« My Strange Beliefs | Main | Posting on Politics »

December 31, 2007


Robin, I don't think its just teens that over-estimate the benefits of a particular event occurring, I think all of us do it. Or at least Daniel Gilbert would have us believe so. It just seems to me, without even experiencing sex once, teens vastly outweigh the benefits, and once they have the experience their 'nexting apparatus' comes more in sync with the real benefits, but like the rest of us, is still way out of proportion to the real amount of happiness we are going to get out of doing X.

It's not in the realm of HIV, but speaking of drugs...

When I was 15, what worked on me was this: my father convinced me that he feared for my life (he called me said he dreamt that I had died in a car accident). He asked me nicely if I would please not drink or smoke or do drugs. I was touched. I agreed. I kept my promise. It wasn't a rational evaluation, it was a gift I could give to my father.

Don't bother with *logic* to convince kids to do something or not do something! Work with premises instead. Establish a caring connection and make your feelings known.

(My own son is 14, by the way. You bet I'm using this on him.)

But these benefits might be less important in the modern world -- teens who will move away to college, and then away again to a job, should invest less in impressing home town associates.

Actually, the circumstance of teenagers being largely sheltered from the rest of society might be having the opposite effect. Social aggregation among teenagers seems to be extremely stratified, so a teen might be willing to pay a very high price in order to secure the cooperation of his allies.

It's quite likely that the situation would improve if teenagers were allowed to mature sooner and provided with better social role models; this is what happens in traditional societies.

If you're honest but uninformed, does that make you dishonest?

Also, The NYT article wasn't entirely clear. It seemed to suggest that teens overestimate the numerical probability of adverse events occurring as a result of risk-taking activity, but it seems possible that teens underestimate risk by underestimating the negative value of these events (fits with whole teens-have-underdeveloped-frontal-lobes-which-is-where-value-is-encoded sort of Oprah Winfrey's neuroeconomist of the month type of thing).

do teens consistently report that they find sex disappointing?

"Humans may have evolved to take teen risks to impress..."
Robin, you are jumping to a evolutionary pyschology explanation for no reason. I find this jarring.
Assuming it is true that "teens overestimate the benefits", this fact does not require an evolutionary pyschology explanation.
It would be like asking, why is it that humans can't lift 1000 pounds? Could it be that long ago, when humans lived in small tribes and knew all other members of the tribe, it was good to be limited in strength so you didn't accidentally pick up a large boulder and drop it on people?
Abilities (physical or mental) require an explanation to show why it was beneficial for evolution to preserve them. In contrast, limitations in abilities do not any explanation other than noting that there would be a non-zero cost to increase the ability.

"...the more plausible explanation is that teens overestimate the benefits, not the costs..."

And who are we to say they are flat out wrong? Who tells a 16 year old kid 'the risk of getting pregnant is 1 in 12, but the enjoyment gained from the experience will be only 350adm' (anti-dust motes). What price enjoyment? You've already said kids are better at analysing risk.

"...without even experiencing sex once, teens vastly outweigh the benefits..."

But these are human beings we're talking about. Their genes are screaming start reproducing at them. From an ev-psych or even genetic point of view, the cost (possible pregnancy) is not in fact a cost, it's the benefit (chance of your genes continuing their line). Add to that how cool and fun sex looks to an adolescent's eyes; is it any wonder we see what we see?

I know the counterarguments here too - if kids want to protect themselves against pregnancy with condoms, how can you argue from a reproduction point of view? Easy - kids don't want to have babies, but the physical urge is strong enough that they're willing to take those massive risks anyway. When I'm hungry, it's not my conscious mind thinking 'hmmm, haven't eaten in a while, better get some food.' It's a deep, chemical, physical urge to fulfil a survival function. As James tells us above, against a force like that, rationality ain't got a chance in hell. Same rules when telling kids it's irrational to eat too much sugar. Particularly when that 'rationality' relies on a value judgment as to the utility of an early sexual encounter. It was certainly the holy grail through my early years. ;)

Are we less rabid for sex as we grow older? Certainly. Is this because we get better at analysing risks and benefits? That might have something to do with it, but for me it's down to the fact that as teens we're horny 99% of the time, and this drops off as we age (no pun intended.)

Jor, if all of us equally overestimate benefits of risk , there is no special reason to limit teens

James, but why trick them if they aren't mistaken?

Anon, good point about teen culture being new - not clear which way it cuts though.

Alexa, you don't think teens understand what it means to die?

Briar, I was just arguing the theory isn't crazy, not that is it true or obvious.

Ben, are you saying teen preferences are just wrong?

It's worth noting that a pregnancy resulting from teen sex isn't (necessarily) an adverse outcome from a natural selection perspective. That's in contrast with other examples like HIV infection.

In an environment where one is surrounded by known liars (one's parents), why would one give much weight to a lecture on the dangers of an activity that the lecturers themselves CLEARLY partake in? This is especially true when they're being cagey about it (at home) or over-formal (at school). And failure to talk about the benefits is obviously because they're large and those giving the lectures want to deny that.

The topic of how to deal with those who seem less rational than yourself, without claiming that you're perfectly rational yourself, is a very difficult one. I think it needs to start with specific objective predictions and work from there. I'd also like to point out that "Teen" is a pretty wide range. Age 13 and age 17 are very different places on the road to rationality.

I have some objections to the article as well:
Surveys seem wrong for this kind of probability estimate. A lot of people don't really understand how much less likely 1/500 is than 1/12, and a frightening percentage will say that 1 in 20 is far more than 5%. Additionally, a lot of survey-takers will try to "guess the authority-given right answer" rather than estimating their true beliefs.

I also question the logic behind their recommendation of a "gist" as a decision methodology. They seem to be saying to ignore risk and reward calculations in favor of intuition. Robin's recommendation - honest discussion of both risk and reward - seems likely to lead to better decisions, and may even be more effective in manipulating our youth into behaviors we want. Especially if you include the social aspects - for most middle-class non-religious teens, the social risk/reward outweighs the physical by a long shot. This is true for sex obviously, but probably even for other risks.

I read Ben's point as "maybe the teens are right". How do we evaluate the value of the bonds formed by shared risk taking, or the freedom to research the benefits of something that most of society seems to be lying about, or even the happiness of a good orgasm?

Teen brains are immature. They physically work different to adult brains.
The prefrontal area is less important in decision-making.
James Bach above, is acting correctly in appealing to teen emotions.

Another series of MRI studies is shedding light on how teens may process emotions differently than adults. Using functional MRI (fMRI), a team led by Dr. Deborah Yurgelun-Todd at Harvard's McLean Hospital scanned subjects' brain activity while they identified emotions on pictures of faces displayed on a computer screen.5 Young teens, who characteristically perform poorly on the task, activated the amygdala, a brain center that mediates fear and other "gut" reactions, more than the frontal lobe. As teens grow older, their brain activity during this task tends to shift to the frontal lobe, leading to more reasoned perceptions and improved performance.


I think it's very plausible that they are overestimating the benefits and I like that part of your explanation. What I object to is how you immediately leap into evolutionary pyschology to account for why they might be overestimating when there is a vastly simpler explanation for why this might be happening (namely that it is much easier to overestimate something that it is to correctly estimate something).

How do you imagine honest paternalist interventions going? With alcohol and drugs I can see how it would go and yes, it would probably be moderately effective. But you're suggesting parents sit junior down and explain to him that sex isn't really that great? The kid will never believe another word you say.

Briarandbramble makes a good point. We must always be cautious in our use of EP. You wouldn't want to give the just-so story crowd any ammunition(ideas are soldiers :P)

I read about this same research on MR awhile ago and the conclusion of "gist" was repulsive to me then and now. I see it as basically "I *know* what's best for teens and if we can't reason them into it we will have to scare them into line." In the article the researcher said something like "any sane adult *knows* it's *never* rational to risk a one in 6 chance of dying for *any amount of money*). I think this goes to show that a little statistics can ruin a person and only with a lot of careful thinking can people take their 'book-learning' of statistics and risk analysis and actually manage to use it informally on emotional issues.

Adult genes and teenage genes must, at least some of the time, be at cross-purposes.

We should not be surprised if, among other things, adults espouse views that embrace differing discount rates than those used by teens. For example, a parent's genes may be best served by particular teen choices, while the teen's genes may be best served by alternative choices.

I'd risk a 1 in 6 chance of dying for "enough money to maintain my present lifestyle indefinitely without working or being supported by others". Then again, I may be undervaluing the negative utility of death...

Remember how it was to be adolesent : fun is here, now, immediate, and intense. Risk is abstract, remote, elsewhere. The NYT authors, Reyna and co-workers, make this point strongly in this paper :
The bias is not related to estimating probabilities, it's timescale related. Adolescents live in the present.

All I see are estimates of the odds, not the costs. Young people are, more or less by definition, lacking in all sorts of life experiences, both painful and pleasurable. That makes it pretty difficult to really get a handle on the personal costs and benefits of actions.

It's all very well to say that a teenager thinks the chances of getting pregnant from unprotected sex are, say, 50-50 (greater than 1/12). But do you really think most are capable of forming any kind of accurate intuition about what that pregnancy would really mean? All the potential hardships and heartbreaks and joys? I couldn't have. And the odds are pretty meaningless without it.

Insofar as consequences are seriously considered, the complexity and distance of it all tend to wash everything out, so it's often easy to end up on, "oh, f--- it, let's have fun."

All of which is to say that continuing attempts to educate teens about the actual consequences of things, as opposed to just the numerical odds, is at least somewhat helpful, not entirely "dishonestly paternal".

Your post was insightful and interesting. Please don't feel the need to spoil it with just-so conjecture about the evolutionary basis of these behavioral traits. It might be true, but it might not. It's going to be extremely hard to prove this either way and I don't think that this speculation adds much.

Many commenters (not Robin in the post) are assuming that teens take too much risk.

There is no good reason to believe this.

James Bach: I would much rather have kids who were able to make a sober assessment of risks and rewards than ones constrained by nakedly emotive arguments. Not least because it would be very sad for them to miss out on the rewards of drinking and doing drugs.

Paul Crowley, an emotive argument need not be constraining or dishonest. Even as they seek increased independence, many teenagers are quite loyal and respectful to their elders and superiors: if one understands that excessively risky behavior may cause emotional distress to one's parents, it's quite admissible to take this into account.

My thrust was simpler than a few people seem to make it out. I see no reason to make the assumption that adults know better than adolescents, particularly when a value judgment comes in.

Robin - I'm saying the exact opposite. When it comes to judging risk against something as subjective as pleasure, there is no right or wrong. There's what you do and what you don't do. By all means educate kids on the risks, just don't presume to have any divine adult-authority. This isn't just theoretical, this is good practice in trying to educate young people to do things. Anyone ever tried telling a kid not to do something? "I know better, because I use a different part of my brain to you. And I say wait till you're married!" Good luck with that one....

Be rational, explain the risks, and if you've done your job properly, I think most people would be surprised at how rational (most) teens actually are - see the original post by Robin for some evidence! There are as many stupid, irresponsible adults as there are stupid, irresponsible kids. The only difference is that the latter don't seek election to office.

I think Paul Graham had a good explanation for why parents' advice tends to err on the side of safety:

>All parents tend to be more conservative for their kids than they would for themselves, simply because, as parents, they share risks more than rewards. If your eight year old son decides to climb a tall tree, or your teenage daughter decides to date the local bad boy, you won't get a share in the excitement, but if your son falls, or your daughter gets pregnant, you'll have to deal with the consequences.


The comments to this entry are closed.

Less Wrong (sister site)

May 2009

Sun Mon Tue Wed Thu Fri Sat
          1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30