The Swarming Intelligence Of Ant Colonies And Slime Molds
If you’ve watched a colony of ants hunting for food along a trail or the pulsating movement of a slime mold, these organisms might seem to swarm in an undirected, random way. But researchers studying these seemingly mindless organisms are learning that they actually assemble themselves into a collective unit capable of decision making. The findings could be used to develop algorithms for artificial intelligence. For more on the research, check out SciFri’s Macroscope video below.
Luke Groskin is Science Friday’s video producer. He’s on a mission to make you love spiders and other odd creatures.
SPEAKER 1: That sound means it’s time to talk about our new “Macroscope” video. It’s up now at sciencefriday.com. Video producer Luke Groskin joins me in studio to talk about it. Welcome back, Luke.
LUKE GROSKIN: Hi, John.
SPEAKER 1: So the “Macroscope” video this week looks it ants, slime molds, and the New Jersey turnpike. What’s the connection?
LUKE GROSKIN: So besides the fact that you don’t want any of those near your home, they’re the field of research for Dr. Simon Garnier of the New Jersey Institute of Technology who runs a swarm lab. And what he’s interested in looking at is at communal behaviors, at group decision making.
How do ants or unicellular organisms, how they decide to do things? What’s the mechanisms? What’s the systems? And also, how do we as people do the same? How do we decide to work together and how does our systems work together?
SPEAKER 1: OK, so let’s take ants first. You look at ants on the ground. They seem to just be roaming around randomly, but they are very organized.
LUKE GROSKIN: Sure. Yes, they are. And the question is, how? Because there’s no leader. The queen’s not sending out orders. Go over there. Go get those mealworms and bring them back.
So all decisions are made communally, and they’re done through kind of a system of communication, for most species with pheromones. And Dr. Simon Garnier, he researches one species called the army ant. And the army ant is very nomadic.
They wander around the forest. They don’t set up and build colonies underneath the ground, and that means they have to be extremely efficient at foraging. And what that means is that they have to work together to make the most efficient pathways.
Now, Dr. Garnier and his colleagues have done experiments with them where they set up gaps where the ants have to kind of find a way to get to bridge a gap. And if you watch the video, you’ll see this incredible time lapse footage where you see in– you see these ants build a bridge or a ladder to cross a gap. It’s really fantastic behavior. And what they found was they think that these ants, when one ant is being stepped on, that’s the cue for them to just stop. So there’s a programming inside the ant to just stop, and you repeat that over and over and over again, and eventually you get a bridge that crosses a span.
And now, what’s really cool is not only can you build a bridge, the bridge can move. So the bridge could be like the George Washington Bridge. More people want to go downtown that day and the George Washington Bridge angles itself more downtown to create a shorter path. The ants do this.
Now, there’s no leader. There’s no decision. Nobody says it’s time to do this, but they do it anyway.
SPEAKER 1: And you have to see this video, because this is the coolest thing, these ants making the bridge. I’m John Dankosky and this is Science Friday from PRI, Public Radio International. And so Luke, so now ants, we know what ants look like, but we never think about slime mold. Tell us how slime mold fits into all this.
LUKE GROSKIN: OK, so slime mold. Imagine, John Dankosky. Now we take all the walls of his cells and we remove all the cell walls. So now he’s just a blob of John Dankosky. Now we take out his brain, and then you tint him yellow, and that’s kind of what slime mold is.
It lives on the forest floor, and it loves moisture and darkness, and it just meanders around looking for food. Now, it has no eyes and it can only feel for food. So how does that make a decision? How does that creature decide, oh, it’s time to go over here. It’s time to go over there.
So what they saw, what they realized is that the slime mold will pulsate. It’s really, really gross. It just kind of grows in specific areas and just kind of– it pulsates is the best way to describe it. And the more rapid the pulsation is at a specific spot, the more fluid and resources get pumped to that spot.
Now, they’ve done experiments at this lab where– in New Jersey– where they watch the slime mold. They put more food on one side of a track and then they put less food on the other. And 80% of the time, the slime mold goes to the side that has more food.
SPEAKER 1: Hold it. If we flip a coin, we make a decision. It’s like 50-50, right? But this slime mold’s right 85% of the time?
LUKE GROSKIN: Yes. It’s absolutely way, way better than anything we could do. And you can apply the same sort of concept to economics or buying a new computer. You don’t know what the best model is out there. You only have a set amount of information.
So which one are you going to go with? The slime mold has no eyes. It has no sensory equipment besides its touching food, but yet it makes the decision right 80% of the time.
SPEAKER 1: And one of things we hope to learn out of all this is not just how to maybe go to the Casino and bet or maybe play the stock market, but also how to get around and how to, I don’t know, get off the New Jersey Turnpike.
LUKE GROSKIN: Oh yeah, because I mean, you want to find the quickest route out of the New Jersey turnpike, which is a horrendous road if you’ve ever been on it. And what Dr. Garnier’s hoping to do is to take these models, these organismal models, and apply them to other fields. So what can we learn about criminal behavior? Or what can we learn about traffic systems? How can we design these systems better so that we can actually achieve the same efficiency as some of these other creatures?
SPEAKER 1: So we can learn a lot from ants and slime mold. I never knew. Luke Groskin is Science Friday’s video producer. Check out this “Macroscope” video “The Road Best Traveled” at sciencefriday.com. Thanks so much, Luke.
LUKE GROSKIN: Thank you.
SPEAKER 1: One last thing before we go. A few weeks ago, we told you about astronomical confusion over a distant star with the friendly name of KIC 8462852. Astronomers using the Kepler Telescope found that the light signal from that star dipped and dimmed a lot is they observed it in ways that suggested something complicated was periodically moving between the star and our telescopes, something like a swarm of comets, or maybe an alien structure.
Well, the SETI Institute has trained its Allen Telescope Array on the star for more than two weeks looking for two different kinds of radio signals, and there’s been silence. So far, the SETI Institute scientists say no evidence of deliberately produced radio signals can been found in the direction of KIC 8462852.
So we may have to look elsewhere for new friends. Oh well. At least there’s a new Star Wars movie coming out.