sense of the extent of the problem.
So what does it mean? So a young person becomes pregnant. Well, it turns out that parenthood is the leading cause of school dropout among teenage girls, the leading cause still. Just like when some of us were growing up, it has not changed.
We think of this is a solved problem. It's not. Dropping out in today's world is a worse outcome than it was 20, 25 years ago. Because the demands on us for levels of education are higher than they've ever been. And yet this is the number one, preventable, clearly, cause.
And what about the outcomes? Well, let's compare teen mothers to young people who wait until they're 20 to 21 to have a child, at least 20 to 21. All right? So fewer than 2% of teen mothers have a college degree by age 30. Only 40% of teen mothers ever graduate from high school, compared to 75% of young women who wait to become pregnant. Obviously, 2% versus 9% going to college.
Children of teen mothers are more likely. So you see this kind of going to the next generation. It isn't just the young person themselves who becomes pregnant. It's their children and their children and so on. So this propagates, this risk taking defect propagates to that next generation.
Children of teen mothers are more likely to drop out. 2/3 of children born to teen mothers earn a high school degree. But this compares to 81% of people who bear children later in life. So that's 66% approximately to 81%.
Children of teen mothers score lower on measures of readiness to come to school, things like cognition, language, communication, and interpersonal skills. And their outcomes are different. The children of teen mothers do worse in school. They're more likely to repeat a grade. They're less likely to complete high school than the children of older mothers. And they have lower performance on standardized tests. So these are the children of the children who are being left behind educationally.
So there's a strong relationship between academic failure and teen pregnancy. And this causation goes-- and if you're thinking to yourself, well, which way is the causation? Is it that becoming pregnant leads to bad educational outcomes? Or is it that bad educational outcomes, failure, lower achievement in school leads to teen pregnancy and dropping out?
And the answer is yes to both of these. The causation goes in both directions. So if you care about education, if you care about education, you've got to care about teen pregnancy. You've got to care about these kinds of risk taking behaviors.
And this is only touching on some of the risk taking behaviors. Obviously, think about all the others, too, and how they can compromise the educational achievement of young people. So this is something we have to understand.
Now that I thoroughly have depressed you and talked about all the horrible outcomes of being poor and being at risk and all of that, I do want to give you a ray of hope. Does anybody know who this is. Does anybody recognize him?
This is Richard Carmona. He was US Surgeon General of the United States of America in the Bush administration. Richard Carmona is actually somebody that I met in Tucson, Arizona.
I met him the first time when we were working on a grant on Christmas Eve. All right. All right. But it was for youth.
We were both up-- you know, we were in a building. They had to let us in, this committee. We were writing some grant with a county agency to try to get some money to help teens at risk. And he was very much committed to that.
And one of the reasons he was committed to that is because he had been a youth at risk. This is a man who was born in Harlem, who grew up and got in trouble. He was sort of a juvenile delinquent. I'm not telling this as something that everybody doesn't know. He talks about it.
He got in trouble with the law. He was having trouble in school. He was heading down the wrong path. He was one of the millions of Hispanic American young people who were at risk.
And he was a high school dropout. So we should write him off, right? Based on all the statistics I just gave you, we should say-- but here he is. He's in this fancy uniform.
So something happened in between the writing him off, the high school dropout in trouble with the law, and becoming a doctor and, eventually, surgeon general of the United States. And he went to Vietnam. He got his GED.
He became a medic. He loved jumping out of helicopters and taking risks, because he was a risk-taking type of individual. He became a trauma surgeon, ultimately, after he became a nurse, and so on, and so forth.
He opened up a trauma center in southern Arizona. He joined the SWAT team because he loved taking those risks, you know? But he became somebody that, instead of being a statistic, a casualty of these risk factors that we talked before, he became a success story.
And now he's given back. And he continues to be interested in the fate of young people. So I wanted to tell you, si, se puede!
Which, loosely translated, means "yes, we can." Right? So despite all of the depressing statistics and the negative consequences, we have to keep these outcomes in mind. Because this is the possibility. And this is what we work for every day. And I know many of you do. So thank you.
So I'm going to go through a lot of things fairly quickly in this talk. But don't worry. There are many boring publications available on my website that you can download at this email address.
All you have to do is put your email address in and it sends you back the publication. We love our new software. And thanks to Kareem Booker for making that possible.
So I'm going to be drawing some of the statistics and so on that I'm going to be talking about from a review paper I wrote with Frank Farley, published in 2006. This paper is available free. It's an open access paper. So if any of the work is useful to you, feel free to post it on your websites, to send it to people, and so on. And that's a result of the Association for Psychological Science.
I'm also going to be drawing from a special issue of a journal called Developmental Review. This is a 2008 special issue. And it featured some of the leading theorists in this area. So it's up to date and current.
And that includes people like Baruch Fischhoff, who talks about behavioral decision making; Meg Girard and Rick Gibbons, who talk about things like the prototype willingness model. It's sort of an extension of the theory of reasoned action. Don't worry. Later on, if people want to go into detail, I can tell you about it.
There's a wonderful couple of articles on the brain. And they're written at a level that, I think, they're very tutorial and easy to read. I found them, at least, that way. And I think they're really a great entree to that kind of literature on how the brain changes with age.
And then, of course, our lab has a contribution there. We talk about emotion. And then, finally, Cass Sunstein, who is a policymaker and legal scholar, who talks about the policy implications for understanding how teenagers make decisions. So I'm going to be drawing on some of these sources.
So back to the real world, if you have a program and you want it to be effective, there's basically three questions that they say in the literature we should all be asking ourselves. And those three questions are what's the ideal, what behaviors should the program foster? That's a should question. As scientists, sometimes, we shrink from that question.
But if we want to change people's future, and we want to change their behavior, we have to think, what's ideal? What are the right attitudes and the right behaviors? And this is very kind of scary a little bit, right? But we have to think about that seriously.
Descriptive, how are adolescents making decisions currently when you don't intervene in any way? And then, finally, this third thing, which often isn't recognized, which is that bridge between the ideal and the real. You can't just say people ought to be perfect. And they ought to do this right. And they ought not to have problems.
You know, you need a bridge, a practical way to go from where people are to where they need to be. Right? And that's what prescription's about. It's not just do the right thing. You have to have more than that.
So there's a science of all three of these things. Right? So better decisions, ideal, what should you do, what's a rational decision, description? How do teenagers actually make decisions? And prescription, how can we help teens make better decisions.
So let's start with the first one, the normative ideal. Now you would think, based on your life experience and so on, that's a simple question. Right? Don't drive drunk in the car on the expressway at 95 miles an hour. Simple, right?
Well, when you get into it, it turns out it's not so simple what people should do. Some people think it's very normal for teenagers to take risk. It's part of evolution. That's part of the evolutionary thing, especially if you're a young man.
You want that mate. You have to show off. You have to prove that you can take physical risks. That's the way. You know, it's in our genes to do this, to explore.
They study adolescent rats, for example. Adolescent rats explore a lot. They're searching for a date, I gather. I don't know. I don't know much about rats.
But on the other hand, of course, in modern society, if you just are driven by this evolutionary imperative, then it's, what? Early procreation and so on and there's this need to get this protracted education that doesn't quite jibe with what evolution says is the adaptive right thing to do. So sometimes we fight against the evolutionary urges, don't we?
Well, then there's traditional economic models. And economic models often say, well, the rational thing to do is to reach your goals. What if you're a teenager? What are your goals?
What if your goal is to get drunk every Saturday and that will make you happy. And you will reach your goals. Right? Is that a rational decision? And these are serious debates. Like, how do you know what the right decision is?
Well, often adolescents want to maximize their immediate pleasure, the short term goals as opposed to long term goals. I hate to break it to you young people, but that does change. Well, it does, the time. There has been research on it.
Your time horizon does get longer as you get older. And you're seeing more far out, as opposed to weighing that immediate a little bit more. And that changes developmentally.
So we know that for most people, there are changes in goals when you move from being a teenager to being an adult. And we call that developmental difference. Right? There are developmental differences.
And it's important, I think, for policy we decided to promote these positive long term goals as opposed to the immediate short term goals, which might emphasize things like immediate fun. All right? Not that you shouldn't have any fun. I'm not against fun.
Remember that I said that. All right? All right. Thank you.
So this is a concrete example. Now I have a lot of statistics in this paper about suicide. But this is a concrete example. And I think it gets at-- it illiustrates the concept I'm trying to get at here. This is a real note that was published in the Cornell Daily Sun on September 14, 2005.
And it's the report of someone who says 10 years ago, I tried to kill myself in the A lot. Now the A lot is a parking lot. This person tried to commit suicide.
And this letter was published saying that they're grateful that they were prevented from committing suicide in the A lot 10 years earlier. And this is a letter to the Cornell community and also, this person thanks their parents, the paramedics, and so on for saving them.
So their short term goal, obviously, was to be successful in killing themselves. But as many studies have shown, at least for the survivors, we don't know about the people that don't survive suicide attempts, but for the survivors, the vast majority say it was a permanent solution to a temporary problem. And I'm glad I was stopped from doing this.
So as policy people, as people who think about affecting the lives of other people, we have to think about what would people ultimately prefer, not what they want necessarily at the moment. And it's this kind of thing that I keep in mind. So we think about the future self. So let me give you a little bit of drill down into these developmental differences that research has identified. And we're going to talk about some concrete examples, too.
So in principle, if you look at the adolescent mind cognitively, they're at the apex of cognition. Your memory is the best it will ever be, right? Sorry. But there other things that are more important than memory, like judgment.
And so they're capable of rational decision making. They can solve logic problems very well in the abstract. But life is not abstract. Life decisions come with the heat of emotion, the heat of passion, the things you're not familiar with, and so on. And that's when this impeccable, logical machine of the teenage brain sometimes can break down.
So research has shown for example, that the presence of peers affects risk taking more in younger people than in older people. There's this wonderful study by Larry Steinberg where he has the simulated driving machine. Anybody hear about this? It's so cool.
Anyway, it's in the lab. But the only difference is, so you're driving along, and if you see a yellow light, if you hit the gas, sometimes you can make it through the intersection. But every so often, you crash into a wall.
So the key is to make it through that intersection and get the maximum number of points without killing yourself by crashing into the wall. Right? So guess what happens? Of course, the younger people drive faster, take more risks, crash into the wall more often.
But here's the key part. All you have to do is put some of their friends in the room with them, and guess what? Risk taking goes up.
These friends don't have to say anything. They don't have to be go for it! Do it! Oh, yeah! None of that.
They're sitting quietly. They're just there. That's it. That's all you need.
And the effect on risk taking for teenagers is bigger. There's still a significant effect on college students-- yes. Sorry about that.
And then you get older and the effect goes away. You're as risk taking when they're in the room, your peers, as when they're not in the room. So just the mere presence of the peers causes the risk taking to go up.
And obviously brain maturation just changes. Seeing a lot of advertising by the insurance industry talking about the teenage brain and auto accidents, and I'm going to talk a little bit about the details of that. But yes, the brain continues to mature through the decade of the 20s and so on. So things are not completely developed and done during these teenage years.
And the thinking process itself changes. And this is some of the newest data that we have. And it's been generated right here at Cornell.
And that is that people shift-- and I'm going to talk about this again. Because it's a very counterintuitive idea. But young people shift from a much more logical pros and cons, weighing the costs and benefits of taking the risk, to a more categorical kind of risk avoidance that you get when you get to be older.
And by categorical, I mean for the risks that are really catastrophic. Not all risks, but things, like, oh, getting into the car with a drunk driver, having unprotected sex, that sort of thing. And you get to be a certain age and you go, just don't go there. Right? Don't take that risk.
It doesn't matter if this potential sexual partner's had few partners or a lot of partners. You just don't necessarily-- you just don't go there. You don't have unprotected sex.
You know? They could be more risky. They could be less risky. You don't do that.
So the analogy I like to use, and this is something I use, actually, in some of our training with young people is Russian roulette. OK? Everybody know what Russian roulette is?
It's amazing. All the teenagers seem to know, right? TV, they cover Russia-- I have no idea.
So Russian roulette is you have six chambers in the gun. And one of them has a bullet in it. And you twirl the thing. And you put it up to your head and you go click, right?
Right. So if I were to give you a million dollars-- see? Aren't you glad you sat in the front? If I were to give you a million dollars, would you play Russian roulette for a million dollars?
VALERIE REYNA: OK. Good. Oh, thank God. Your mother's so happy. And we have it on tape, too.
That's good. Because what do you think teenagers say? Half of them say yes.
Because a million dollars is a lot of money. And there's only a 1 out of 6 chance. There's sort of a logic to it. Right? I mean, it's very mathematical.
And that's exactly my point. Our recent research has shown that adolescents approach this like a math problem. And you weren't thinking-- does it matter if I, like, fewer bullets, more bullets, would you play? Like, if there was, like, 1 out of 8, would it be OK?
VALERIE REYNA: OK. Good. Phew. I'm really glad. Thank you. Exactly. And now adults are going, like, that would be crazy. Right? That would be irrational to play this.
So half the kids will say yes right away. Because it's a million dollars and it's only one out of six. The other half of the kids say, OK, for a billion, I'd do it.
It still doesn't make it right. It's irrational. That's a crazy way to think. And yet, if you look at it from a rational, economic model point of view, they're sane. And we're crazy. Right?
It's the other way around, right? Because they're doing the cost-benefit analysis. How many bullets, how many dollars, there ought to be a point, right, where let's make a deal. Right? No! This is not the kind of thing that you make a deal about.
So we think of it that way. We think of the bottom line gist as being determinate. The bottom line of this, are you crazy? You could die! You know?
So I say to them things, like, this might hurt. You might not die right away. Yeah. We know that. We know all about it. But it's only a 1 out of 6 chance.
So you see there's a difference in the way we think. So adolescents are much more likely to do kind of a cost-benefit computational analysis, whereas adults do this much more primitive in one level categorical bottom line gist. And I'll show you some data to that effect. All right.
So decision processes develop. And one of the surprising things that came out of this review of the literature is we're often told that young people think they're immortal. Has anyone heard that? OK.
Everybody says that. When you're interviewing so-called experts and so on, I say so-called, because they may have evidence. They may not. They have opinions, though.
And they say, OK, young people just don't realize that they could die. They think they're going to live forever. That turns out not to be true.
When you ask young people, in fact, they have an exaggerated sense of their own mortality on average. They overestimate the probability that they're going to die before they get to be age 25. And some people think that that might promote risk. Well, I'm going to be dead, anyway.
But that doesn't really turn out in the data to really be true. But if anything, they certainly don't think they're immortal. They think they're more mortal than they are. It's the opposite of what we used to think. Apologies to David Elkind and so on.
So they overestimate a lot of the key risks as well. So we think they underestimate. We think, oh, they think they're invulnerable. And they don't understand the risks.
If they understood the risk-- I know you gotta go-- if they understood the risks, then they wouldn't take these risks. Right? Turns out if you ask them what's the probability of dying of lung cancer if you smoke, and so on, they overestimate the probability.
What's the chance of getting HIV from a single act of unprotected sex? They way overestimate that probability. So why do they take risks then?
They overestimate the probabilities because the benefits outweigh the risks. The risks are very high, but the benefits are even higher. And they rationally calculate them, like a balance beam, right? And then out comes the risk taking behavior. So it's in spite of the fact that they overrate the overestimate the risks that they take, they engage in these behaviors, in spite of that.
So let's look at some real data. This is 255 high school students age 14 to 17. This is from my laboratory.
And we asked them, on a 0 to 100% scale, what was the probability that an average teenage girl who didn't have any symptoms at all would have any one of the following sexually transmitted infections. And you'd think, OK, do they know what chlamydia is, gonorrhea, HIV/AIDS, HPV, human papilloma, syphilis? Yes, they do. They get this in health class. Right?
And we asked if they didn't know. We would tell them what it is. But the vast majority knew what we were talking about. And they had a very definite and high estimate.
And it turns out if you take the highest published estimate of the actual risk, that's what's in the dark bar right here. And as you can see, they way overestimate the probability that a sexually active girl would have any one of these sexually transmitted infections. So they're overestimating the risk, not underestimating the risk.
If you take out all the data that's just 50-50, so you think, well, maybe that's just uncertainty, they're still hugely overestimating the risk. So it's not that they think they're invulnerable.
And here's a slide. Don't worry about reading all this. I'll translate it for you.
It starts off with the youngest kids in fifth grade here. And it gets to the oldest this way. So fifth graders in blue, seventh graders in red, then ninth graders in yellow, and this is young adults.
And this is three different kinds of consequences. What's the probability that you will get ill from ingesting alcohol? What's your perceived probability, 0 to 100, that you will get involved in a drug-related accident, you will be a driver in a drug-related accident, that you will be a passenger in a drug-related accident? And this is their own estimated probabilities.
And as you can see, as they get older, those probabilities go down, not up. So they think it's hugely likely that these bad things will happen to them when they're younger. The younger teenagers and middle school students here, and even elementary students, they overestimate their risk relative to young adults. And by the way, everybody's overestimating the risk in this graph.
So this has been replicated now in many multiple, independent laboratories that, in fact, teenagers overestimate their risk. And so do, in fact, other age groups. But they do more so.
Now does perceived risk matter at all? We said that they overestimate the risk. And this is a large study called Monitoring The Future that tracks risk perception and risk taking over many, many years.
As you can see, it goes from 1975 to 2003. Risk perception does matter. OK? So how much you perceive the risk of something seems to be related unbelievably well to actual use. So here you have the actual use of marijuana among 12th graders. That's the little round bars right here.
And here you have the perceived risk of marijuana among 12th graders. And as you can see, the lower the perceived risk, the higher the behavior. And you can track this over-- they just track each other amazingly well. This is real, self-reported data over many, many years.
So risk perception matters. And we are getting the message across in health classes. You're at risk. You're at risk. You're at risk.
And they're going, I'm at risk. I'm at risk. I'm at risk. Yeah.
And it does have an effect. And the point here is that it's not that it's just completely chaotic, impulsive behavior. Their behavior is a computational function of the perceived risk and the perceived benefit.
It's like a math problem. Right? That's how they're thinking. So it's no wonder that their behavior tracks their risk perception.
So this is kind of a complicated slide. Don't worry. I won't spend long on it. And basically, if you go to the paper, there's more detail.
But this is an experimental study using the Iowa Gambling Task with young people. And it starts off with very, very young people. That is what, like 6-year-olds, something like that, 6 to 9-year-olds, and then 10 to 12-year-olds, 13 to 15, and then 18 to 25, or college students.
And basically, there are decks of cards. And you can sample from the decks of cards. And this is used as an assessment task for people with brain damage of a very specific type.
And guess what the effect of this brain damage is? It makes you take risks. You have this kind of very focal damage.
You go out there, and you buy used cars that you can't afford. And you divorce your wife. And you borrow money you can't afford. And you do all kinds of things like that that involve risk taking.
And yet your cognitive scores, your IQ is still very high. So this very focal kind of risk taking, also the people who do that tend to do poorly on this card task that involves drawing from different decks. So there's basically four decks. And two of them are good decks and two of them are bad decks.
And the good decks are the decks that eventually pay off, slow and steady. You don't win big. You don't lose big. Slow and steady, eventually, you work hard. And in the end, you make enough to survive.
Does this sound like a sort of parable of life? Right? Right? But there's not those big possibilities of big wins and big losses.
The bad decks have big wins, but big losses. And guess where you end up in the end? In the poorhouse, all right? You end up behind in the end.
So you have that thrill of those big wins. But in the end, it's bad. You end up owing money.
So eventually you get feedback when you draw these cards of which ones are the good ones, slow and steady wins the race, and which ones are the high risk decks. And eventually what happens, people start off, oh, I likely that high win. Uh, oh, right?
And this is the bad deck. So you slowly don't prefer the bad deck anymore. About the end of many, many trials of learning about outcomes and consequences, you don't pick the bad deck anymore. You pick the good deck. This is if you're older.
This learning is much, much slower and not so good if you're younger. So it takes you longer to catch on punishment, pain, punishment, pain! The School of Hard Knocks just-- it takes more knocks to learn the same thing if you're younger than if you're older. And you don't separate them quite as clearly as you do.
So your ability to learn from consequences gets better and better as you get older. So we say, oh, well, we'll let them out in the world and they'll learn. Well, hopefully they'll survive. Because it will take them more opportunities.
So the early adolescents, the older adolescent, these are two different people. Right? So the older adolescent maybe could learn from experience and bad consequences. The younger adolescent might benefit from monitoring and not having opportunities to take risks. So that's just part of the story.
So risk taking is predicted by adolescents' perceptions of risks and benefits. Important risks are often overestimated. This is sort of a summary.
They don't believe they're invulnerable. That's a myth. Benefits, however, loom larger than risks for young people.
There are benefits to some of these behaviors to them. And we have to acknowledge that. And learning from punishment occurs more slowly. And benefits loom larger than risk.
But then the question is why? And I'm going to give you some sense of some of the theoretical ideas that have been used to try to explain this. And I'm going to just go over, I hope, just the highlights of that.
So fuzzy choice theory is a theory that we've developed over the years based on laboratory risk taking tests, as well as looking at real life surveys, and so on. We do field studies and we do lab studies. And we try to bring those two together.
And the idea here, of course, is to explain why adolescents perceive risks and benefits and yet take more risks than adults, to explain how risk taking changes developmentally as people get older, what changes, and hopefully provide guidance about educational approaches that can reduce risk taking. In our most recent work, we've taken some of these laboratory theoretical ideas and we designed an educational intervention. We did a randomized clinical trial with three arms, just like they do in studies of medicine.
And we looked at both short term outcomes, but also long term outcomes, things like pregnancy rates, how often you have unprotected sex, that sort of thing. And it was a curriculum design, based on our theoretical principles, to reduce this unhealthy risk taking. And we compared it to a standard health curriculum and a control group that got-- another thing that we thought would be helpful wasn't related to these kinds of outcomes.
And in fact, and this is very recent data, we reported a significant benefit for the arm that was based on the theoretical work that we've done here. That after 12 months, young people who had gotten this more gist-based, bottom line, categorical kind of education about risk-- not how many bullets are in the chamber and how big is the payoff-- but don't even think about it kind of thing, that whole kind of intervention produced fewer adverse outcomes and kids were protected longer. So we're writing that up right now.
So let me give you a kind of example of some of the laboratory tasks. This is a scenario task that we based on real life stories that people had told us and had told the folks working with us. Sonya's 17. She's a junior in high school.
She's been sexually active with many partners in her life and has never used condoms. Sonya met Juan at homecoming last fall and they've been a couple ever since.
Juan has never had sex before. At a party on Saturday night, et cetera-- so we have two people. They're about to. And they're a third party.
This is not yourself. This is Juan and Sonya. And I want to ask you how you perceive the risks and benefits if these two people were to have sex. All right? And so we asked young people what did they perceive the benefits of having sex for Juan? What do they perceive the benefits of having sex for Sonya?
Do they perceive it as none, which we coded as zero, small, medium, and large? So people actually picked from none, small, medium, and large. And the risks of having sex, do you see those as none, low, medium, or high? And subjects checked off a verbal label.
And what do you think they found? Do you think the benefits of having sex in our culture are equal for men and women, for this man and the young man and the young woman? Aha. We shall see what they thought.
This is the control scenario. And in this scenario, both people had never been sexually active before. So it's sort of a baseline. So Juan and Sonya, imagine that neither of them had sex before. They're about to have sex.
Well, the male subjects perceived more benefits to either Juan or Sonya having sex than the female subjects. So male teenagers said there's more benefits to be had by having sex here. They perceived, looking at the same scenario, they saw more benefits than that.
But let's break that down. Was that true across different cultural groups, different ethnic groups? What would you think about traditional cultures versus less traditional cultures, how they would perceive gender differences and the potential benefits of sex? Remember, we're thinking that teenagers use perceived benefits to make their decisions. Right?
So here we broke this down into the Anglo teenagers, mixed, meaning one parent is Hispanic, Mexican-American in this case, and one is not, and then the Hispanic kids, where both parents were Hispanic. So the idea is this is the most traditional acculturation. And this is probably a less, more standard American acculturation here.
And as you can see, as you go more toward the traditional culture, you see much bigger disparity in the perceived benefits. This is the perceived benefits for the male subjects versus the female subjects. Female subjects here perceived lower benefits for this third party who would have sex and the male subjects perceive more benefits.
So you have a very definite gender difference here in the perception of the benefits of having sex, less so here, and no significant difference among the Anglo teenagers. So the perception is in the eye of the beholder. Remember, they're all looking at the same stories, but they're seeing different things.
So same results for the characters. Juan is perceived as receiving more benefits than Sonya. And of course, they're looking at the same story. But they perceive a different gist.
Now what about if different age groups look at the same story? Do they perceive the same level of risk? Remember, we talk about benefits on the one hand and risks on the other determining their decision.
Well, older teenagers actually perceive more potential risk in this situation. And all we said is do you see the risks as none, low, medium, or high. And on average, the older subjects were perceiving significantly more risk. Not hugely more as a group, but receiving significantly more risk than the younger ones.
Now that kind of flies in the face of some of our stereotypes. Right? Because it's the older teenagers who we see are taking more risks. And when we look at them under controlled conditions in hypothetical risks and lab situations, actually, the older teenagers take fewer risks in the lab tests, too, than the younger ones.
So as you increase in age, you actually steadily decrease your risk preference. And this is with lots of other things controlled. You know, when we look at risk taking the real world, everything's going on. So there's lots of other factors at work.
So we can look at the real world. We can look at the lab. And we compare them. But under the controlled conditions, where everybody has the same payoffs and the same risk opportunities, risk preference goes steadily down with age.
And we've replicated that in study after study. And other people have replicated it, too, in independent laboratories who didn't necessarily agree to begin with, right? So that's a pretty stable result.
So if your risk preference is steadily going down, right, and your risk perception, meaning you see danger is going up, why do they take more risks? Why do teenagers take more risks? Well, they have more opportunity.
So compare the older adolescent to the younger adolescent, the younger adolescent to the elementary schooler, who has more opportunity to get into trouble? It's the older adolescent or teenager. So my thesis is, if you gave an eight-year-old a beer in a car--
--it would be worse than if you give an 18-year-old a beer in a car. But they can't get the beer. And they can't get the car. At least, let's hope so.
It's harder for them to get the beer and the car. And that's what we call risk opportunity. And that, too, has been studied.
So some kids just have more opportunities. There's more people coming up to them in certain areas. When you're around risky peers, this accounts for the peer effect.
When people say, hey, why don't you get in the car, you know? We're going to the corner. And yeah, you've had a few drinks, the driver, but come on. We're going to have fun.
So if I never ask you that question, if I never invite you to come in the car, you don't have that risk opportunity. So you have to now resist that risk opportunity. But maybe if you don't have it available, you don't get into trouble. So risk opportunity is a very important environmental factor that interacts with what we call the endogenous or individual factors about people. OK. So supervising or monitoring is the key, especially for younger teens.
Now does the brain change? I got to show you pictures of brains. Today is not complete unless you've seen a picture of a brain. So we've got to show you [INAUDIBLE] in the brain. OK?
So this is sort of a classic picture. This is from five years of age all the way, and this is, like, different age points, to 20. And as you can see here, this is the gray matter is the more hot color stuff. And the cool color is not.
And as you can see, just the morphology of the brain changes as you get older. And during this critical period of adolescence, what is happening to the brain? You are losing gray matter. Remember, that's the stuff you think with.
So you're losing gray matter. But what's happening is it's pruning. It's becoming more efficient. I would say you're sort of on the express route to gist. That's what I would say.
So in other words, instead of deliberating and thinking all the possibilities and going in 60 directions, as a result of experience, that experience molds the brain. And you prune away the things you're not using, the things you don't need. You myelinate more, so you get faster to that bottom line.
And this all happens, a lot of this brain development is going on during this period of adolescence. You know? They're not done yet.
And then, this is one of my favorite examples. This is some work by Abby Baird. So remember the argument that we're making from the point of view of fuzzy trace theory, think of the Russian roulette example, is that teenagers are taking risks because they're thinking of the risks and the benefits. They're weighing them, right?
Whereas when you get older, and you're mature, which most of us get to most of the time, then you're just thinking of that bottom line, the categorical just don't go there. So there was this study. And I love this study.
And Abby Baird asked teenagers the following kinds of questions. These are good idea questions and bad ideas. Good ideas were things like the following.
Is it a good idea to eat a salad for lunch? Yes, of course. Good, class. You can yell out the answers to this one. OK.
Is it a good idea to set your hair on fire. No. That would be a bad idea. Very good, class. Is it a good idea to swim with sharks?
VALERIE REYNA: No. That would be a bad idea. OK. Now good. I was a little worried there for a minute.
So good ideas, this is reaction time. How long did it take you to say yes, it's a good idea to the good idea and no, not a good idea to the bad ideas? OK? Now the good news is that the teenagers did say that it was a bad idea to swim with sharks. That's the good news.
The bad news is they thought about it. Right? So they took significantly longer to respond no, it's not a good idea, when asked is it a good idea to set your hair on fire. Really, that was the question.
They thought about it. OK? Now, OK, so far, so good. At least they said no.
But the question then is, what parts of the brain are lighting up. So we have this very intuitive theory, fuzzy trace theory, that says that how adults are thinking is are you kidding? Set your hair on fire? Right?
Whereas the adolescents are doing what? Reasoning logically, computing the answer, calculating the expected value and expected utility. Right? So what parts of the brain ought to light up?
Well, if you look at this is the right fusiform. And this is the differential activation. This is for adults versus adolescents.
So adults are lighting up the right fusiform. What does the right fusiform do? That's the face perception area. That's the images. That's big teeth from Jaws coming for you. OK?
That's what's lighting up in your head if you're an adult. But not if you're an adolescent. The left fusiform, yeah, you see some pictures in there in the head lighting up for the adult. But not for the adolescents, right?
Here you got the insula. Now the insula is a part of the brain that has to do with things like your gut response, disgust, things like that. So what are the adults are going? Eh. Right?
See that face? Eh. Right? They're doing that. So that's what their brain is doing. Ooh, set your hair on fire? Ooh, that might hurt.
And that's what the adults are doing. The adolescents aren't doing that. But what about this part of the brain? This is the right dorsolateral prefrontal cortex. And what does the right dorsolateral prefrontal cortex do?
Reason, exactly. That is the seat of reason. The adolescents are reasoning about swimming with sharks, yes. They're weighing the costs and the benefits of Russian roulette.
So that's the part of the brain that the adolescents are lighting up, but not the adults. So you see that fits with this very counterintuitive theory that their reasoning, well, this is effortful, executive function. This is reasoning. They're computing the answer to whether to take a risk or not, whether to engage in dangerous behavior.
So laboratory and public health evidence converge on this notion, that adolescents, in fact, are more logical than adults. They quantitatively trade off the risks and the benefits. Russian roulette is an example of that.
Adults avoid risk because of an increase in what we call gist processing, this bottom line, qualitative, just don't go there kind of thinking. They process risk information qualitatively, often categorically. So even if the benefits are bigger than the risks, they still don't take the risk. Because it's not a math problem anymore.
And adolescents who think like adults are more likely to avoid risk. All right. So we have more data. I know, we're-- yes.
So I'll give you a little bit more data just to show you that we're not just making this up. This is a study of almost 600 kids that was published in Psychological Science. And, again, you can get this on the website for the details.
This turned out to be a multicenter trial in three states because I moved. It wasn't started out that way. But it became a multicenter trial in three states.
This is a majority-minority population, like you find among real adolescents in America today. The average age at the beginning of the study was 15 and a half. Fewer than half were sexually active. A little bit more than half were female.
And we asked a number of different questions. Now the verbatim questions, sorry about the technical terminology. But the verbatim questions are just this logical, detailed, exact way of thinking. And the gist questions are the bottom line, qualitative, categorical, adult way of thinking. And we'll get to concrete examples in a minute. OK?
And we predicted, in fact, that if you ask about risk in different ways, you would recruit different ways of thinking about it. So within the same person, we have the potential to be a teenager or an adult in our mind. We have the potential to think about things in a much more quantitative, mathematical way on the one hand. And on the other hand, there's this other little voice, which is the mature voice, that thinks about things in a fundamentally different way.
And these are inside our own mind, even as adults. And there are different things we can do to draw out one kind of thinking, that's what we did in our intervention, versus drawing out the other kind of thinking. So one of these ways is to ask about risk in different ways.
So we asked things like specific risks. Like, how likely is it that you will have HIV/AIDS by age 25? These are teenagers. And they said, you know, very unlikely, likely, so on, and so-- up to very likely.
So this is where-- Likert scales and so on, strongly disagree to strongly agree. And this is a very reliable scale. So they answered a series of questions like this. Very specific risks, how likely are you to develop a sexually transmitted infection,and so on and so forth. Specific risk, verbatim scale.
There is 0 to 100 probability that they personally would develop a sexually transmitted disease, on a 0 to 100, very precise kind of estimate. And they gave us an estimate. And they could tick off any number between 0 to 100. What are the chances that you will have one.
Now here's a gist scale. Now this is exactly the sort of thinking that other interventions to help kids make healthy choices, they're trying to get them to not think like this. We're trying to get them to think like this. This is very simplistic thinking.
It says it only takes once to get HIV/AIDS. Right? Anybody's parents ever say that? It only takes once to get pregnant. Right?
Now that's not very mathematical, is it, right? The economists would go, ooh, that's not very-- no. So actually, I had arguments with people in the literature, very esteemed people I respect, saying this is primitive thinking. This is when they have misconceptions, they say things like this.
If they agree with this, however, if you agree with if you keep having unprotected sex, your risks will add up and you will get pregnant. Even low risks add up to 100% if you keep doing it, a very non-quantitative, categorical way of thinking about risk, and so on. This is a categorical thinking about risk.
Agreeing with this, we would argue, protects you, especially if you're 15 and a half. Remember, these are high school students, not older students. So thinking about these things categorical is a more gist way to thinking.
Here's what we call a gist principle scale. The score on this is just how many of these do you tick off. When you think about you having sex, which of the following apply?
Is it a good idea to avoid risk? You know, better to do-- so these are very simple. Better to be safe than sorry. Better to focus on school than have sex, and so on. And it's just how many of these you agree with.
And then, finally, global risks. Overall for you, which of the following best describes the risk of having sex? Is it low, medium, or high?
So here's just a summary of our scales. There's three gist ways of thinking that we think will be protective and there's two kinds of verbatim ways of thinking that we think will elicit actual memories of behaviors about risk. And they should, in fact, produce opposite results in the same people. Now I will show you that.
So risk taking measures we looked at, sexual behavior, like if the teenager ever had sex, yes or no. And behavioral intentions, which is do you think you will have sex before you turn 20, when you are finished with high school, by the time you're finished with high school, and so on. These behavioral intention scales tend to predict behavior. But basically, this is just two ways to measure the same thing, to make sure we're on to-- if everything goes in the same direction, we can believe the data. That's pretty much it.
So if adolescents are in transition and some of them are thinking cost-benefit analysis, that would be verbatim, some of them are thinking our gist-based processing, this all or none, what do we expect? Well, if gist processing increases with maturity and experience, that all or none thinking, and if risk avoidance is a function of gist processing, then risk preference should decline despite rewards or benefits.
And as you mature, you should avoid trading off risks and rewards when the outcome is catastrophic. Again, think Russian roulette. If you understood that example, you're old enough to understand this slide.
So is that what you see? Then you should be less risk taking, for people who are more likely to think of risks in this all or none, just don't go there kind of way. And in fact, that's what we found. So over here are the three gist measures. Let's start with those.
This is the correlation between whether you agree with this categorical risk scale and your behavior or your behavioral intentions. And as you can see, the more you agreed with the categorical risk perception, the more you perceived, yes, the [INAUDIBLE] risks are going to add up to 100% if you keep doing it. Yes. OK?
The more you agreed with that, the less likely you were to have sex, or to have behavioral intentions to have sex. Gist principles, the more of those you ticked off, this correlated minus 0.6 with behavior and behavioral intentions. The moral of those, you said avoid risks. Better to be safe than sorry.
I have a responsibility not to put my partner at risk. I have a responsibility to my parents. Yes, that matters to young people. They ticked off simple principles like that that inversely-- the more of those they ticked off, the less likely they were to have had sex and the lower the behavioral intentions these young people had to have sex.
AUDIENCE: Protected or unprotected?
VALERIE REYNA: Well, we looked at that, too. And we broke that down. But this is just the overall measure I'm reporting that. That, too, exactly.
And then global risk, we looked at prophylaxis, unprotected, lots of different outcomes. But yes, exactly. And then global risk perception, the higher you perceive the risk in this global, totally kind of undifferentiated question, the more you said yes, if I were to have sex, it would be a high risk. The less likely you were to have sex. It was protected.
Now remember, we're asking the same people these questions. The very people who said if I were to have sex, it would be high risk. When we asked them specifically, on a 0 to 100 scale, what's the probability that you would get the specific outcome of sexually transmitted disease, then they thought, well, what was I doing last Saturday night.
And so very specific questions queued verbatim memories of actual experience. And what happens then? You say, OK, I'm not at risk. You know? Right?
And therefore, you're having more sex. Because you don't perceive any risk there. But what did you do last Saturday night?
Well, actually, there's a positive correlation there between the quantitative perception of risk. So if you ask the same people the question a slightly different way, they thought of what they were doing, their behaviors. And they said, I am at high risk.
So if you ask it one way, they'll say, you know, I do have a high risk, now that you mention it. I am going to get HIV by the time I'm 25 if I keep doing what I did last Saturday night. But then you ask the global question, they say they're at low risk, the same people.
So inside the head in teenagers, you have this computer, whirring along, calculating the risk-benefit analysis. And right next to the computer is this incipient, mature adult, that really is capable of thinking in a more gist-based way where we just have to draw them out. That would be my interpretation of these data. So that's opposite results in the same people.
And this just summarizes. Risk takers rated themselves as high for specific risks. Perceived global risk is low. They realized they were at risk when they were given specific cues to their own behavior. But they denied they were at risk when the cues [INAUDIBLE] global attitudes.
And risk avoiders rated themselves as low for specific risks. They were home on Saturday night. OK? But they perceived their global risk, if they were to have sex, as high. And older were, in fact, more likely to endorse the simple gist.
So this is not just that younger kids are simple-minded. This doesn't account for this result. In fact, it's the opposite. You get more bottom line gist-based in your thinking as you got older, not for the younger kids.
So you have your risk taker one, your reasoned, deliberate, trading off of risk, that's when the right dorsolateral prefrontal cortex is lighting up and you're thinking about the risk, and you're reasoning logically, and you take the risk because the benefits outweigh the risk. Right? That's one kind of risk taker.
You have your second risk taker, who I have not talked about, which is your impulsive risk taker. This happens, too. Right? There are young people who take risks not because they've calculated it, but because they're in the heat of the moment. There's that lack of impulse control. That's also true.
But that's a different causation than the deliberator. And when I look to the literature through these many, many studies, it turned out that every study, just about, that had looked at risks and benefits perceived risk and benefits, that the risk taking behavior was actually predicted by perceived risk and perceived benefits. So we think of adolescents as just being impulsive.
But in fact, what the data show is that you believe what they say about what they perceive the risks and benefits, their behavior follows logically from that. So it's not as though, oh, my God, I took a risk. And I regret it the minute I took it.
That's some of it. I'm acknowledging that that's some of it. And other theories focus on that. But that's only a part of the story. And it may be a smaller part of the story than we think. And then, of course, we have our risk avoiders who rely on gist.
So summary, right, it's about time. It's a hot room. And thank you.
So many studies show that adolescents' perception of risks and benefits predict their behavior and their intentions. They take risks despite overestimating the risks, because the benefits outweigh the risks. Some of this may sound familiar.
You have three types of risk takers. You have your risky deliberator, but also your risky reactor who's responding to emotion and impulse. Both of these things are routes to risk taking, but they're different.
And the way you deal with someone who has these different motivations would be different. The educational program should be different for these two. And then you have you're gist-based risk avoider who engages in less analysis, but less risk taking as a result.
Now, this is an example of, I think, a story that highlights this other kind of risk taking. And I want to acknowledge this. Again, so many theories emphasize this. And I want to indicate that I agree with those theories.
So does anyone know who this is?
AUDIENCE: Rudy Galindo.
VALERIE REYNA: Very good, Rudy Galindo, exactly, bronze medalist, Kristi Yamaguchi's former partner, and so on. He went to the heights in figure skating, very original. I liked him a lot, still do.
But he had a brother who died of AIDS and a coach who died of AIDS. And after that, and he went through this with them, and this was this coach who brought him up from LA and really was a mentor to him, very close to him. And they both died of AIDS.
And what did Rudy Galindo do? He had unprotected sex. And he became HIV positive after that.
So, again, it's not that he didn't perceive the risks. Of course he perceived the risk. His brother died in his arms. He lost people very close to him.
So he knew, intellectually, what the risks were. But he still took that risk. And I think about this kind of preventable tragedy when I think about the work that we do. So we have to keep that in mind.
He's still living with HIV. And he's still got a website and he does things and so on. So he's hanging in there. But we want to be able to prevent these kinds of outcomes.
So conclusion-- I think we could probably read through those pretty quickly. Obviously, the conventional wisdom is that young people just don't perceive their risk. They think that they're immortal.
Study after study has shown that that's false. Everybody has a little bit of an optimistic bias about their risks, by the way. If you compare yourself to your neighbor, you think you're at lower risk.
But both of you are overestimating, often, your objective risk. You're just overestimating it a little less for yourself than for your neighbor. And this developmental difference, in fact, doesn't seem to be there.
Young people know that they're at higher risk than their parents are because they're doing things their parents aren't. And so they realize that they're at higher risk, which is very logical. So the things we've assumed in the past are not meant to be true.
Obviously, the object of interventions is to enhance the accuracy of risk perceptions. That can do what? If kids are overestimating the risks, what happens if we tell them what the real risk is?
Right? And you saw the one example I gave of how the behavior tracks the risk perception. So this presents a dilemma for public health people. Right? Because they're thinking accurate, accurate. If we can just tell them that it's one in whatever, in fact, that would probably promote risk taking.
I'm not telling you what to do about that. I'm just telling you what the implications are. It's kind of scary.
Experience, obviously, is not a good teacher. This notion that if you drink that you learn to drink. One of the reviewers of one of our papers said, well, in Europe, we learn to drink. And I don't know why I'm having a Russian accent. But vodka, give me vodka and I will learn to drink the vodka.
So it turns out that's not true. Guess who has the most binge drinking teenagers and adolescents? Countries where they drink more, right? So this is not the case. In fact, there's a dose response relationship. The more exposure to alcohol, the more bad outcomes you get. That's just the way it is, unfortunately.
So this is about drinking age and risky deliberators and identifying factors that move adolescents away from considering degree of risk and amount of reward, letting them think less computationally and, we think, in a more gist-based manner. And of course, there are implications that have to do with monitoring. When we know that young people are calculating the risks and they're right about the objective risk, study after study has shown that, on a whole range of risk taking behaviors, pregnancy, crime, lots of outcomes, that amount of supervised time correlates with negative consequences. So if you have more supervised time, you have fewer negative consequences.
And we have replicated that result. But many, many people have found that. And that's particularly true for younger adolescents. Simply-- and I don't mean knowing the names of the friends, I mean, being able to see the friends in front of you. So if you're playing basketball, it's hard to be having unprotected sex during the game.
All right? You learned it here first. So that's the sort of thing I'm talking about. I think I'll end there and take your questions. Thank you.
We've received your request
You will be notified by email when the transcript and captions are available. The process may take up to 5 business days. Please contact firstname.lastname@example.org if you have any questions about this request.
Valerie Reyna describes the developmental differences in the way adolescents make decisions and reviews her research regarding why adolescents perceive risks and benefits and yet take more risks.
Reyna is a professor of human development and psychology in the College of Human Ecology at Cornell.