Investing Lessons From Slime Molds

During their latest episode of the VALUE: After Hours Podcast, Taylor, Carlisle, Forehand, and Carbonneau discussed Investing Lessons From Slime Molds. Here’s an excerpt from the episode:

Jake: Better do some vegetables, huh? All right. So, yeah, we wouldn’t want to miss out on this one either. We’re talking about slime molds. [laughs] God forbid that we miss that. So, this passage that we’re going to go is inspired by some work that Robert Sapolsky has done in this new book that he just came out called Determined. So, what is a slime mold to begin with? It’s like billions and zillions of these single-cell amoebas that join forces and to grow and spread like a carpet over a surface. They ooze around mindlessly in search of food. Maybe they’re not so mindless. The individual cells are interconnected by these tubules that can stretch and contract depending on the direction. Somehow this collection amoebas, without any apparent centralization, has this problem-solving capabilities that you just wouldn’t believe.

Researchers have done some really amazing experiments around this. So, here’s the setup. Imagine that you spritz a dollop of slime mold into this little plastic well, and it leads down to two different corridors. One of the corridors has a single oat flake in it, and the second corridor has two oat flakes. And apparently, slime molds love oat flakes for some reason. But similar to the hive mind like insect strategy of sending out scouts that bees and ants use, the slime mold expands into both corridors, and it reaches both of the food sources. But within a few hours, the slime mold reacts, and it retracts from the single oat flake corridor, and it heads to the one with two. How does it know? Like, how do all these things that–? It’s not like they’re talking to each other. Well, and if also, you could stick the slime mold into two different corridor mazes of differing lengths, and it ends up finding the shorter route. You could stick it into a maze with a bunch of dead ends and this brainless slime mold finds an optimal solution to its beloved oak flake.

Japanese researcher did an interesting study. He plopped a slime mold down into this strangely shaped like walled off area with oat flakes at very specific locations around it. And at first, the mold expanded and formed tubules connecting all the different food sources to each other in a bunch of multiple ways. It’s kind of a mess. Eventually though, the tubes retracted, and it ended up leaving close to the shortest path length connecting all the different food sources. Now, here’s where it gets interesting. The walls that this researcher put it in outlined the exact shape of the coastline around Tokyo. And the slime molds were deposited where Tokyo would be on the map. So, the oat flakes corresponded to the suburban train stations around Tokyo.

And out of this slime mold emerged a pattern of tubule linkages that were statistically similar to the actual train lines linking the stations that had been built. So, a slime mold, without a single neuron in it, had done the work of numerous urban planners.

Tobias: I’m just impressed that the humans got the same point that the slime molds did.

Jake: Well, I was going to make the joke that– I’m sure if we asked our friend, Moses Kagan, what he thought about LA city planners. He might say they also were operating without a single neuron.

[laughter]

Tobias: Sorry, stepped on it.

Jake: I know. All right. Actually, even in computer science, there’s this famous– It’s called the traveling salesman problem, and it’s like an optimization thing. It follows like if you’re given a list of cities and the distances between each pair of the cities, what’s the shortest possible route to visit each city once and return back to the origin city? Actually, Carl Menger, who’s the son of the famous Austrian economist, Karl Menger, was one of the first mathematicians to make real progress on the traveling salesman problem in the 1930s.

Anyway, so, how does the slide mold actually do it? Let’s get into that a little bit. It’s actually a three-step process, which mimics the ants and bees strategies. There’s scouts that go out, and that’s the slime mold oozing all over. And then there’s quality dependent broadcasting. And then rich get richer recruitment. So, let’s go back to our first version of the two corridors where one oat flake or two oat flakes. The slime mold will initially ooze into both corridors, and this is like the scouting phase.

And then when the food is found, the tubules contract in the direction of the food, pulling the rest of the slime toward it. And the better the food source, the greater the contractile force generated in the tubules. This is that quality dependent broadcasting that’s effectively a form of communication. And the tubules that are a bit farther away dissipate the force by contracting in the same direction and increasing the force of contraction and recruiting more behind them, basically. And eventually, it pulls the whole slime mold towards the optimal pathway.

I’ll spare you from going into all the gory details, but it turns out that the way that our neocortex wires itself is a very similar strategy to the slime mold. Your neurons will send out scouts to connect with other neurons, and they’re climbing along these things called like radial glia, and there’s these reinforcing mechanisms to attract other neurons to hook up where there’s better connections found. So, basically, like bees, ants, slime molds, your brain wiring, it all happens without a master plan or constituent parts really knowing anything beyond their own immediate neighborhood.

And then there’s one more little branching mechanism that I’d like to share, just because it’s so freaking wild to me. All right. in your circulatory system, each cell in your body is at most only a few cells away from a capillary. That’s where the blood feeds the nutrients, expels the waste, moves things around, like it’s the transportation system. Well, the circulatory system accomplishes this by growing around 48,000 miles of capillaries in the average adult. So, 48,000 miles worth of capillaries inside of you. And yet, that 48,000 miles only takes up about 3% of the volume of your body.

Science is freaking amazing, isn’t it? Or, nature, I guess is. So, anyway, there’s more than you wanted to know about slime molds. I don’t have any real investment takeaways from that other than just the emergence of solving things perhaps using simple systems can lead into much more complex behavior than you would ever imagine.

Jack: Well, I’m wondering if I need to replace myself with slime molds to build our multifactor strategies. Maybe I could align oats in such a way that it can select the factors or something. [Tobias laughs] I don’t know, it’s something I got to look at. Maybe I was worried about AI. Maybe I should really be worried about slime molds.

Jake: [laughs] The original AI.

Justin: I got some investing thoughts from that. I was thinking like, it got me momentum investing, like the initial scouts are the early guys in these stocks, and then they send the signals of the market that there’s an opportunity here, and then you’ve got the reinforcement of more investors coming in, which drives momentum. And then, I don’t know, I was trying to weave in some things there.

Jake: Yeah, that’s farther than I made it. That’s great.

Jack: Its great stuff though, Jake. You should have been a teacher, dude. Well, you are a teacher, but you could have been a teacher, [chuckles] science teacher.

You can find out more about the VALUE: After Hours Podcast here – VALUE: After Hours Podcast. You can also listen to the podcast on your favorite podcast platforms here:

Apple Podcasts