A developmental psychologist explains the surprising cognitive effect of having friends near.

Question: What is one of your most unexpected recent findings?

Laurence Steinberg: The more recent finding is this peer effect, where -- we did a study where we had people come into the lab to take a batter of risk-taking and decision-making measures on a computer. And we asked them to come and bring two friends with them. And then we randomly assigned them to either do these tasks by themselves or to do it with their friends watching them. And we had teenagers, college undergraduates and adults, you know, 30 or so years old. And so what we found in this study was that the mere presence of peers doubled the number of risks that adolescents took. It had a 50 percent increase among college students and no effect at all among adults. Now, this was really interesting to me. Nobody had shown this experimentally before, although it's consistent with the idea that most adolescent risk-taking occurs when they're in groups and with their friends.

So we just recently -- my colleague Jason Chein, who's a neuroscientist, and I just recently took this experiment into the FMRI. So we have the subjects with their brains being scanned, and they're lying down on the magnets, the computer screen is in there, they can do the task, the same task. We have them come with two friends; their friends are in the next room. Their friends can see their performance on a monitor that's in the next room, and their friends are miked into the magnets. So on cue, when we tell the friends, they say, okay, Dave, I'm watching you now, all right? And we compare people's performance on these tasks when they know their friends are watching them and when they know their friends can't see them at all. So they're not even in the same room with them; they're just aware of their presence. I mean, they're not encouraging them to take risks or anything like that, because we script what they're allowed to say. And we've just discovered --we're in the process of writing this up right now to send out to a journal -- that the presence of peers activates the brain systems that are in this reward circuit that we've been talking about, in kids, but it doesn't in adults.

And so we think that the reason that we saw the peer effect is that when you are with your friends when you are a teenager, it increases the salience of the rewarding aspects of a risky decision, which then of course is going to make you pursue the risky decision rather than be risk-averse. So we're really excited about this finding. And we are just working on a paper this morning in which we looked at the impact of the presence of friends on people's preferences for immediate versus delayed rewards. And we're just doing this in the scanner now, but in our behavioral research we see the same thing. So we have a sample of 19-year-olds, and they do this task -- it's called a delay discounting task -- where it basically measures how much less of a reward you'd be willing to take to get it now rather than waiting. When the 19-year-olds do this with their friends watching them, they make decisions that are comparable to what 14-year-olds make. But when 19-year-olds do it with their friends not watching them, they behave like adults. And so again it's this impact of peers on reward processing that seems to be going on that may play a role in understanding risk-taking. So when we first saw -- you ask what gets me excited -- you know, when we first saw these brain scans, we couldn't believe it because it looked too beautiful to be true. And we peaked kind of early on; we'd only run, you know, a couple dozen subjects. And now we've done the whole experiment, and it's just amazing. It has held up, so we're really, really jazzed about this.