What Is Intelligence?

Steven Pinker
Psychologist, Harvard University

How did human intelligence evolve?

Steven Pinker: Human intelligence and intelligent behavior don't just come from having a whole bunch of stuff packed into our skull like meatloaf. The actual organization of behavior goes on at the level of the individual nerve cells and their connections, and we have 100 billion nerve cells, probably 100 trillion connections. It's just mind-boggling to think of all the different ways in which they're arranged in a baby's head. And a lot of our evolution consisted not just in getting more of this stuff but in wiring it in precise ways to support intelligence.

It's very likely that the changes in the brain didn't happen overnight. There wasn't one magical mutation that miraculously allowed us to speak and to walk upright and to cooperate with one another and to figure out how the world works. Chances are there were lots and lots of mutations over a span of tens, maybe even hundreds of thousands of years that fine-tuned and sculpted the brain to give it all the magnificent powers that it has today.

The reason that we live differently today from the way the cavemen lived is not because we have better brains but because we've been accumulating all of the thousands of discoveries that our ancestors have made, and we have the benefit of a huge history of inventions that we communicate non-genetically, through language, through documents, through customs. This is called cultural evolution; some people call the units of cultural evolution memes—little units of memory or knowledge—and we've been accumulating them for tens of thousands of years.

*****************************************************************************************

Nicholas Humphrey
Psychologist, London School of Economics (emeritus)

Is social intelligence the key to our species?

Nicholas Humphrey: Social intelligence is the ability to take in information from the social world and use it to make predictions about what's going to happen next in the community in which you are living. It is so familiar to us. We do it, of course, all the time. We spend our time wondering what's going to happen next in our own lives, in our friends' lives. We're always analyzing, trying to get ahead of the game by interpreting behavior, reading minds, and then, like a game of chess, moving several moves ahead of where we have got to now to see what possible outcomes there could be.

I first became interested in this partly because when I first went out to Africa to work with gorillas, I realized that the life of the gorilla in the forest is actually a rather easy one. It's got plenty of food all around it. It's got very few predators. The physical environment wasn't presenting them with difficult problems. There was another dimension to their lives, and that was the social world.

To manufacture new relationships, to solve the squabbles, to take advantage of the opportunities for mating or friendships as they came along—all those were setting problems for the gorillas, which perhaps really did need full-scale intelligence. And so from there I took off with an idea, which became known as the social function of intellect and gave rise to ideas about social intelligence having been the prime mover in the evolution of higher primates, including human beings.

*****************************************************************************************

Rodney Brooks
Roboticist, MIT (emeritus)

What is artificial intelligence?

Rodney Brooks: Artificial intelligence, to me, is trying to get computers and robots to do stuff that if people did them, you'd say, "Oh, they're demonstrating their "peopleness." And artificial intelligence is trying to get at that stuff that humans do. There are four challenges, and if we can make progress on any one of these challenges, our robots would be a whole lot better than they are today.

The first thing is to get a robot to be able to have the visual object recognition capabilities of a two-year-old child. And the stuff that a two-year-old could do, like recognize a shoe or a chair, that's the really hard stuff that we can't do well yet. So there's object recognition capabilities of a two-year-old child.

By the time you get to be four years old, a child knows all of syntax of language, is really good at understanding different accents. So the language comprehension of a four-year-old child. A six-year-old child can tie shoelaces—way more dexterous than any of our robot hands. So let's aspire for the manual dexterity of a six-year-old.

And lastly, by the time you get to eight years old, you understand social interactions. They're able to put themselves in people's heads and understand what's going on between each other. If we can make progress in any of those four, our robots are going to get a whole lot easier to interact with and a whole lot more useful.

*****************************************************************************************

Seth Shostak
Astronomer, SETI Institute

What would extraterrestrial intelligence look like?

Seth Shostak: In the SETI business, if you will, the search for extraterrestrial intelligence, our definition of intelligence is very operational, very pragmatic, very simple really. And that is, if you can build a radio transmitter or a powerful laser so that we can pick up signals from you at light-years distance, then from our point of view, you're intelligent.

Certainly one of the most controversial aspects of this whole endeavor is whether intelligence is a common evolutionary development or not. I mean, the question boils down to this: If I give you a million worlds with life and let them sit around for a couple of billion years, what fraction of them will ever cook up anything that can build a radio transmitter. So maybe there are very few worlds where you actually have intelligence. We don't know.

Today, biologists are looking into this, evolutionary biologists, and they seem to be of two minds. On the one hand, there are quite a few of them who think, you know, intelligence is a very rare evolutionary development. Don't count on it. And yet there are other people who point out that there are other species besides our own that have gotten quite a bit cleverer in the last 50 million years. Think of dolphins, or for that matter, octopuses or some birds and, obviously, apes and so forth. Some of them are not terribly closely related to us and yet they still got more intelligent. So their point of view is that intelligence will eventually spring up on many worlds. We don't know the answer to that, and maybe the only way we're going to find the answer to that is by finding intelligence somewhere else.

Producers:

Susan K. Lewis & David Levin

Interviewers:

(Pinker) John Heminway; (Humphrey) Melissa Salpietra; (Brooks) Michael Bicks; (Shostak) David Levin

Designers:

Sonali Patel, Jon Hayward & Tara Taylor

Developer:

Daniel Hart

Images

(brain illustration)

© Sebastian Kaulitzki /iStockphoto

(Pinker)

Rebecca Goldstein/courtesy Steven Pinker

(neurons)

© ktsima /iStockphoto

(skull casts)

© WGBH Educational Foundation

(books)

© Tatian /iStockphoto

(Humphrey)

courtesy Nicholas Humphrey

(chess)

© Rene Mansi/iStockphoto

(gorillas)

© Geunter Guni/iStockphoto

(Brooks)

courtesy Rodney Brooks

(yoga robot)

© Julien Tromeur/iStockphoto

(robot with phone)

© Viktoriya Sukhanova/iStockphoto

(robot handshake)

© Dirk Freder/iStockphoto

(Shostak)

courtesy Seth Shostak

(telescopes)

© cbpix/iStockphoto

(planets)

© Sergey/iStockphoto

(dolphins)

© Krzysz/iStockphoto