A couple arrive at a fancy restaurant and they’re offered the wine list. This establishment only has two bottles on offer, one costing £5 and the other costing £25. The second bottle seems too expensive and the diners select the cheaper one. The next week, they return. Now, there’s a third bottle on the list but it’s a vintage, priced at a staggering £1,000. Suddenly, the £25 bottle doesn’t seem all that expensive, and this time, the diners choose it instead.
Businesses use this tactic all the time – an extremely expensive option is used to make mid-range ones suddenly seem like attractive buys. The strategy only works because humans like to compare our options, rather than paying attention to their absolute values. In the wine example, the existence of the third bottle shouldn’t matter – the £25 option costs the same amount either way, but in one scenario it looks like a rip-off and in another, it looks like a steal. The simple fact is that to us, a thing’s value depends on the things around it. Economists often refer to this as “irrational”.
But if that’s the case, we’re not alone in our folly. Other animals, from birds to bees, make choices in the same way. Now, Tanya Latty and Madeleine Beekman from the University of Sydney, have found the same style of decision-making in a creature that’s completely unlike any of these animals – the slime mould, Physarum polycephalum. It’s a single-celled, amoeba-like creature that doesn’t have a brain.
Physarum spends most of its life as a large mat called a ‘plasmodium’, which is a single cell that contains many nuclei. The plasmodium searches for food by moving along like an amoeba and sending out a network of tendrils. Its search patterns are very sophisticated for a brainless organism. A Japanese group found that if they placed the mould among food sources arranged like Tokyo’s urban centres, it created a network that closely resembled Tokyo’s actual railway system. The slimy network was optimised to transport nutrients to the main plasmodium.
Scientists have long since discovered that you can run simple decision-making experiments with Physarum by presenting it with several food sources and seeing how it behaves. Typically, the plasmodium touches all the potential meals and then either ‘decides’ to move towards one, or splits itself among many.
Latty and Beekman did one such test using two food sources – one containing 3% oatmeal and covered in darkness (known as 3D), and another with 5% oatmeal that was brightly lit (5L). Bright light easily damages Physarum, so it had to choose between a heftier but more irritating food source, and a smaller but more pleasant one. With no clear winner, it’s not surprising that the slime mould had no preference – it oozed towards each option just as often as the other.
But things changed when Latty and Beekman added a third option into the mix – a food source containing 1% oatmeal and shrouded in shadow (1D). This third alternative is clearly the inferior one, and Physarum had little time for it. However, its presence changed the mould’s attitude toward the previous two options. Now, 80% of the plasmodia headed towards the 3D source, while around 20% chose the brightly-lit 5L one.
These results strongly suggest that, like humans, Physarum doesn’t attach any intrinsic value to the options that are available to it. Instead, it compares its alternatives. Add something new into the mix, and its decisions change. The presence of the 1D option made the 3D one more attractive by comparison, even though the 3D and 5L alternatives were fundamentally unchanged.
This style of ‘comparative valuation’ may seem uncannily human, but it’s also one that’s shared by hummingbirds, starlings, honeybees and many other animals. In fact, Latty and Beekman think that it’s a “common feature of biological decision-making”. Certainly, it’s a much easier process – comparing two nearby options is less “computationally intensive” than making absolute judgments about each of them.
But how does Physarum make decisions at all without a brain? The answer is deceptively simple – it does so by committee. Every plasmodium is basically a big sac of fluid, where each part rhythmically contracts and expands, pushing the fluid inside back-and-forth. The rate of the contractions depends on what neighbouring parts of the sac are doing, and by the local environment. They happen faster when the plasmodium touches something attractive like food, and they slow down when repellent things like sunlight are nearby.
Despite being a single cell, each part of the plasmodium acts like a tiny individual, reacting to information from its environment. By combining these reactions, the entire plasmodium flows towards things it likes and away from things it doesn’t, all without a single conscious thought. It’s the ultimate in collective decision-making and it allows Physarum to perform remarkable feats of “intelligence”, including simulating Tokyo’s transport network, solving mazes, and even driving robots.
Reference: Proc Roy Soc B http://dx.doi.org/10.1098/rspb.2010.1045
Image by Johnna Gott
More on Physarum: Slime mould attacks simulates Tokyo rail network
If the citation link isn’t working, read why here
August 10th, 2010 at 10:07 pm
Slime mo[u]lds are just like Americans. They have an excuse, being physically brainless. What’s ours?
August 10th, 2010 at 10:16 pm
This is amazing. I didn’t think it would be that cool of a read, but I think I need to share this. I just think about the years of evolutionary selection that arrived at a creature with traits that let it act like this.
August 10th, 2010 at 11:23 pm
I will definitely have to use this in teaching Botany 130 next spring. Slime molds rock! Thanks, Ed!
August 11th, 2010 at 12:35 am
Great article. I’m not sure I understand the experiment depicted in the video, though. What was the conclusion?
August 11th, 2010 at 2:15 am
That’s funny. most comments marvel at how smart is mold instead of wondering why we are as stupid as slime … I wonder if there is another species with such a strong power of denial …
August 11th, 2010 at 2:26 am
It’s not just fancy restaurants that use this technique to modify our decision making – there’s an equivalent in political theory too, known as moving the Overton Window:
http://en.wikipedia.org/wiki/Overton_window
I guess this is more proof that politicians just treat us all like slime (molds).
August 11th, 2010 at 3:26 am
This all boils down to emergent complexity, which has got to be one of the most fascinating observations in the natural world. A single bee is just that, a single bee. But put many of them together, and you’ve suddenly got a swarm with a hive mind. A single neuron? Not much. Lots of interconnected neurons? A brain.
The same holds for this slime mold. Put lots of ‘decision makers’ together, and you’ll get amazingly cool complex behaviour, as these little guys show. More really is different. 2+2 does not equal 4. It doesn’t even equal 5. It equals apples! (from Kevin Kelly).
Pity the youtube video shows the hive mind failing / being lazy
.
August 11th, 2010 at 3:47 am
@lucas : well said.
This all shouldn’t come as much of a surprise considering we are all collections of brainless cells
August 11th, 2010 at 4:14 am
This makes me very, very happy.
August 11th, 2010 at 7:42 am
Nathan Myers, you’re a bigot.
August 11th, 2010 at 9:54 am
It is well known that even some materials have memory, why not this amoeba type of being? we have yet much to learn about our own home.
August 11th, 2010 at 10:54 am
This is fascinating. If it wasn’t so hot I would have more to say, but I’ll leave it at that. I think right now the slime mold has more thinking power.
August 11th, 2010 at 1:23 pm
> The simple fact is that to us, a thing’s value depends on the things around it. Economists often refer to this as “irrational”.
The weirdness of the concept that other life forms have a similar decision-making characteristic as humans is removed when one considers that ‘irrational’ is the wrong word to describe the decision.
‘Irrational’ implies that there is a flaw in the decision making process, which is not the case at all. In nature, it only makes sense to compare alternatives. The more information we have about our environment, the more we understand our situation, and therefore the effective absolute values of our options do change.
Just because we can artificially create a situation which fools the decision making process into choosing something different doesn’t mean that it’s broken. It just means that we’re not omniscient.
It shouldn’t surprise anyone that living organisms share certain algorithms that allow them to survive, one of which is the ability to adapt to the world around them when it changes. When the world changes to include a $1000 bottle of wine, it isn’t irrational or incorrect to take that into account, even if it’s the result of another person artificially trying to manipulate you. Yes, it will cause people to spend more than they would have otherwise, and yes, it is a less-than-honest tactic especially if the $1000 option only exists for that reason, but it doesn’t make the people WRONG. Tricking someone’s perception into making them choose something different doesn’t mean that their decision making skills are broken.
August 11th, 2010 at 2:22 pm
@Killgore9998 – Agreed. This is why I put irrational in quote marks. I suspect the study’s authors share a similar view.
August 11th, 2010 at 3:04 pm
Fascinating. It reminds me of the Overton Window idea, which I first encountered in
I wonder, computationally, what the slime moulds are doing to achieve this.
August 11th, 2010 at 3:05 pm
Sorry — URL of WWVB blog seems to have been considered illegal HTML.
Which I first encountered in
http://vannevar.blogspot.com/2009/07/rhetoric-golden-mean-overton-window.html
August 11th, 2010 at 3:07 pm
Sorry. that’s
which I first encountered in
http://vannevar.blogspot.com/2009/07/rhetoric-golden-mean-overton-window.html
August 11th, 2010 at 3:57 pm
Jeffimix (#10): No, I’m an American. Again, the slime molds have an excuse. What’s yours?
The title should have been ” Humans make decisions like brainless slime mould”.
August 11th, 2010 at 5:01 pm
I wonder if any of them would choose MAD?
August 11th, 2010 at 8:46 pm
Well written killgore…
August 11th, 2010 at 9:54 pm
Clearly these are decisions based on nothing more than “difference” which is apparently hard-coded into animals, is very easy to implement biologically and is “good enough” in terms of evolutionary survival. This makes sense because “difference” is the fundamental definition of information: a bit is defined as the difference between being either a 0 or a 1 which might logical or physical (as a voltage).
I also strongly agree with @ Killgore9998: the flaw is in presuming “rationality” is a reality. Cognitive science and experimental psychology are pretty clear about this: rationality is, at best, a non-unique emergent decision system that can arise in a brain. Mostly humans using heuristics for decision making which are often congruent with some degree of localized rationality but not rational in a formal predicate calculus sense.
August 13th, 2010 at 12:20 am
According to a recent Newsweek article (http://www.newsweek.com/2010/08/05/the-limits-of-reason.html), “failures of logic” are due to the need to win arguments. I never realized that slime molds are the stars of a debate team.
August 13th, 2010 at 11:51 am
To clarify, I also agree with Killgore9998 (#13) and J Gruszynski (#21). This is why I put irrational in quote marks. Could probably have made that clearer.
September 5th, 2010 at 6:42 pm
I just want to clarify something said in the article. Economists call that kind of decision-making process, when we compare our options with other, RATIONAL.
I mean people compare prices with other prices to see what is the relation between them. That is not irrational, it’s just what we do.