Friday Weird Science: Slime Molds take it to the highway.

Dec 28 2012 Published by under Friday Weird Science

We humans really like to think we are rational planners. In some examples, we are. Carefully designed modern suburban communities, with walking distances to stores and parks, etc. Require some careful planning. So do the stamped grids of many modern cities. But what about the routes between? You could always just take a straight line from A to B, if you wanted. But if you never accounted for the lay of the land, you could come up with some really unfortunate roads. So when it comes to laying highways and larger roads, many take advantage of roads that where there before them. Now a highway, once a dirt road, and before that, a track. Many transport networks therefore evolve organically, taking into account the lay of the land and things in between.

But how "organic" is it really? Is it really efficient? And can a slime mold tell the difference?

Yes. A slime mold. These fascinating creatures (I want one for a pet) have been featured here before laying out the train map of Tokyo.  But that's not the only city to receive the benefits of a slime mold's expertise!


slime mold

Adamatzky, et al. "Are motorways rational from slime mould's point of view?" arXiv:1203.2851v1, 2012.

The authors note the organic rise of roads, the first original tracks just the most efficient ways to get from A to B, with regard to difficulty of terrain. These tracks were bound to change as people started to transport large amounts of things (say, cattle) along them, demands change and the roadways do too. But do the networks always stay efficient? The authors bring in the slime mold.

Why slime mold? Well, it is easy to raise, doesn't take up a lot of room, doesn't eat a lot, and is easier to deal with than, say, ants. The slime mold, Physarum polycephalum, is a single cell with LOADS of nuclei. Usually a species of single cell organisms, the slime mold cells band together in times of scarcity. Then, it heads on the hunt. It searches for particles of food, and surrounds them, secreting an enzyme to digest the food. But if there's more than one source of food available, what's a mold to do? It forms a network of of protoplasm, with each strand connecting a food source, so it can suck up as much food as possible in the area.

And slime molds are very efficient. If you assume that the way the slime mold covers all its bases is, in fact, optimal, well, then you can use it to look at how optimal our transport networks really are.

All you need is some oats, an agar plate covered in the shape of your region of interest, and one very hungry slime mold.

The authors of this study wanted to compare 14 different regions: Australia, Africa, Belgium, Brazil, Canada, China, Germany, Iberia, Malaysia, Mexico, The Netherlands, the UK, and the USA. I'm not really sure why they picked so many areas of severely different size, because you'd think that for the big areas, you'd have different needs than for small.

But anyway, they set up the agar maps, and seeded the oats where all the major cities were. Observe:

Screen shot 2012-12-26 at 10.35.28 PM

The slime mold gets put on an oat, placed in the capital, and left to do its thing.

Screen shot 2012-12-26 at 10.36.06 PM

You can see here what the slime mold made and how it compares to the actual road map.

When the analysis was done, the best matches between existing roadmap and slime mold went to Malaysia, Italy, and Canada. The most fault-tolerant (with several alternate routes and no problems with jams) went to China, The Netherlands, Canada, and Italy. When all other measures of a good "road" were analyzed (cohesion, economy, extreme regions, average shortest paths, etc), the winners were Malaysia, Italy, and Canada. The worst were Mexico, Brazil, and the US. China, along with Belgium and Canada won a special prize for being the closest match for a slime mold (um, yay?).

But though China may win for a close slime mold match, this doesn't necessarily mean the network was the most economically or even the most distance efficient. It was just…the most like the slime mold. So the jury is still out as to whether or not we should put a slime mold on the problem when planning the next highway system.

Andrew Adamatzky, Selim Akl, Ramon Alonso-Sanz, Wesley van Dessel, Zuwairie Ibrahim, Andrew Ilachinski, Jeff Jones, Anne V. D. M. Kayem, Genaro J. Martinez, Pedro de Oliveira, Mikhail Prokopenko, Theresa Schubert, Peter Sloot, Emanuele Strano, & Xin-She Yang (2012). Are motorways rational from slime mould's point of view? arXiv arXiv: 1203.2851v1

2 responses so far

Leave a Reply