This site is not maintained. Click here for the new website of Richard Dawkins.

← Thinking ahead: Bacteria anticipate coming changes in their environment

Thinking ahead: Bacteria anticipate coming changes in their environment - Comments

riemann's Avatar Comment 1 by riemann

This is truely exciting and scientific way of thinking at its best. However i have one difficulty getting my head around. I fail to see how this new emergent adaptation is one that's "learned" by the bacteria, rather than a particular genetic variaton selected by artificial selection to adapt to the newly introduced feeding patterns.

"And sure enough, after a few thousand generations, an ecologically fit strain of microbe emerged which did exactly that. This happened for every pattern of cues that the researchers tried."
This sentence highly suggests to me that it's ordinary selection at work, not an undetermined (by genes that is) way of dealing with enviroment at large, or even one determined but with a loose leash. I am not even sure what anything less than a fully determined trait would mean for a bacteria. Surely it's not the individual bacterium that acquires the new trick, but its descandants? The best i can relate to the premise of this experiment is this: "Genetic variation to adapt for any feeding pattern exists almost readily for bacteria, and therefore the ones that have these variations tend to fare better than the rest of them." Right? But isn't this the very definition of natural selection? If so, what's the fuss all about? I am sure though it's me who's missing a point, rather than the researchers. Further elobarations would be much appreciated.

Thu, 19 Jun 2008 17:25:00 UTC | #186413

Greyman's Avatar Comment 2 by Greyman

It's about how the regulatory mechanisms controling the activation of other genes, and thus the cell's metabolic processes, behave as though they are anticiplating changes in their environment.

They have a network of trigger and responses so that the bacteria can respond to changes in one condition just in time to take best advantage of the opportunity presented by an another changing condition before it happens.

"Oh, it's getting hotter.  I better switch over to aerobic respiration so I'll be ready for the oxygen when it gets here."

More remarkable, these algorithyms change over generations.  The population learns to recognise when rising temperatures means they should switch to aerobic respiration, and when it means they should switch to anaerobic.

"Oh, it's getting hotter.  I'd better switch over to anerobic respiration so I'll be ready for when the oxygen level drops."

The process of selection is optiming the response triggers patterns over generations, enabling the bacteria population to learn.  The simulation they used demonstrates how this is possible without a designer reprogramming the cells.  The networks of trigger and responses evolve into working algorithyms; but the code they develop displays a lot of useless kruft.

Design without a designer.

Thu, 19 Jun 2008 20:53:00 UTC | #186467

jaytee_555's Avatar Comment 3 by jaytee_555

I had only read a half dozen lines of this article when I thought, hey!....what's all this 'thinking ahead', 'learning' and 'smart' stuff? This is all very interesting, for sure, but it is STILL basic natural selection.

I see "Riemann" came to the same conclusion, and hit the nail on the head by pointing out that it's not the individual bacterium that acquires the new trick, but its descendants.

Once again, it seems that the science is good, but the reporting is misleading.

Thu, 19 Jun 2008 21:08:00 UTC | #186473

mordacious1's Avatar Comment 4 by mordacious1

Hey, these little guys have survived for a wee bit of time now, and will be here when we are long gone. The fact that they are adaptable this way doesn't suprise me. Finding the evidence that they sense cues is good stuff.

I wonder why god made them this way? She sure thought ahead, since...oh, wait, she produced mankind and bacteria at the same time, forgot.

Thu, 19 Jun 2008 21:57:00 UTC | #186478

8teist's Avatar Comment 5 by 8teist

I enjoy reading your posts Mord 1. Maybe we are the bacteria and bacteria the intelligent lifeform,they are the ones that don`t need religion.

Thu, 19 Jun 2008 22:23:00 UTC | #186483

King of NH's Avatar Comment 6 by King of NH

This is interesting as a bacterial study, but also in how it uses terms like 'learn.' It's easy to say that bacteria can't learn without a brain, and that this is just natural selection. But then what is learning? Obviously, this is not one organism learning a new behavior in one lifetime. There is a difference there. But there is no special magical box in our brain that 'learns' and 'thinks.' I guess I'm very interested to know if our brain cells are using the same mechanism to learn that the bacteria is using, just with better cooperation and specialization of cells. Perhaps the generational, natural selection/learning of bacteria is very, very similar to what we mean by 'learn.'

Thu, 19 Jun 2008 23:06:00 UTC | #186489

mordacious1's Avatar Comment 7 by mordacious1


Well I know E. Coli don't need religion. There's enough crap where they live.

Thu, 19 Jun 2008 23:29:00 UTC | #186497

8teist's Avatar Comment 8 by 8teist

They live in heaven ?

Thu, 19 Jun 2008 23:37:00 UTC | #186498

mordacious1's Avatar Comment 9 by mordacious1

One can define heaven anyway they want, since it doesn't exist, and I'm sure the lower GI tract is Nirvana for them.

Thu, 19 Jun 2008 23:47:00 UTC | #186500

riemann's Avatar Comment 10 by riemann

Fair point, King of NH, but there are fundamental differences between how a bacterium regulates its behaviour and, for instance, we do. As in any aspect of nature, there are of course all sorts of intermediate stages from fully genetically determined bactaria behaviour to not-so-much genetically determined human behaviour, which Dan Dennett dubbed "The Tower of Generate-And-Test." You can check out what he means from the link below. Seen in this light, the difference between cognition or lack of it really, functionally matters. Therefore phrases like "thinking" and "learning" cannot be used as mere metaphors with vauge definitions. That's the reason i objected to the conclusions of the experiment. of generate-and-test

Fri, 20 Jun 2008 00:56:00 UTC | #186514

Telic's Avatar Comment 11 by Telic

To predict mealtimes accurately......

Pavlov's Bug experiment :D

Its certainly interesting, but I have to agree with the objections others have made about use of words like "thinking" and "learning".

This is the kind of inaccurate use of language that scientists should not indulge in; Just read the recently posted news item about Einstein...

Fri, 20 Jun 2008 01:56:00 UTC | #186541

Greyman's Avatar Comment 12 by Greyman

No.  Thinking and learning are the proper terms.

The bacteria have the collective ability to solve a problem and react with learned behaviour.  The solving and learning just does not take place on an individual level and they store the information in gene regulation networks rather than networks of neurons.

Using the Tower of Generate and Test definitions, riemann provided above, individual bacterium are first tier, or "Darwinian", creatures but over generations a population acts effectively like a single second tier, or "Skinnerian", creature.

Fri, 20 Jun 2008 02:21:00 UTC | #186551

He'sAVeryNaughtyBoy's Avatar Comment 13 by He'sAVeryNaughtyBoy

Thanks Greyman, I would have jumped to the conclusion that this was meerly natural selection in action but your last two posts explain the significance of this very well. Thanks.

Fri, 20 Jun 2008 02:51:00 UTC | #186563

kornyjorge's Avatar Comment 14 by kornyjorge

i for one welcome our new bacterial overlords.

Fri, 20 Jun 2008 13:51:00 UTC | #186950

mordacious1's Avatar Comment 15 by mordacious1

[rant]: If you've been reading these posts for awhile, you may realize that it is one of my pet peeves to attribute human atributes, like "deciding to grow a hard shell for protection", to other species. It's usually journalists translating science articles for the "laymen" and it pisses me off. Make the laymen more knowledgeable, not less. [Rant ended]

Fri, 20 Jun 2008 13:58:00 UTC | #186955

Thor'Ungal's Avatar Comment 16 by Thor'Ungal

i for one welcome our new bacterial overlords.

Welcome? they've been dominating the food chain for a while now. Where do you think the idea of black magic came from. A guy you wish harm to is walking along and starts to get progressively ill, no one stabbed him and he ate the same meal you did so you review what you did the previous day while thinking how much you'd like to stick the fellow.

Bacteria, can't live without them. Likely to kill you in the end. Almost certain to get at your corpse when you're done with it.

Aka above us in the food chain.

Oh, I too welcome our old bacterial overlords.


Fri, 20 Jun 2008 20:08:00 UTC | #187050

King of NH's Avatar Comment 17 by King of NH


Therefore phrases like "thinking" and "learning" cannot be used as mere metaphors with vauge definitions.

No, you're absolutely right, here. As a scholar, I can't stand arbitrary definitions. But here is the very base of the problem. What, exactly, does it mean to think? Where is that line? This is not a metacognition Descartian "wow man, that's deep" question. This is a serious question that needs to be answered before we can say that generational adaptation does not count. If we do say, to the end goal of defining thought, that such change over time does not count, then we have moved closer to finding where the first 'thought' originated and what it was (as in physical attributes, though I'm almost positive the first thought was something like 'I gotta pee').

Sat, 21 Jun 2008 23:11:00 UTC | #187431

Telic's Avatar Comment 18 by Telic

Thinking and learning are absolutely NOT the correct terms to use.

A clear case of muddy thinking from scientists who have read too much science fiction and have a vested interest in getting carried away with their own conclusions....

Comparing natural selection to a multi-generational 'genetically stored' algorithm is fine in one sense; But just confuses things and invites misunderstanding.

This is Natural selection 'solving' the problem at a hard-wired individual organism level over multiple generations of mutations; It doesn't involve thinking, or choice at the individual's software level.

Tue, 24 Jun 2008 01:51:00 UTC | #188437