Dr. Dan Romer: Back in the early 2000s, research was starting to come out, some of it done by the National Institute of Health, very reliable, trustworthy source, suggesting, based on brain scans, looking at young people from like ages 8 to 30 and scanning their brains across this age period, what became clear was that there were relatively dramatic changes in the structure of the brain over that time period. It's not that the brain got any bigger or smaller, it's that certain structural features of the brain were changing, the most interesting of which was a decline in what appeared to be the gray matter, what seemed to be a pruning process. So, we actually lose gray matter as we age through adolescence, and an increase in white matter. White matter is the myelinated fibers that surround a lot of neurons and enable fast communication between different parts of the brain.

It looked as though if you just took that as face value, it wouldn't be clear what it meant, and so when it was first presented, it was sort of suggested that, well, this is a sign that, during the adolescent period, the brain is learning and assimilating information, and so we lose, we prune the parts of the neuronal structure that we don't need anymore, we keep the ones we do, we increase connections between them, and the brain myelinates to enable faster connection.

That was the initial interpretation, but then, sort of in the middle of the 2000s, some neuroscientists and developmentalists came to the conclusion that this could be an explanation for why adolescents engage in a lot of risk-taking, and of course, risk-taking in the negative sense: using drugs, unprotected sex, driving in a way that's not safe. Things like that, things that cause injury and harm.

There was no clear reason for that. A lot of neuroscientists started doing research, looking at brain differences, activation differences in relation to what might be called risk-taking behavior, at least, in the laboratory, and came up with studies that suggested there may be some validity to that. So, the teen brain suddenly took on a new meaning in the sense that it wasn't fully developed, and this myelination process, in particular, didn't look as though it was complete until mid-20s or so.

It became popular to say the adolescent brain was undeveloped, and by implication, unable to control its behavior, the behavior that's driven by the subcortical part of the brain, the part that creates emotions and drives for behavior: sex, and drugs, things like that. That was soon to be fully developed by the time we got to, at least, mid-adolescence, whereas the upper part of the brain was still developing and we continue so until at least the mid-20s.

Then, we had magazine covers talking about. It must have been several TIME Magazine, National Geographic, all of these popular culture magazines picked up on it, and it became part of the culture that the adolescent brain is not fully developed until 25 became the magic year, and therefore, that's why adolescents do all the dumb things they do, and then there's the implication that we shouldn't let them loose, basically, because you got to watch them, keep them home down on the farm, that kind of attitude. Because, if we let them out, they're going to do all these dumb things because they can't control themselves, and we just have to wait until their brains are fully matured before they can become fully functioning people.

That's sort of the backdrop where we come in, and we're not the first to sort of question that, but I think we put together the evidence that suggests that interpretation may be a bit of an overstatement and actually be driven more by stereotypes than anything else.

Jesse Lawler: Funny, because when we think about the "bad things" that teens do: sex, drugs, and bad driving tend to be, I guess, the big three, and in an ancestral environment, the vast majority of drugs weren't available. Having teenage sex when you were a cave person probably wasn't a terribly maladaptive thing to do. Obviously, bad driving wasn't an option you had. I mean, you could chase the woolly mammoth faster, but that was about it. It's interesting that none of these things that are the big red flags in modern society would have really been all that bad, I guess, in the environment that our brain's developed for.

Dr. Dan Romer: That's one of the arguments that this is evolutionarily, it was adaptive because having sex at age 15 wasn't such a bad thing when you lived in a cave, and maybe there weren't as many drugs around, although that's probably debatable. I'm sure there were all kinds of hazards that would be present for an adolescent, even evolutionarily speaking.

Jesse: There are certainly ways of getting yourself in trouble. I guess, since this is a human universal that teenagers are risk-taking, intuitively, there must be some sort of adaptive advantage to it or it would have been smoothed out a long time ago.

Dr. Dan Romer: Yeah, for sure, and some of the people who were pushing this point of view started to soften their way of talking about it and saying, "Well, it's probably evolutionary-adaptive, but it still creates a problem especially in the modern context." So, I think the people who proposed this model kind of started to soften. But, you still had some of these folks still talk about teenage drivers, you put them together in a car, next thing you know, they're going to have an accident. That kind of scare is still talked about. Adolescents can't control themselves when they're together with each other. Here they are with their unformed brains, and then you put them together with each other, and next thing you know, they're going to be even more out-of-control.

There was always another side of it that would accentuate the downside of risk-taking. Of course, the whole point of talking about risk-taking was always in the negative sense, as you said. There's a lot of risk-taking people do that, if you don't do it, you're never going to get out of your cocoon, become an adult.

Jesse: Yeah, it seems like that's part of the defining of one's sense of self needs to be taking the risk of pissing off your parents.

Dr. Dan Romer: Yeah. I mean, if you can't start doing things that your parents may not always agree with, that may not be the most adaptive way to transition to adulthood. You can't live with your parents and expect them to run your life for the rest of your life.

Jesse: Tell me about your alternate model, your alternate hypothesis for brain development. You're looking at the same data but drawing pretty different conclusions. Tell me how that came about.

Dr. Dan Romer: One of the things we did was we looked at what exactly are adolescent risk-taking patterns like. If their model is true, adolescents should be more likely to take risks than children, for example, because they claim this is a heightened period of risk even worse than for children. So, we looked at developmental patterns in risk-taking, and there's some kinds of risk where, if you tell people exactly what the risks are -- so, for example, you can get $10 for sure, or I can give you a chance to get $20 with a coin toss. So, you either get the $10 for sure or you can get $20 50% of the time.

That kind of task is very common. Adults will take the 10 bucks. They don't want to take the risk. Children are much more likely to take the risk, and adolescents are right in the middle. So, they're not particularly more risky than children, and they're only slightly more risky than adults. They're not particularly strong in effect. So, you get pretty much of a linear, or what we might say a monotonic decline in that kind of risk-taking.

Jesse: Are we talking even odds in this, so it's basically the expected value in both cases the same?

Dr. Dan Romer: Yeah, it's just you have to wait for it. So, being willing to wait is a sign that you're not impulsive, and that kind of response is correlated with other behaviors like using drugs, and doing risky things. So, it's a marker of risk-taking that has some real-world application, but it doesn't fit their model.

Then, there's another kind of risk-taking in which you don't know what the odds are of whether it will be good for you or bad for you. That kind of risk-taking does seem to be higher in adolescents. So, if you don't know, specifically, how this is going to turn out for you, which is different from these tasks where we know the probabilities, we know their length of time, we know all the parameters, but if we don't know, that's where adolescents are more likely to take a chance.

This is correlated with a phenomenon called "sensation seeking". So, adolescents experiencing novel things: "Should I try that drug or not? Yeah, give it a shot. What are the chances it's going to be bad for me? Not that high. What are the chances that I'm going to get HIV from having unprotected sex with my girlfriend? Not that high. Do I know? No." So, those kinds of behaviors are more likely to occur with adolescents because they're novel, they don't have experience, and they don't know, for sure, how it will turn out for them, and so they're having optimistic bias.

Jesse: How do you test that in the lab? Because, it seems like it's harder to test an unknown downside because you can't do anything actually dangerous with kids in an experiment. How are you mocking up that as a real choice?

Dr. Dan Romer: Well, they do experiments in which they give people the option of trying the risky choice without knowing how it will turn out. So, it could turn out really good or it could turn out really bad. You just don't know. When you do that, adults will hardly ever take that choice. They always take, "I'll take something that I know for sure better." Adolescents are willing to take the risk if there's a good chance of a payoff, and in fact, it turns out that they're actually more irrational in that regard because, by not being so risk-averse, they actually do better. They will lose out some of the time, but they will also, on average, they do okay.

This is in laboratory settings, and we also know, just from drug use, things like that, adolescents are more likely to try a drug that they don't fully know what it will do to them. They might know what it does to people on average or to their parents, but they really have no idea what it will do for them. That's more likely in an adolescent than in an 8-year-old. An 8-year-old is going to say, "I don't want to take that," but a 15-year-old will say, "Yeah, I'll try a little marijuana or I'll try smoking a cigarette. Why not? What's going to happen to me? My uncle does it." So, they're exploring, they're willing to explore.

We argue, based on that kind of evidence that what neuroscientists, when they looked at the brains of adolescents, they found they were likely to evaluate these unknown kinds of risks more highly and the brains would reflect that, and they interpreted that as evidence for what was called an imbalance. The prefrontal cortex is not in balance with the subcortical part. We argue that that's just exploring, and it isn't necessarily going to be harmful.

Now, there is a form of impulsivity that is really insensitive to risk, and we know that happens because people get addicted to drugs, and they can't stop themselves, and that can start in adolescence. Those kids who have that problem tend to be more impulsive, and you can see that even earlier than adolescence. It's the characteristic that you start seeing as early as ages 4 and 5 especially in conditions like ADHD. ADHD is a very commonly understood, although not fully understood brain development condition that has impulsive characteristics and puts a kid at risk for later having behavior problems, drug problems, things like that. That's a condition that I think is just prior to adolescence. All adolescence does is enable the person to do riskier things, especially if they're not being watched or not given stimulants which can sometimes help to reduce that condition.

We argue that, yeah, there are kids who do that, and they may be more likely to be at risk when they're adolescents because the things they can do are riskier, and there may actually be a slight increase in their impulsivity.

Jesse: Are the results that they're looking at there, is it drug use or really problematic drug use, drug abuse where it's having a noticeably negative impact on the kid's life?

Dr. Dan Romer: Kids that are exploring, they will use drugs at quite a relatively high rate compared to those who are not. This is an individual difference now, which tends to be higher in adolescents, but they don't tend to get addicted. The kids that are more likely to get addicted are the ones who have impulse control disorders, because then, when they have a bad experience with a drug due to some effect that the drug has on them, not the least of which is failing to do well in school and then becoming unable to stop, which is in the property of the drug - it creates this dependence.

The kids, who are sensation-seeking, don't tend to have that problem as much, and that's because they actually have pretty good cognitive control. So, that's another piece of evidence that goes against this imbalance theory, because the kids that are doing the exploring are actually not so bad off in terms of their ability to control themselves. They're actually better off than the kids who have the impulse control disorders. The ones with those disorders tend to have poor cognitive control, and so if they end up using a drug, they're going to have a harder time quitting, and this is true for tobacco or any other addictive substance.

The imbalance, really, is just in this subset of kids, some of whom have ADHD, others have other problems, or they may be growing up in an environment that doesn't train self-control, often, for good reason. Like, if you're growing up in poverty, you may not be so wise to be controlled, and it may be a completely adaptive kind of behavior in those environments. So, it's not just a personality thing unique to the individual. It's something that can be cultivated due to the environment one's in. So, yeah there's all kinds of factors that can lead to that, not all of which are necessarily so maladaptive.

We basically deconstructed what it means to be a risk-taker, and found that kind that goes along with being an explorer doesn't have a deficit in cognitive control. The kind that does have a problem, if they're exploring, it's because they're a minority of adolescents, and they do have cognitive control problems, but they're really not typical, and there may not be anything particularly maladaptive other than, for the explorers, that they might try something that could be harmful to them in the immediate case. But, on average, they learn from it and don't have long-term problems. They don't have gambling addictions, and drug addictions, they learn how to drive and control themselves eventually.

Jesse: Right. It's not that you shouldn't burn your hand on the hot stove. It's that you shouldn't repetitively burn your hand on the hot stove.

Dr. Dan Romer: Yeah, and the impulsive kids will have a problem more so with doing things that are unhealthy for them. We also looked at brain data. There weren't that many, but there were several studies that showed that if you looked at the brain's development, it wasn't necessarily -- you know, this myelination thing, there was one study that found that the more myelinated the brains of adolescents were, the more risk-taking they were, not less. Other studies found no relation.

There's also a lot of research that's come out that shows that the brain networks that control behavior that are in the prefrontal cortex and connect with other regions, what happens during adolescence is they actually become more self-contained. It's not so critical for them to branch out and reach other parts of the brain. It's more important for them to become self-contained, because that's how the brain can concentrate and control itself is if these cognitive control networks sort of look inward. So, working memory, for example, is a very important component of cognitive control. The networks that control that become what's called segregated as the adolescent brain develops. They don't branch out. They become, actually, somewhat, segregated.

But then, later in adulthood, these networks become less segregated, and the interpretation, we don't fully understand it, but one interpretation is that that's a necessity for the brain to connect with all the things it's learned. So, it's no longer able to control itself so well, so it has to rely on all the experience it's gained. As we move into adulthood and later, this is the part of our model that we attribute to what's called wisdom, and wisdom develops over time with experience. If you explore when you're an adolescent, you're going to gain wisdom that will do you well later when your cognitive control capacity is actually declined.

Once you get into your late 20s and early 30s, you start losing some of these cognitive abilities: working memory starts to decline, short-term memory starts to decline. These are all standard biological processes having to do in part with dopamine transmission, which peaks in adolescence and helps with the brain's cognitive control system. It also excites the motivational system, but it doesn't seem to be doing that any more than it enables the control system.

But, as we age, we have to depend on our wisdom. We can't depend on the ability to think and concentrate as well as we could when we were adolescents. So, that's the other part of our model is that there's a growth in wisdom, and we hypothesized that this wisdom is the component where there's really an imbalance. So, adolescents just don't have a lot of wisdom. They're novices. Everything they do, from sex, to drugs, to driving, to doing well in school, and trying out for competitions, they're going into sports, all the things they're going to try, they don't really know what's going to happen and they're really inexperienced, then they're going to find out, and that's when we find out if we're good at sports, or if we're good at debating, or if we can do math better than other people.

All the things we find out about ourselves happens in adolescence and early adulthood when we do this exploring, and that's where we have a deficit is in the experience component. But, later we gain experience and we don't need that ability anymore to think and concentrate the way we were able to do when we were adolescents. So, that's how we kind of put this all together into what we call this lifespan wisdom model.

Jesse: It feels like you're kind of standing up for the brain and saying, "Hey, the adolescence isn't this pockmark on our mental development where things go south for a while. There's just a couple of different signals there that when you segregate them into separate piles, the well-known risk versus the exploratory risk, everything sort of makes sense in a progressive way.

Dr. Dan Romer: I think so. It never made sense to us to argue that adolescents had a developmental deficit, even considering all the arguments about evolution and how our brains are now evolutionarily maybe maladaptive, which is one implication you would get because of modern society being so prone to having risks that weren't present back in our earlier evolutionary past. Aside from that part, it didn't seem to make sense that we would all be here and get through that -- you know, we get through adolescence, nearly all of us, pretty well.

Jesse: I feel like adolescence gets a bad rap. It's like I remember being 9 years old, 10 years old, and already was hearing, "Oh, it's so hard to be a teenager. Teenagers are so crazy." It's like they get this idea kind of sunk into their head before they're even there that it's going to be this really rough time, and I feel like if it wasn't advertised that way leading into it, it might not actually be quite so bad.

Dr. Dan Romer: Yeah, I agree. I think we sort of stigmatize adolescence, and stereotype them, and I think that's because adults have a hard time with them. Parents don't enjoy, always, the adolescent period, because the kids want to do things that the parents have had experience with that they don't think is smart, or things that the parents do.

Jesse: Yeah, it's okay for me but not for you.

Dr. Dan Romer: Of course, the parent knows how to handle it, has had experience, and can possibly control that kind of behavior better or knows what to avoid. The adolescent's still learning all that, and it's difficult to raise an adolescent. I think adults are prone to this stereotype: the adolescent is a problem, they do things that are risky, they do things that we wouldn't do, they don't listen to us, they do all these things that are upsetting to parents.

Jesse: You flip that the other way, and you had a kid that never made any mistakes and didn't do anything wrong, then they would never get the experience of learning how to deal with a mistake or pull themselves up and dust themselves off. Mistakes are part of the process.

Dr. Dan Romer: Yeah, I think that's true, and we sort of forget about that when we talk about it as a brain issue. This is a deficit in the brain, and then we have to protect adolescents, not allow them to do things, prevent them from driving, going out, and having sex. This perspective sort of says that there's probably some sort of ideal level of experimentation that parents should allow. This is not a new idea either. It's been around for a while called authoritative parenting. There are parents who won't let their kids do anything, there are parents that will let their kids do anything. Neither of those are good.

Jesse: Goldilocks parenting.

Dr. Dan Romer: Yeah. The best kind of parenting is the kind that says, "Okay, you can go out, but I want you home at a certain time, and who are you going with?" you know, putting some parameters on what they do so that the chances of them having a bad experience are minimized.

A good example of how this works in a policy level is graduated driver licensing, which came into vogue 10 or 15 years ago, and it's now shown to delay kids going on the road on their own, and spending time with their parents in the car before they can get a full license, and that helps. It doesn't solve all the problems, but it does help. It does reduce some accidents, and it's a form of authoritative parenting. You don't just let the kid have the keys and say, "Go make mistakes." You kind of gradually help them get to the point where they can drive on their own.

The same argument would be true, I think, for things like alcohol, which is so commonly used in our culture. It becomes a problem when kids go to college because they're not 21 and half the campus is 21, and so it puts kids in a funny situation where they have to drink undercover, and then that creates all kinds of problems. I think this is not a popular thing to say, but I think it would make sense to allow kids, under the supervision of their parents or some other adult, to drink alcohol, as long as they didn't overdo it, and you can argue that some of the things that happened that are bad with alcohol happen because the kids aren't being supervised and they're hiding.

Jesse: When you push something underground -- I mean, we kind of saw that with prohibition of alcohol for adults. It's like it created this whole criminal enterprise out of it, and it's not quite the same when you're just having to hide alcohol use from your parents, but it kind of does force all this dishonesty and stuff like that wouldn't need to take place if you could do it in the light of day as a teenager to some reasonable degree.

Dr. Dan Romer: Yeah, I mean, I grew up in New York, and at the time when I was growing up, 18 was the age, and there were kids who would go out on a Friday night and drink at 18 or even younger, and I guess what happened -- well, we know what happened - it was in the '80s. The Federal Government, under pressure, for good reasons -- I think Mothers Against Drunk Driving were upset that a lot of young people were driving drunk, and adults as well, and there were no harsh penalties for that, so they started encouraging people to have a designated driver, and they also started encouraging government to make 21 the age that you could start drinking, and whether that has been a wise decision or not, I think, is still up in the air, although many people think that that's been settled, that that has saved lives.

On the other hand, the Canadians, their legal drinking age varies by province, but it's 18 or 19, that's the highest, and if you look at what happened with them, they also got influenced by Mothers Against Drunk Driving. They also had a decline in alcohol fatalities just like we did, but they didn't put in the 21 law. So, I'm a little skeptical about the 21 law because they're hiding and they have to get tagged up before they go out and all the things that happen as a result of that.

Jesse: This is probably one of the more salient examples where a neuroscience model really has direct policy-level ramifications of what laws get written, what the social norms are, how society operates. What are some other areas where you see if people accept your lifespan wisdom model that that has direct follow-on consequences for social policy?

Dr. Dan Romer: Another thing that we've done some research on is drug testing in high schools. This is another effort on the part of adults. It's all well-intentioned to prevent adolescents from even trying a drug. So, the thought is if we randomly drug test, and it's done randomly only because the Supreme Court has ruled you can only do that to a limited degree, and because you can't test everybody without suspicion, that's such a gross violation of privacy. At least, you could argue it is.

The Supreme Court has agreed that, at the high schools, any school, can test its kids on a random basis, and we sort of think that's a bad strategy because, first of all, it suggests that we don't trust the kids and that any use, at all, if you detected someone having used marijuana three months ago, and it shows up -- because marijuana does stay in your system for quite a while. Alcohol doesn't, ironically, so these drug testing systems will pick up things like marijuana. They won't pick up alcohol, which is far more serious worry, and if the kid does test positive, it could be because all he did was smoke some marijuana three or four months ago, and they're not permanent users, and they're not really at risk of any addiction. They were just exploring.

We think it's not a good strategy, and there's a research that suggests that it's an ineffective strategy, because what the kids do is they know what drugs are being tested, so they just are sure not to use those drugs during the school year. They use other drugs that won't be detected as easily, including alcohol. So, there's been evidence to the effect that other drugs go up in use, and the ones that tested may go down, but the kids don't report any less concern about those drugs, or any less interest in trying them. They just know that, if they're in a school that tests, it's not smart to use it during the school year.

It's really not a very smart strategy. The better strategy would be to, A, watch for kids who might be having problems and get them into treatment. We would advocate just doing surveys and letting kids report on their own if they're having a drug problem or any other mental health problem, and make it confidential, and if the school finds out that a student is having a problem, they can then go to them confidentially and suggest how to get help, or be on the lookout for kids who are not doing well suddenly, their performance drops, they don't show up, things like that, go after those kids. But, just the kid who's exploring doesn't seem like a very smart way to go, and it's not all that effective either.

Jesse: Has there been any pushback as you put these papers out and these ideas, any counterpoints?

Dr. Dan Romer: I will tell you that we had a hard time getting it published. It took us a couple years. We had at least five different reviewers, some of whom read it more than once because we revised it. So, advocates of the imbalanced point of view are not very open to alternative interpretations. We finally found the right journal and the right people who would appreciate what we were saying. So, it took a while.

Now, I think what will happen, since it just got published and it got some attention, they will now come after us and push back. How they will do that, I'm not sure, but yeah, I'm sure the people who advocate the imbalance model will not be convinced by our review, and they will try to poke holes in the evidence we present. It will go on, and we'll keep doing research to try to show that we're right and they're wrong.

On the other hand, maybe they'll just say, "Yeah, you're right," and move on. That's a possibility.

Jesse: Are there people that have sort of built careers around this idea of the imbalance model that they really have an important dog in the fight, I guess, as far as their own prestige levels?

Dr. Dan Romer: Absolutely. I mean, it's almost like a cottage industry. People have written books. I won't name names, but I know, in talking to some of these people, they love the attention and they say they could spend their lives talking to various groups: parents and other groups about the teenage brain, and how do you deal with the teenage brain? There's loads of self-help books about it, and the people who write those do very well. What we're saying goes a little contrary because they like to start with the imbalance model and say, "Yeah, the teenager, the brain is just not developed," and boy, oh boy, does that cause problems.

Jesse: The macro level of effects of teenagers behaving in risky ways, nobody's denying that. That's their one way or the other, and I feel like people could make a pivot and still be able to write books about how to raise your teen.

Dr. Dan Romer: Sure, and I think there's authoritative parenting, which is an idea that's been around for quite a few years is a very nice way to think about it, and you just build up from there. So, yeah these self-help people will find other ways to send their message, I'm sure.