 Hi, I'm Gary Marcus. I'm a cognitive psychologist at NYU, and I study how children learn language. And lately I've been studying how anybody can learn anything at any age. And one thing is the number that just keeps coming up over and over again is this idea that you need 10,000 hours of practice to get great at anything. I've seen it so many times in so many different books, and it's just not true. If you read these books, you'll hear stories like Bill Joy, who programmed computers for 10,000 hours and started Sun Microsystems, or The Beatles, who practiced in the cavern club for many, many hours. And all of these books point to the same set of data. The data were gathered by Anders Ericsson, who's the expert on expertise. And he found that violinists, by the time they were 20, had practiced for about 10,000 hours. So that's a very interesting piece of data, but it's just an average. So there are lots of things that actually go into how good you are, like talent, like how competitive the field you're in. So there are, for example, bands like Anvil, some of a few of you might recognize, who practiced for a way more than 10,000 hours for 30 years and still hardly anybody knows who they are. And then there were Ramones, who practiced for about two weeks before they discovered a new way of playing. It became famous. What you really want, of course, are systematic data. Some of the best data come from chess playing, and sure enough, the average number of hours that master level chess players practice before they got to be masters was 11,000 hours. But there are some people that only took 3,000 hours, other people that took 23,000 hours, and some people, at the end of the study, after 25,000 hours, who still hadn't gotten there. Which brings me to the topic of chick-sexing. I know that's what you all came to hear about tonight. This is the fine art of grabbing a chicken, a newborn chick, looking at its genitals and deciding, is it a boy or is it a girl? And this is a classic example of practice. It takes thousands and thousands of hours before people get any good at it. And you can verify this in the lab. You can show people photographs of boy and girl chicks. And the experts, sure enough, they do better than novices. But the other thing that you can do in the lab is instead of having people learn by trial and error by looking over the shoulder of the masters, you can tell them what they should actually do. And then in two hours, the novices are almost as good as the experts. So it really matters how you practice. If learning a new skill doesn't necessarily take 10,000 hours, does it still require that you start by the time you were 10? That's certain banjo players have done. This is a question of great personal interest to me because I started playing guitar just before I turned 40. So this is where my science intersects with my personal life. And the idea here is called the critical period effect. So the notion is if you're going to learn a language, you need to start by the time you were three. If you're going to play golf, you better start when you're five. And if you have amblyopia, you better get a cure right away. But this idea, too, although it's very popular, it's a meme that's out there in the atmosphere, is not, in fact, true. It's wildly overrated. So for example, one recent study showed that you could take adult amblyopics, who we all thought couldn't learn anything new, and you can help their visual acuity, their spatial attention, and so forth by the complicated intervention of simply putting an eye patch on them and having them play Medal of Honor, a first person shooter game. And it actually improves their vision. The best data actually come from experimental data with animals, where you can really control what the learning is, what the practice is. And the textbook example of critical periods is with barn owls. You put a prism in front of them. And it's sort of like a virtual reality experiment. It distorts the entire world. And in all the textbooks, they tell you, if you do this for an adult barn owl, they can't really do it. They can't do it like a child. But a new data has shown that if you make them do it incrementally, step by step, instead of dealing with a very large distortion all at once, and you do sort of a little bit each day, the adult owls do much better than anybody ever expected. So if you are thinking of learning something new, think of Michelle Steele. She took up downhill skeleton at age 18. And six weeks later, she came in sixth on the World Cup. And she made the Olympics after just 14 months of practice. So to close, I want to say something about something new that I have been learning to do, which is to play guitar while riding a unicycle. So I want you to take three things with you. If you want to learn something new, don't give up just because you're past puberty. It really matters how fast you learn. It's not just about how many hours you put in and how intelligently you practice. Thank you very much.