Teen learning exaggerates rewards – Bad decisions and brain development

Teens tend to make some seriously stupid decision (including teen drinking and driving), at least when compared to younger kids and older adults. We’ve all heard that brain development during that part of life plays a role in this but the question is: What exactly about brain development makes teens more risky?

There are a number of options – 1) Teens could have less control over all aspects of their behavior as their prefrontal cortex finishes developing, 2) Teens may be over-sensitive to rewards, putting too much emphasis on the positive value of stimuli they’re exposed to, or 3) Teens might just be less sensitive to the negative consequences of their action, which lets them take risks others just wouldn’t.

A somewhat recent study coming out of UCLA (and I just have to congratulate my colleague Jessica Cohen for getting a 1st author Nature paper!) suggests that at least to some extent, oversensitivity to positive reward-signals may be the answer and brain development has a lot to do with it.

You see, as we navigate around the world, our brains register expectations about situations – Touching the hot stove will be bad, eating that piece of candy is good, and opening a door to walk outside of one’s house is neutral (all examples, your expectations might differ). Many of those expectations are created through experience with the world and our brain uses them to make decisions about actions to take and actions to avoid.

When we expect something to be bad, or neutral, and it turns out to be good (or if it started out as good but turns out to be even better), short dopamine signals in the midbrain including the striatum, Ventral Tagmental area, and the Nucleus Accumbens tell the brain “Hold on, we made a mistake, this is good stuff!!!” This is the process by which our brain updates it knowledge about things that are good in general and addiction is thought to exist partly because drugs mess with this process in ways that likely create too many such “good” correction errors when drug use is concerned. But back to the study.

Teens and reward prediction study

Researchers used a learning task, conducted while participants were in an fMRI scanner, to assess the brain activity associated with decision values (response to the stimulus) separately from the prediction error (response to the feedback during the exercise). The task was a relatively simple “game” whereby participants had to classify abstract symbols as being either “Northern” or “Eastern” and would get small monetary rewards if they answered correctly. Remember that in reality, the researchers weren’t interested in what the participants got right but rather in the way their brains reacted to the stimuli and the feedback about their classifications.

One of the interesting results from the study was that only teens got faster at responding for symbols that became associated with large rewards. That should already be telling us something about teens response to reward… The major finding was that teens seemed to produce the strongest positive error-prediction signal of the three age groups.

What does this mean? This suggests that indeed, teens have these newly acquired large responses to unexpected rewards but have not yet acquired the inhibitory capacity they will fully develop in adulthood. So when their brains tell them something has repeatedly turned out to be good, or better than expected, they are more likely to go for it without much thought. Younger kids are not yet sensitive enough to the reward and older adults have the improved ability to control their actions. It makes so much sense!

This result is not meant to remove all responsibility from teens when it comes to making appropriate decisions. But it should inform our expectations and the way that we can help teens deal with this biological reality. Making sure that your teen is aware of the consequences is still important, but if we don’t do our work to make certain that their environment supports their decision making we may be in for some seriously unpleasant surprises. We’ve written before about teen drinking and I don’t think the phenomenon is going away but one way to keep everyone safe is by making sure that no drinking teen has their car keys available – no keys, no decision to drive, no DUI accident. A little planning can go a long way towards avoiding a disaster even if there’s a little gap here when it comes to wisdom and we can’ completely stop teen drinking and driving.


Jessica R Cohen, Robert F Asarnow, Fred W Sabb, Robert M Bilder, Susan Y Bookheimer, Barbara J Knowlton & Russell A Poldrack (2010). A unique adolescent response to reward prediction errors. Nature Neuroscience, 13, pg 669-671.

2 responses to “Teen learning exaggerates rewards – Bad decisions and brain development”

  1. Interesting study. Drinking teens, nor any age person, should not drive. The difficult part is that most teens are not drinking around responsible adults, to implement the no driving rule. Parents certainly can emphasize the importance of the designated driver in these situations. What occurred to me was the idea of parents of teenagers rethinking a reward system for positive behavior and how much that may play into their living healthy lives throughout their teen years.

Leave a Reply

%d bloggers like this: