Early Migration To North America Likely Wasn’t A One-Way Road

11:40 minutes

a landscape shot of low rolling hills as far as the eyes can see. clouds dot the blue sky
Bering Land Bridge National Preserve. Credit: National Park Service

The story of how early humans migrated to North America might not be as simple as we once thought. The prevailing theory was that ancient peoples traveled from Siberia to modern-day Alaska using the Bering strait as a land bridge. But new genomic research, published in Current Biology, reveals movement in the opposite direction, back to Asia, as well.

Ira talks with Sophie Bushwick, technology editor at Scientific American, about the new research, and other top science stories of the week, including a new AI voice generator, a green comet visible visit in the night sky for the first time in 50,000 years, and how a specific atmospheric weather pattern caused historic flooding in California.

Donate To Science Friday

Invest in quality science journalism by making a donation to Science Friday.


Segment Guests

Sophie Bushwick

Sophie Bushwick is senior news editor at New Scientist in New York, New York. Previously, she was a senior editor at Popular Science and technology editor at Scientific American.

Segment Transcript

IRA FLATOW: The story of how early humans migrated to North America just got a little bit more complicated and interesting. You probably remember the story pretty well. Early humans, living in what is now Siberia, crossed the Bering Strait into modern-day Alaska. And here’s where it gets even more interesting. Because new genomic research shows movement in the opposite direction, too. These early migrants traveled back again to Asia.

Joining me now to explain this fascinating new research and other top science stories of the week is Sophie Bushwick, technology editor at Scientific American, here in New York.

Sophie, welcome back. Always good to see you on Science Friday.

SOPHIE BUSHWICK: Thank you. It’s good to see you for the first time in 2023.

IRA FLATOW: And Happy New Year to you. Now, let’s talk about this. What’s the new information we’re learning here?

SOPHIE BUSHWICK: So researchers took the remains of 10 individuals from what we now call Siberia, that area of the world. And they did a genetic analysis. And these are the remains of humans who lived at various periods of time. The oldest ones are about 7,500 years old. And yeah, they tested their genes. They said, who are your ancestors?

And they found evidence that they had Native American genes. And this suggests that there were not one, but multiple times when Native Americans either crossed over or people from Siberia crossed to America and then came back.


SOPHIE BUSHWICK: So there’s a lot of genetic mixing in there that suggests that these populations were not just traveling to a new place and staying there forever. That there was back and forth going on for thousands of years.

IRA FLATOW: So how does this new research change, then, our understanding of early human migration patterns?

SOPHIE BUSHWICK: Well, it shows that these early nomadic humans were traveling very large distances. Some of these remains they tested, they weren’t just in the area around the Bering Strait.


SOPHIE BUSHWICK: They were much further away, in the far eastern part of modern-day Russia. And in these same populations, they’ve found genetic influence from other groups. They found that people from Japan probably had genetic mixing with people from what is now Russia. So there was a lot of travel going on in the ancient world.

We think of ourselves as being at the most connected time in human history, with planes and stuff. But our ancestors were no slouches either.

IRA FLATOW: I love it when new stuff like this comes up, right?


IRA FLATOW: It makes everything so much more interesting.


IRA FLATOW: All right. Let’s move on to some rather depressing news this week, as if we need any more of it. And this one’s about the state of our glaciers. And not those in the Arctic or Antarctic, but the mountain glaciers, right? Half of them are expected to melt by 2100. I hate to hear that.

SOPHIE BUSHWICK: And that’s the best case scenario.


SOPHIE BUSHWICK: Yes. So if the world managed to limit warming to 1.5 degrees Celsius, then we would still expect half of all the mountain glaciers in the world to be melting by 2100. But the problem is that that’s if we limit warming to 1.5 degrees. Right now, we’re on track for 2 degrees, based on the limitations that countries have made in the attempts to reduce emissions. And that would melt about 60% of the mountain glaciers. And each time that number goes up each time that amount of global warming increases, we can expect to lose more and more glaciers.

IRA FLATOW: And so we’re talking about major mountain chains in places.

SOPHIE BUSHWICK: Absolutely. Yes. So there’s glaciers in places like– big mountain chains in South America. You might think of the Alps. These glaciers are very valuable not just as sources of tourism for people who want to check them out. Also, in some places, they have a spiritual significance to Indigenous populations.

And they also provide a reservoir of water. So when it warms up, water from the glaciers might melt and flow down from the mountains and supply people who live at lower elevations. So these are very, very important parts of our world. And the risk is not just that we’ll lose them, but where are they going to go? They’re going to go to the ocean and contribute to rising sea levels.

IRA FLATOW: Loss of the pack, that is important.

SOPHIE BUSHWICK: That’s a big deal. Yeah.

IRA FLATOW: A big deal.

Now let’s move on to a team of researchers in China, announcing earlier this week that they have figured out how to break one of the most common digital encryption methods. And they’re using quantum computing?

SOPHIE BUSHWICK: That’s right. So researchers have known about an algorithm for breaking this type of encryption for a long time, called Shor’s algorithm. But the problem is you would need a quantum computer with about a million qubits– that’s quantum bits, the building blocks of a quantum computer– in order to run this algorithm.

Now, these Chinese researchers have said there’s a different algorithm, confusingly named Schnorr’s algorithm, as opposed to Shor’s algorithm, which can be– it was written originally for a regular classical computer– but they said, if you run it on a classic computer, you can break encryption using a computer with just a few hundred qubits, instead of a million or more.

IRA FLATOW: So is that the big breakthrough here, they don’t need a supercomputer, they can run it on a–

SOPHIE BUSHWICK: Well, it sounds, really, like a big breakthrough on the surface.

IRA FLATOW: I’m worried now.

SOPHIE BUSHWICK: Right. Don’t worry. Once you start looking a little closer, you realize that I don’t think we have anything to worry about yet.

So first of all, if you wanted to run it on a quantum computer, you would need to have extremely low rates of error. It would have to be extremely accurate. And quantum computers just aren’t that accurate. They’ve got to be kept in these very pristine, specific conditions in order for all these qubits to remain entangled and in the right state. And it’s just essentially really– at our current level of technology, it’s nearly impossible to get the error levels down to the point where this would work.

And then the other issue is it’s unclear how long it would take this quantum computer to crack encryption running this algorithm. So it could crack it, but how long would it take? Would it take hundreds of thousands of years? We’re not sure how much faster we’re going to be able to break it, to the point where it would be like a practical solution. So until they’ve got a better sense of that, I would say, don’t worry.


SOPHIE BUSHWICK: I mean, the big problem that we’re worried about is down the road. Like, let’s say today a hacker steals some encrypted data. And then they hold on to it for several years. And then maybe, eventually, the technology catches up. And then quantum computers get good enough to run these algorithms that break encryption. And they have this cache of data that they can now break into. I think that’s the worry that a lot of cryptographers have.

But there also, at the same time, people aren’t just sitting around waiting for this to happen. They’re working on new forms of encryption that hopefully even a quantum computer couldn’t crack.

IRA FLATOW: Well, let’s get a little deeper into the digital world. And I’m talking about artificial intelligence. Because this week Microsoft researchers announced a new tool, called VALL-E. You can recreate someone’s voice just on three seconds of recorded audio. I’m fearful about my job now. How does that work?

SOPHIE BUSHWICK: So this is work from Microsoft researchers, but they actually used a tool developed by a different tech giant, Meta, the company formerly known as Facebook. But Meta has a tool that these Microsoft researchers took advantage of, which can basically break down your three-second clip of audio into all these little discrete components. And then the Microsoft researchers built a tool that you use these components to teach a model how to make a voice that’s imitating the original clip.

So yeah, they say that using only a three-second audio clip they can reproduce a voice. But something I’m grateful for is that they have not released this yet. A lot of other AI tools, people are free to play around with. You can play around with ChatGPT, you can play around with DALL-E. But that’s not the case for this. The Microsoft researchers have said, first of all, they’re not going to release it right away.

And second of all, when they do, they’re going to include a sort of watermark in any recordings or audio that’s built on this tool so that you can tell it’s not original. Which would help cut down on using a deepfake to spread misinformation or to try to blackmail someone, for instance.

IRA FLATOW: Well, you could tell it’s not original in mine if there are no dad jokes in it.


SOPHIE BUSHWICK: That’ll be the dead giveaway.

IRA FLATOW: But this is not the first program, right, to use AI to recreate someone else’s voice, or even video?

SOPHIE BUSHWICK: Right. The difference here is how fast it can do it, how little of a sample it needs in order to recreate that. A lot of people who build deepfakes of someone’s voice, they might want to look at hundreds of hours of recorded audio, as opposed to three seconds.

And the other thing is that it also reproduces the emotional tone of the voice in the three-second clip and also what’s going on in the background. So if I’m talking cheerfully in this clip and I’m talking on a telephone– so it’s got a certain sound quality to it– any imitations of that clip are going to sound like I’m talking on the telephone and I’m cheerful.

IRA FLATOW: Wow. Wow. Wow. This is so interesting. We’re going to devote a whole segment to this coming up in the near future, talking about AI and the explosion of all these tools.

SOPHIE BUSHWICK: It’s really fascinating technology.

IRA FLATOW: Really, really, really. Let’s go off to space. There’s a new comet and it’s green. And you can see it in the night sky, right?

SOPHIE BUSHWICK: That’s right.

IRA FLATOW: It’s coming around for the first time in 50,000 years.

SOPHIE BUSHWICK: I know. It orbits the sun at a very far distance, which is why we haven’t seen it in 50,000 years, as you say. And it’s going to be visible in the Northern Hemisphere in the early evening and also possibly in the early morning. If you want to look for this, you’re going to probably need binoculars, although it might be bright enough to see with the naked eye towards the end of the month. But what you’re going to want to do is look for the North Star, Polaris, and it should be around there. And it’ll have this fuzzy green glow around it.

IRA FLATOW: And Friday night, tonight, it gets closest to the sun, right?

SOPHIE BUSHWICK: That’s right. That’s right.

IRA FLATOW: I’m going to look for that. I remember seeing Comet Hale-Bopp. I remember– and that was years ago– but it’s one of the most exciting things you can do if you’re an amateur astronomer is actually see a comet for yourself. But that’s another story.

Let’s end on a tasty note. Utensils that can make your food sweeter or saltier without sugar or salt.

SOPHIE BUSHWICK: That’s right. So people have known for a while that a whole bunch of factors influence the taste of your food other than the contents of the food itself. Some people try to eat with cutlery made of gold or different materials. They found out that cutlery with different textures can change the way something tastes.

So previously, researchers had developed these chopsticks with a very mild electric current running through them. And the idea is that that current moves around sodium ions in food. And so when you eat food with these chopsticks, it tastes salty even if there’s not salt in it.

IRA FLATOW: Oh, no kidding.

SOPHIE BUSHWICK: Right. So if you’re trying to cut down on your sodium but you don’t like bland food, you could have a low-salt meal but eat it with these electric chopsticks and that would stimulate your senses a little more.

IRA FLATOW: No one is getting zapped with these chopsticks, though, right?

SOPHIE BUSHWICK: Right, right. This would be a very mild current. And those are actually supposed to come to market this year, in 2023. So they might be available for purchase soon.

And then there’s a spoon that’s not quite as close to being available. But some student researchers actually developed the idea, the design, for a spoon they call Sugarware that would have these bumps on the bottom of it. And the bumps would be coated with a material that interacts with the sweet sensors on your taste buds.

IRA FLATOW: There you go. Thank you, Sophie. Always exciting stuff. Thanks for taking time to be with us today.

SOPHIE BUSHWICK: Thanks for having me.

IRA FLATOW: Sophie Bushwick, technology editor at Scientific American. She’s based here in New York.

Copyright © 2023 Science Friday Initiative. All rights reserved. Science Friday transcripts are produced on a tight deadline by 3Play Media. Fidelity to the original aired/published audio or video file might vary, and text might be updated or amended in the future. For the authoritative record of Science Friday’s programming, please visit the original aired/published recording. For terms of use and more information, visit our policies pages at http://www.sciencefriday.com/about/policies/.

Meet the Producers and Host

About Shoshannah Buxbaum

Shoshannah Buxbaum is a producer for Science Friday. She’s particularly drawn to stories about health, psychology, and the environment. She’s a proud New Jersey native and will happily share her opinions on why the state is deserving of a little more love.

About Ira Flatow

Ira Flatow is the host and executive producer of Science FridayHis green thumb has revived many an office plant at death’s door.

Explore More