As we get really proficient at a video game, our brain changes as we are playing it.

It is not the same when you fist learn the game and when you are an expert and you play it,

the brain activity is not the same. In fact, there is even less brain activity once you

are an expert. You kind of make everything automotized and easier. So we are studying

how that process unfolds in the brain. But then, even more interesting I think, we are

studying how your brain activity when you first come into the lab, before you have not

even played the game at all, how much that can predict your future improvement in the

game and your future improvement at other things like whether your memory improves or

your reaction time improves outside of the game.

So the game is a really boring old game that was invented here at the University of Illinois

in the 1980s to study some basic cognitive tasks like memory and reaction time.

So you have to fly around this space environment. And if you have every played those types of

games, you know that they are hard and your plane goes flying around. So lots of the early

learning is subjects learning to control this ship in space. But then, at the same time,

we have some other tasks that are embedded in the game. So there are these complex processes

that are kind of built on top of this basic flying game. They play the game for their

first couple of times with EEG on -- with a full head of EEG on. We can control each

individual aspect and we know the exact milisecond where things come of the screen, which allows

us to time-lock our EEG data so we can find exactly which type of brainwaves or brain

activity were happening. So we get their brain activity when they are first learning to play

the game. And then they play the game for twenty hours over the next month. And then

we bring them back and we compare how much they have improved during that training and

go back to the first set of brain activity and try to make some predictions.

What we found that was really interesting is that people that learning the play the

game the fastest, they had the most alpha oscillations - alpha waves - when they first

played the game. Alpha waves are a particular type of brain aciticity where big groups of

your neurons fire together in snychromy, then times every second. The people that improve

had more power alpha waves. So it was not that they were happening faster. It is that

they had bigger groups of neurons that were all firing together and going off together.

So maybe their brain was more coordinated in its alpha.

Even not when they are playing -- so we bring them into the lab and have them do some other

unrelated tasks on the side that are not even the game -- and we can use their alpha waves

from that side task, before they have even played the game at all, to predict how fast

they will learn this game.

The first EEG finding in the 1930s that Hans Berger put EEG on his son and found that,

when his son closed his eyes or stopped paying attention, that this alph activity would go

up; there would be more peaks, ten times every second. And so that was thought of as kind

of just the idling process of the brain. But it has come to be found in many more brain

areas that this is a more controlled kind of inhibition. So it is not just an "off"

that happens whenever you are not doing anything. But you can selectively chose which parts

of your brain to inhibit with these alpha oscillations.

We did not try to predict who would be the best when they were done playing.

We were not able to predict that as well. What we are able to predict is kind of the slope that

people learn. So we do not know where people are going to end up after a certain amount

of time. We just know the tragectory they are going to take to get there -- which is

perhaps more interesting. Because the endpoint is probably just a function of different people's

learning slopes. But the real difference is, can you learn it in a second or does it take

you a little longer? That is what educators and trainers and the military would want to

know more about -- which are the people can get it in a second and which are the people

that you need to draw out the training a little bit more.

Another interesting thing is that it has become clear that you can give people feedback in

real-time about their alpha; which is something we have not done here but something we want

to try. So you can measure their alpha and then tell them how much they have and say,

"Okay, increase it." And they do not necessarily know how to increase it. But they will try

something. And if it works, you will give them feedback and slowly you can get them

to boost up or dampen their alpha. And we might try to do this when people first start

playing a game and see if we can manipulate their future learning. Which would be really

cool is you can really change people's proficiency to play a game just by boosting how they go

into it that first time. That would be really amazing.

<MUSIC>