Trained Microbes!
by JoyYou've heard of a Flea Circus… now get ready for the Germ Circus!
Thinking Ahead: Bacteria Anticipate Coming Changes In Their Environment
LOL!!! Something a bit more than Shapiro's "cellular intelligence," researchers at Princeton have demonstrated some interesting intelligence in e.coli per anticipating future conditions and turning genes on or off based on that acquired knowledge.
In addition to shedding light on deep questions in biology, the findings could have many practical implications. They could help scientists understand how bacteria mutate to develop resistance to antibiotics. They may also help in developing specialized bacteria to perform useful tasks such as cleaning up environmental contamination.
Huh. An understanding of evolution as endogenous adaptive mutagenesis looks to "have many practical implications?" Who'd a thunk?
The researchers say that their findings open up many exciting avenues of research. They are planning to use similar methods to study how bacteria exchange genes with one another (horizontal gene transfer), how tissues and organs develop (morphogenesis), how viral infections spread, and other core problems in biology.
By golly, here we have actual biological scientists at an Ivy League institution and publishing in Science reporting that life anticipates the future at the most rudimentary level and adapts itself accordingly. Who was it who predicted years ago that science would eventually come to accept an EAM-ish version of intelligent design in biological evolution because it offers better solutions to 'problems' the RM-NS paradigm simply cannot explain?
Very cool.
June 20th, 2008 at 12:25 pm
Yeah, I know I'm being smug here. I've had a bit of a rough spring out in the real world, so it feels pretty good to 'win' something for a change.
The idea that cells use genes as tools for their survival, propagation, adaptation and evolution is NOT that scary. It's just the polar opposite of the idea that genes are "selfish" independent entities that use cells for their survival, propagation, adaptation and evolution.
So, Culture Warriors. What say you about this evidence of intelligent design in life and evolution? Is this science? Should the researchers be expelled from academia for heresy? Should the journal Science be taken to task and forced to issue a retraction?
Comment by Joy — June 20, 2008 @ 12:25 pm
June 20th, 2008 at 12:40 pm
Dawkins can include this in a revised publication.
But where's the evidence for angels and a deity?
Comment by Bradford — June 20, 2008 @ 12:40 pm
June 20th, 2008 at 1:00 pm
Bradford:
LOL! I do sometimes feel a bit left out around here when the critics start harping on theology and such, having been an EAM supporter from the first moment I encountered it at ARN, and had mturner flesh it out in the off-topic lounge. I don't have a deity in this race, since nothing science has ever said or ever can say about evolution could 'prove' anything about deities (and still be science). Religion/spirituality has no need for science. Science, on the other hand, has often been inspired by spiritual insight. It has always been thus.
Comment by Joy — June 20, 2008 @ 1:00 pm
June 20th, 2008 at 1:06 pm
Whoever claimed that bacteria could not adjust their regulatory networks over multiple generations to come up with such responses?
And IF this really is ID research, then it bears mentioning again that there is no discrimination against ID research, as long as actual science is being done.
Comment by hrun — June 20, 2008 @ 1:06 pm
June 20th, 2008 at 1:14 pm
http://genomics.princeton.edu/...
Comment by Rock — June 20, 2008 @ 1:14 pm
June 20th, 2008 at 1:16 pm
Did anybody here read the original Science paper? It has absolutely nothing to do with EAM. In fact, it is based on a model assuming random mutations (see quotes below the abstract).
Here's the abstract:
The authors used a modeling approach based on - guess what? - RM+NS.
Some quotes:
Comment by Raevmo — June 20, 2008 @ 1:16 pm
June 20th, 2008 at 1:18 pm
Joy wrote:
I don't know who made that prediction, but whoever it was must be pretty disappointed in the paper. Far from undermining RM + NS, it shows how RM + NS can produce predictive biochemical systems that don't depend on neurons!
Which makes this statement all the more puzzling:
Sorry you've had a rough spring, Joy, but what did you 'win'? I thought you were a skeptic of RM + NS, not a proponent!
Comment by robin — June 20, 2008 @ 1:18 pm
June 20th, 2008 at 1:24 pm
It's a fascinating paper, but where is the evidence for ID? To answer the rest of the questions: Yes. No. No.
Comment by Raevmo — June 20, 2008 @ 1:24 pm
June 20th, 2008 at 1:29 pm
Another quote from the paper:
So here we have a process of random evolution giving rise to a representation of the external environment.
This must be a huge blow to people like Bradford who believe that such representations cannot be produced except by an intelligent designer.
Comment by robin — June 20, 2008 @ 1:29 pm
June 20th, 2008 at 1:45 pm
Thanks for the pdf, Rock!
Raevmo:
Hahahaha!!! Assumptions and more assumptions. You know what they say about "assume" don't you?
So… including "dogma" in your experiment to preclude heretical findings means you've proven random causation? Funny. That never worked before, why do you think it works now?
I recall a few lengthy threads just in the last couple of weeks insisting that ID supporters drop the word "Intelligent" from their lexicon because it entails both anticipation of future conditions and the ability to learn. Here we have a double-edged research establishing that e.coli anticipates future conditions and learns from that information, altering its genetic expression suites and occasionally altering genes (have the sequences been done, or is this an assumption too?) to adapt to what it learns and expects.
Need I remind you that gene expression is epigenetic, not genetic? That EAM postulates all adaptive evolutionary developments begin via epigenesis, with gene duplications, expression markers, etc. only added to the equipment after the new phenome proves its adaptive worth?
Science is never going to call anything in the biological realm "Intelligent Design," nor is it ever going to appropriate any version of ID even if their end results look exactly like it. No one here expects it to, so you won't be disappointed that I am not surprised that they don't call it EAM.
Comment by Joy — June 20, 2008 @ 1:45 pm
June 20th, 2008 at 1:50 pm
Raevmo:
No doubt that's why they programmed the "central dogma" into the computer to prevent heretical results. Sometimes evolutionary biology sounds more like religion than religion.
Comment by Joy — June 20, 2008 @ 1:50 pm
June 20th, 2008 at 1:51 pm
How does this account for bacteria being able to "anticipate and prepare" for changes in their environment?
Comment by Doug — June 20, 2008 @ 1:51 pm
June 20th, 2008 at 1:58 pm
More precisely, we have a computer simulation of a random evolution process.
My oft stated position is that a causal trail retraced leads to purposeful intelligence at some point in the trail. There is no possibility of selected mutations until there is an elaborate self-replicating system in place, able to express, maintain and replicate, with great fidelity, a genome storing needed genetic information.
Comment by Bradford — June 20, 2008 @ 1:58 pm
June 20th, 2008 at 1:59 pm
Raevmo asked:
Obviously not. Worse still, it looks like they didn't even read the article that Joy linked to — or if they did read it, they didn't understand it.
Look at this quote:
So here we have a blind process of evolution producing networks capable of solving logic problems — and the networks were cobbled together in a distinctly undesignlike way.
As Raevmo asked: where is the evidence for ID?
Joy quotes the Science Daily article (emphasis hers):
I'm glad you emphasized that, Joy. Because of their success with their RM + NS approach, they plan to apply the same method to a wide spectrum of core problems in biology.
Cool! Thanks for bringing this paper to our attention, Joy!
Comment by robin — June 20, 2008 @ 1:59 pm
June 20th, 2008 at 2:10 pm
Joy:
I don't know, but I assume you will tell me soon. Anyway, the paper (and 34 pages of electronic appendix) shows very nicely how random mutations and selection based on the resulting mutant phenotypes can give rise to "intelligent" biochemical networks. The predictions based on those assumptions are confirmed with experimental evolution of E. coli. Seems like a slam-dunk for RM+NS. Grand of you to highlight this paper.
I think they don't call it EAM because they have never heard of it.
No, I think they programmed it to see if it can predict actual results. And it can apparently. So now you don't like the paper anymore? Now it's suddenly religion? How pathetic.
Comment by Raevmo — June 20, 2008 @ 2:10 pm
June 20th, 2008 at 2:10 pm
robin:
Are you telling me that more than 90 years' worth of an RM + NS approach didn't previously solve the "core problems in biology?" Hmmm… I wonder if there's a reason for that…
Note that the dogma didn't enter the picture until an intelligently designed computer program got involved, which the researchers thought absolutely necessary to explain away the evidence provided by the e.coli themselves. I wonder if there's a reason for that…
You're welcome.
Comment by Joy — June 20, 2008 @ 2:10 pm
June 20th, 2008 at 2:13 pm
Wrong again, Robin.
I read the complete article. Look at how those assumptions of yours just cloud so many things up.
That was a computer simulation, Robin.
Blind process of devising a program to run the simulation? Did YOU read the article?
Huh?
From the article:
I wish when we are debating the relevance of RM + NS robin would be here. It would be a nice contrast to see those saying "no! It's not RM +NS… that's creationist propaganda" with robin's "See!!!! RM + NS… ta da!!"
Comment by Doug — June 20, 2008 @ 2:13 pm
June 20th, 2008 at 2:15 pm
WOW.
Raevmo, who did your mind whip this one up?
Comment by Doug — June 20, 2008 @ 2:15 pm
June 20th, 2008 at 2:22 pm
Joy:
Yes, we know that science is conducted by intelligent humans that intelligently design theories. By this "reasoning" any explanation is an ID explanation. And why do you say they explain it "away"
Comment by Raevmo — June 20, 2008 @ 2:22 pm
June 20th, 2008 at 2:26 pm
robin is very much like the fundamentalist religionistas he pretends to be at odds with. In his world there is only room for strict literal interpretations. A perfectly fundamentalist POV.
Comment by Bradford — June 20, 2008 @ 2:26 pm
June 20th, 2008 at 2:27 pm
I haven't read the paper yet, but it seems they are making the same assumption present in nearly every evolutionary paper dealing with parameters:
"may alter any of the organism's parameters"
The problem is that in order for evolution to work, the "organism's parameters", specifically the ones which are necessary for the simulation, but both (a) exist, and (b) be easily evolvable. The choice of parameters is itself essential. This is why, in life as in computer science, evolutionary algorithms are used to _tune_ algorithms, or at most combine tested pieces of algorithms, not _create_ them. I have never seen an evolutionary algorithm devise its own parameters to tune! Yet that is what is required for naturalistic scenarios of evolution to work.
Comment by johnnyb — June 20, 2008 @ 2:27 pm
June 20th, 2008 at 2:30 pm
Joy wrote:
Joy,
You really should read the paper before commenting further.
This wasn't just a computer study — they experimented with live E. coli populations.
Doug wrote:
Holy crap, Doug! The final three pages of the article deal with experiments on live E. coli strains. The entire article is only five pages long.
If you read the article, it means that you failed to notice that over half of the article was not about simulations, it was about real bacteria!
Comment by robin — June 20, 2008 @ 2:30 pm
June 20th, 2008 at 2:37 pm
Settle…….
Did I say the whole article was about simulations? No.
I was refering to your comment that stated:
"Blind process", "undesignlike way" "cobbled together"…. your buzz words were derived from their simulation.
Nope, just means you still fail to have a clear understanding what we're talking about. But even worse, because this was something YOU were talking about.
Let me help, robin:
again,
and again,
Comment by Doug — June 20, 2008 @ 2:37 pm
June 20th, 2008 at 3:04 pm
I printed off the article and it only came to three pages.
Regardless;
In the 1st sentence of the 3rd last paragraph, "When the researches examined a number of fit virtual bugs, they could at first make little sense out of them"
Robin yelled:
Examining a virtual bug is not the same as doing experiments on live e.coli strains. Because, uhh…well,…. virtual and actual aren't the same.
Comment by Doug — June 20, 2008 @ 3:04 pm
June 20th, 2008 at 3:55 pm
I'm reading the paper. There on the first page dealing with the computer simulation designed to 'explain' the behavior of real, live bacteria, a bolded heading states:
Emergence of predictive behavior in simulated biochemical networks
"Emergence?" These folks are fairly desperate, seems to me. The programmed randomness in the simulation -
Huh. I can change my organismal expression parameters just by thinking about an ice cream sundae. That's NOT a "genetic mutation," it's a transcription-rate change for a gene providing the biochemical process-stimulator (or repressor) associated with my evoked physiological response to the thought. I hate it when they twist jargon so hopelessly that the truth gets buried in a crypt.
Yet in the lab half of this research, the responses of the actual living organisms was found to be related to the previously well-known thermal shock response - which has been noted to have quite specific effects per genetic changes. It's not like these beasties are mutating their genes when they change environments, they're changing their expression profiles (as established in this experiment via microarray). Upregulating and downregulating biochemical processes, not mutating genes into whole new protein-templates. A little like me and my thoughts, changing my physiological state in anticipation of that ice cream sundae I just ordered…
Though perhaps they'll someday get around to actual sequencing and determine that gut-living e.coli are a whole different species than soil-living e.coli. I'll await that research, then ask how the e.coli know to mutate once they get in the gut, and how to de-mutate when they end up back in the soil. Or, as this experiment demonstrated so well, how they know to mutate BEFORE they get into guts or back to the soil.
See, I'm not convinced that what's going on here is genetic mutation. I'd have to see the actual list of SNPs, breaks, substitutions, duplications, etc., etc. before I'd buy that, and then I'd have to ask why that specific set of mutations comes and goes on-demand. "Random" doesn't do anything on-demand.
Though I suspect the issue of confusion lies primarily in yet another assumption made by the computer modelers. That the 'intelligent' behavior must be a product of biochemistry exclusively because e.coli don't have central nervous systems. Whereas it is my strong suspicion that all cellular life forms have central nervous systems, there's just a spectrum of organization and expression in these physical structures that gives us a spectrum of expression.
E.coli do have a cytoskeleton (and ring CPU) made of elements homologous to all the elements in the eukaryote cytoskeleton. Thus e.coli can process information and act on the information learned. This is absolutely UNsurprising to me, but then, I have not invested any metaphysical faith in the idea that living organisms are hapless vehicles for "selfish" genes, or that the causes of biological processes are "random."
It is up to those who have invested metaphysical faith in random causation to establish that causation is actually random in living processes. The ability to anticipate future events upon information from the environment that is processed to provide knowledge the organism learns and responds to does not look random to me. Intelligently designing a computer program to try and explain away the actual findings as random doesn't impress me very much, but it obviously impressed the gatekeepers at Science enough to get this research into print. Looks like these researchers have found a nifty way to game the system.
I now predict it won't be long before scientists make science entirely obsolete (along with themselves) by "proving" with an intelligently designed computer program that intelligence (foresight, learning, etc.) is just an illusory effect of random causation. They can then all be put peacefully out to pasture with the rest of the cows, while the rest of us apply our actually existent intelligence to the many challenges of life on planet earth in the 21st century.
No one will miss them.
Comment by Joy — June 20, 2008 @ 3:55 pm
June 20th, 2008 at 4:02 pm
Doug wrote:
Doug,
I'm talking about the article in Science, not the article in Science Daily.
Now what were you saying about assumptions?
Ah, yes, here it is:
Heh.
Comment by robin — June 20, 2008 @ 4:02 pm
June 20th, 2008 at 4:09 pm
Joy,
Intelligence is real, even if it is an effect of what you call "random causation."
I'm not aware of any scientists who believe that intelligence is illusory. Could you name a few, and provide citations if you have them?
Comment by robin — June 20, 2008 @ 4:09 pm
June 20th, 2008 at 4:12 pm
Robin posted:
Even when you explicitly state the site you got your quote from, you still don't know that's what we are talking about? How do you pull that one off.
I'm supposed to know when you hop from one reference to the other?
Maybe you shouldn't quote from Science Daily and then follow it with this:
Comment by Doug — June 20, 2008 @ 4:12 pm
June 20th, 2008 at 4:20 pm
robin:
And what is "intelligence?" Are e.coli bacteria "intelligent?" Does their adaptive response to environmental cues via foresight involve "intelligence?" Is adaptive evolution then "intelligent?" Is adaptive evolution "design?"
If consciousness is illusion, then all attributes of consciousness are illusory. Including "intelligence."
Comment by Joy — June 20, 2008 @ 4:20 pm
June 20th, 2008 at 4:28 pm
Joy wrote:
Joy,
My question was:
Comment by robin — June 20, 2008 @ 4:28 pm
June 20th, 2008 at 5:01 pm
Do you mean scientist in the very broad sense to include philosophers?
It would appear odd a scientist who believed that intelligence or knowledge was an illusion.
Comment by Doug — June 20, 2008 @ 5:01 pm
June 20th, 2008 at 5:09 pm
Doug wrote:
Doug,
Why are you fixated on this?
You assumed I was talking about one article when I was talking about the other. You made a mistake. So what? Why dwell on this? You're focusing on something has no bearing on the significance of the research.
You did the same thing in this comment on the ID and Morality thread. You went on to say:
That comment, and your behavior in general, are starting to make me wonder whether you want to catch me in a mistake — preferably a stupid one — because you think that would give you some kind of license to dismiss or downplay my arguments without having to refute them.
Whether or not that's what's going on here, it should be obvious that our arguments stand or fall on their own merits. Let's return to focusing on the arguments, and not on which article the word "article" refers to.
Comment by robin — June 20, 2008 @ 5:09 pm
June 20th, 2008 at 5:16 pm
Doug asked:
I'm referring to Joy's statement, which is about scientists:
Doug:
I agree, which is why I'm interested in hearing who Joy is talking about, specifically.
Comment by robin — June 20, 2008 @ 5:16 pm
June 20th, 2008 at 5:18 pm
Agreed.
Sorry for doing that.
Comment by Doug — June 20, 2008 @ 5:18 pm
June 20th, 2008 at 5:49 pm
So can someone confirm or deny that I understand the claims made in the research? Here's my understanding of the paper:
They created a computer model to determine if "RM+NS" might possibly explain the observed data.
1) They observe that actual bacteria can predict when the external environment is about to change.
2) The prediction is based on the change in temperature and predicts that a change in oxygen levels will follow.
3) In response to change in temperature the bacteria turns on or off certain genes in the cell to optimize itself for the oxygen levels.
4) The mechanism that turns "change in temperature" into "turn genes on or off" is encoded in the DNA of the bacteria.
5) They put the bacteria in a new environment that swaps the normal relationship between temperature and oxygen levels.
6) After a few hundred generations the bacteria begin to predict the new reversed oxygen-temperature relationship.
7) This improved ability shows an inheritable change to the bacteria population. A single bacteria cannot learn the new pattern, but variation in future generations can make some members more or less suitable to that environment.
9) The computer model exhibits results consistent with the actual bacteria.
Given my boiled down understanding of the study I don't understand what the connection to ID is. It seems like yet another interesting study supporting current mainstream theories.
Comment by Todd Berkebile — June 20, 2008 @ 5:49 pm
June 20th, 2008 at 5:56 pm
Joy:
No Joy, there's no burying of the truth here. In the model the mutations really were (random) genetic mutations. The parameters that determined the properties of the biochemical network were under genetic control. And the experiments with real bacteria supported the predictions of the model. I understand you are a little upset that the paper you thought supported your favorite theory turned out to do the exact opposite. But that's business as usual in science.
Comment by Raevmo — June 20, 2008 @ 5:56 pm
June 20th, 2008 at 6:01 pm
Todd, you summarize it quite well. I'm still waiting for an explanation for why this study supports ID. So far the only "explanation" has been that the theoretical model predicting the experimental outcome was designed by intelligent beings. Duh.
Comment by Raevmo — June 20, 2008 @ 6:01 pm
June 20th, 2008 at 6:07 pm
robin:
Oh, come on robin! You can do that search as easily as I can, since you've got a computer too. As a participant in the multidisciplinary quest for consciousness pretty much since its inception, I've encountered lots of scientists who consider consciousness an illusion. Admittedly most of those are neuro-types and cog-sci guys, the ai-guys are seriously looking for a real definition of a real phenomenon or they wouldn't be funding the quest. And of course there are many philosophers who are so uncomfortable with the subject that they'll insist it's non-existent. But who pays any real attention to philosophers?
First page of returns (1 - 10 of 509,000 on terms "consciousness as illusion") -
Susan Blackmore
The Biggest Ideas - You're An Illusion
Michael Tye, "Phenomenal Consciousness: the Explanatory Gap…"
Salon: Consciousness dethroned [review]
Amazon: The User Illusion: Cutting Consciousness Down to Size
New Scientist: The grand illusion
Behaviorism of human mind and brain
There are dozens of scholarly articles available in full (pdf) from the Journal of Consciousness Studies, scroll down to "Full Text of Selected Articles" and right-click on the titled in blue, select "download file" from the drop-down menu, and that should do it. You may have to open it in a pdf reader when it's done.
Should I go to page 2, or can you take it from here? It's really quite the lively ongoing debate, this question of whether or not consciousness is worth investigating, or if it's just a grand illusion.
But that's not the subject here. The subject here is foresight and adaptive evolution in response to knowledge gained (learned!) from the environment, in e.coli.
Comment by Joy — June 20, 2008 @ 6:07 pm
June 20th, 2008 at 6:20 pm
Raevmo:
No, Ravemo. Zero genetic mutations were documented in the e.coli at issue - ignoring the pseudo-bugs the computer was programmed to model, which is irrelevant to what was found in the real organisms. For the real organisms, 'changes' were expression-related 100%, as determined by microarray - which profiles gene expression, not gene structure or position in the genome.
This is why the computer model was programmed to insert "random" changes to parameters of expression, not genes, junk DNA, jumping elements, gene number or chromosome number/arrangement. Changes in gene expression in response to hunger, thirst, heat, cold, bad smells, immediate danger or thoughts of ice cream are not random and involve zero genetic modifications. There will be chromatin dynamics changes to present certain genes for transcription, and biochemical triggers to speed up or stop transcription, and associated biochemical cascades enhancing or depressing effects.
None of this involves evolutionary mutation of genomes. It's endogenous response to environmental information processed and learned from enough to enable foresight. i.e., it's dinnertime in half an hour, boy that lasagna smells great, my mouth is watering and I'm starving all of a sudden, so I'd better grab a couple of carrot sticks to tide me over…
Comment by Joy — June 20, 2008 @ 6:20 pm
June 20th, 2008 at 6:24 pm
That is a very different claim than the one you originally made, which was:
You proposed a connection between intelligence and consciousness which I assume is how you are justifying this huge leap, you said:
This is faulty logic, however, and does not justify the leap from your original statement to your revised statement. The fault is that you assume "intelligence" is merely an attribute of "consciousness." Intelligence could be unrelated to consciousness or the relationship could work in the opposite direction. In other words even if consciousness is an illusion that doesn't imply intelligence is an illusion.
Comment by Todd Berkebile — June 20, 2008 @ 6:24 pm
June 20th, 2008 at 6:33 pm
Sure, but if you are thinking about Ice Cream at the moment you conceive a child that won't make your child more prone to crave ice cream. The bacteria experienced an inheritable change. There is more inheritable data in the cell than just DNA alone, but that doesn't change the RM+NS aspect of the study.
No learning behavior was observed in a single bacteria cell, learning was only observed from one generation to the next. You seem to be claiming that the bacteria have an adaptive learning ability that is far beyond what is suggested by this study. The fact that the RM+NS model fits the observation doesn't disprove the sort of purposeful change you are suggesting, but it certainly doesn't support it either.
Comment by Todd Berkebile — June 20, 2008 @ 6:33 pm
June 20th, 2008 at 6:34 pm
Todd B.:
Okay, I can accept that separate designation if you can define it and apply it reasonably. So I'll ask you what I asked robin…
"¢ What is "intelligence?" (definition please)
"¢ Are e.coli bacteria "intelligent?" (per this experiment)
"¢ Does their adaptive response to environmental cues via foresight involve "intelligence?"
"¢ Is adaptive evolution then "intelligent?"
"¢ Is adaptation "design?"
Comment by Joy — June 20, 2008 @ 6:34 pm
June 20th, 2008 at 6:36 pm
And how do you suggest these changes in gene expressions come about?
I wonder, if there is a follow-up study that shows that the basis of these changes is indeed genetic, what will your next argument be? My guess is that it will be: The bacteria learned and then stored that knowledge in their DNA.
Comment by hrun — June 20, 2008 @ 6:36 pm
June 20th, 2008 at 6:38 pm
So you claim that scientists will claim that intelligence 'is just an illusory effect of random causation'. And when questioned for any support for you weird prediction, you want others to define intelligence?
Comment by hrun — June 20, 2008 @ 6:38 pm
June 20th, 2008 at 6:51 pm
Todd B.:
Epigenetic developments are heritable. Even better, some rather spectacular epigenetic expression suites seem to be universal in high-stress situations, even though they appear to have no evolutionary just-so story to account for them. That was quite a nifty surprise.
Thanks for admitting that heritable data isn't reserved to DNA. This has become so evident in the last few decades that it simply cannot be ignored. Evolution is about a lot more than just random mutation to genes/genomes sifted by ruthless physical necessity. I thought everybody knew that by now. Our newcomer appears to be a bit behind the curve on that. Maybe he'll educate himself.
As for the RM+NS "aspect" of the study, it doesn't say anything about the actual nature of the adaptive foresight displayed by these 'trained' bacteria. It's just another toy at Radio Shack that's designed to drive parents crazy.
This is perfectly consistent with EAM. It's a systems view, with a touch of neo-Lamarckism tossed in. If real genetic changes occur in evolution, they occur when a stress-response expression change gets written-in. After it proves useful or opens new opportunities. None of this is directed by top-down 'higher' consciousness, but it does display a degree of awareness and behavioral degrees of freedom by the system itself. Not a single whit of which can be accomplished by a dead cell, even though all the pieces-parts are present and arranged just-so.
[caveat: "random" mutation does occur, by a number of endogenous and exogenous causes. These are the causes of disease and susceptibility to disease primarily, per EAM]
Purposeful agency exists in the living (and functional artifacts designed and built by the living). Dead matter does nothing purposeful on its own.
Comment by Joy — June 20, 2008 @ 6:51 pm
June 20th, 2008 at 6:54 pm
hrun:
Hahaha!!! And exactly HOW do they do that, hrun? Mechanisms please…
Looks like I'll make an EAM'er of you yet! §;o)
Comment by Joy — June 20, 2008 @ 6:54 pm
June 20th, 2008 at 7:04 pm
Joy, I don't know what the laughter is about. I was asking what YOUR response would be, and the above was my guess at your response. So no, it's not me who would be the EAM'er. I simply wanted to gauge would your response will be to the inevitable follow-up study.
And since you are asking about mechanisms: What do YOU think is the mechanism by which the expression pattern in the bacteria changed, if not through associated genetic changes?
Comment by hrun — June 20, 2008 @ 7:04 pm
June 20th, 2008 at 7:17 pm
hrun:
There's a better experiment with e.coli I'm anxiously awaiting the sequencing follow-up on, the one where Lenski got to play with a metabolic development in one out of lots of 'lines' he's stored for years, which looks to be a real way to trace complex trait development through prior (apparently unnoticeable) changes. But I'll take whatever these guys care to throw. Microarray is enough to establish expression changes. It does not establish genetic structural changes at all.
Associated epigenetic changes.
Comment by Joy — June 20, 2008 @ 7:17 pm
June 20th, 2008 at 7:24 pm
Joy,
I will gladly answer your questions:
"¢ What is "intelligence?" (definition please)
I tend to see intelligence as a collection of various behavioral abilities such as problem solving, foresight, memory, learning, and creativity. I don't think its a term that has a useful scientific meaning, scientists studying intelligence tend to focus on specific behaviors as they are more objective.
"¢ Are e.coli bacteria "intelligent?" (per this experiment)
A single bacteria is completely limited to deterministic (although extremely complex) stimulus-response. Still, you could claim the entire bacteria represents a long genetic memory. It also has some problem solving abilities built into it, but it cannot learn or adapt. So you can certainly say that it shares traits with things we would call intelligent. Over multiple generations the bacteria have been shown to learn, so you could make the argument that the species (or perhaps the process of evolution itself) is "intelligent." Since I don't think the term intelligent has any objective scientific meaning such an argument would be metaphysical in nature.
"¢ Does their adaptive response to environmental cues via foresight involve "intelligence?"
The mechanism described in this research seems to be completely stimulus-response so it could be argued that the term foresight doesn't even apply. Change in temperature triggers change in gene expression. They use the term foresight because the stimulus being detected (temperature) is unrelated to the condition being optimized (oxygen levels) by the response. This does not imply a non-deterministic system in any way, so its not "foresight" meaning "deliberate choice" but simply foresight meaning "ability to predict."
"¢ Is adaptive evolution then "intelligent?"
Again, one could certainly make a metaphysical argument that it is. That argument seems off-topic for this thread, imho.
"¢ Is adaptation "design?"
I think the term "design" is perfectly applicable to both telic and atelic artifacts so I'd say yes, but again this is simply a subjective position so I don't see what it really tells us about the objective world.
"Awareness," sure, but what indication is there that this awareness requires anything more than chemistry and physics to be explained? For example, an RTK receptor is "aware" of extra-cellular events, do you consider that it to be purposeful all by itself? Your usage of awareness seems to simply mean "chemically active." I'm not sure how "alive" verse "dead" makes any difference in this scenario either. I can freeze a bacteria and in its frozen state it can no longer respond, purposefully or otherwise, to anything. It can no longer consume or reproduce. Is it dead? Once I thaw it back out it certain becomes alive again. Cells can be in many states between "fully functional" and "completely dead", what is the cut-off point for calling it alive?
Comment by Todd Berkebile — June 20, 2008 @ 7:24 pm
June 20th, 2008 at 8:15 pm
Todd B.:
First, thanks for answering those questions.
Then, I'd agree that at this point in time there is no definitive outline of what it means to be 'intelligent'. That's why the long threads with aiguy awhile back. Henry Stapp predicted to us mere students back in '00 that it would be 300 years before we had a sound scientific definition of consciousness. Intelligence is one of those "I'll know it if I see it" things (like pornography!?).
Still, some adjectives do appear in this paper that strongly suggest some sort of intelligent activity by the system, in so far as "intelligence" can be attributed to a system. I found it very refreshing.
I'd have to disagree, based on this research. Though admittedly bacterial generations aren't much like generations when you get all the way to mammals. I've seen people assert that bacterial life is essentially "immortal" because there's never a progenitor generation - it all goes into the next (exponential) generation via direct division.
It looks to me like the adjectives at work are those that suggest attributes of "intelligence" in the process.
Ability of WHO and WHAT "to predict?"
Nice glissade.
And what is the "artifact" here? The e.coli or the computer program?
I'd suspect it's when exterior forces/agents take over and interior force/agency no longer computes. But my experiences of death are entirely in the multi-cellular realm. What happens to the unseeable critters when I wash my hands with antibacterial soap isn't of enough concern for me to have investigated it much. §;o)
Comment by Joy — June 20, 2008 @ 8:15 pm
June 20th, 2008 at 9:12 pm
To be clear I mean a single e.coli cell has not been show to possess the ability to learn or adapt within its own life span, which I will define as the span between the binary fission that created the cell and the next binary fission to occur or the cell's death. Clearly learning has been demonstrated across multiple generations of e.coli. If this learning is the result of a genetic mutation and that mutation does not occur during binary fission and that mutation is still able to be expressed in the mature cell then you could claim that an existing e.coli cell "learned" as a result. You are also correct that you could define either copy of the cell as "being" the original cell and thus claim the original cell is immortal and has learned. My intent was to draw a distinction these boundary cases that separated out "generational" changes, but I agree that such a boundary is fairly arbitrary. This fairly arbitrary boundary of "self" verse "offspring" in bacteria seems potentially important to the EAM claim but is largely irrelevant to mainstream genetic views.
In this particular example e.coli to predict oxygen levels.
Trust me, I'm no ballerina, more the bull in a china shop sort ;). Its your thread, I'll gladly discuss the metaphysics of whether evolution can be considered intelligent if you want. Short version, my current thinking leads me to say, "Yes, the naturalist view of evolution can be considered intelligent in many of the same ways a human is considered intelligent."
I would say the word design applies to both of those artifacts even though I believe one to be telic and the other atelic.
Comment by Todd Berkebile — June 20, 2008 @ 9:12 pm
June 20th, 2008 at 9:25 pm
Joy:
Robin:
Joy:
Robin:
Doug:
Robin:
Joy:
She then proceeds to do a different search altogether:
Joy, do you realize how much time you could have saved us if you had answered my original question honestly and told us that you had no evidence to support your prediction and that you couldn't name a single scientist who conformed to your stereotype?
Comment by robin — June 20, 2008 @ 9:25 pm
June 20th, 2008 at 9:41 pm
I'm a little puzzled by anyone who would assert that consciousness is an illusion, for I find the statement to be meaningless. Illusions have analogs that are real or at least have real meaning. A stage magician floating in the air may be an illusion, but it's an analog of something really floating. If consciousness is an illusion, what is it an illusion of?
BTW, Joy, this is a fun thread, and I like your style.
Comment by kornbelt888 — June 20, 2008 @ 9:41 pm
June 20th, 2008 at 10:18 pm
An illusion as you use above is a distortion of a sensory perception. It is possible that our perception of consciousness is precisely that and nothing more. Where as magic tricks are distortions of our exteroceptive senses consciousness could be a distortion of our interoceptive senses (equilibrioception, proprioception, and/or nociception). So even interpreted literally you are wrong to jump to the conclusion that the statement is meaningless.
But I think in this case they mean 'illusion' in the sense of, "something many people believe that is false."
"Concousness is what it feels like to have a neocortex." - Jeff Hawkins
PS: I wrote this post simply because I wanted a chance to use a lot of really big words ;).
Comment by Todd Berkebile — June 20, 2008 @ 10:18 pm
June 20th, 2008 at 10:53 pm
kornbelt888:
"Meaning" has no meaning if consciousness is illusory. We're all zombies.
Another illusion.
Thanks!
Comment by Joy — June 20, 2008 @ 10:53 pm
June 20th, 2008 at 11:18 pm
Todd Berkebile wrote:
Well put, Todd.
Of all the topics in the philosophy of mind, eliminativism is perhaps the one most severely distorted by uninformed critics.
When an eliminativist asserts that consciousness is an illusion, she is not denying that the illusion is experienced. That would be nonsensical. She is simply saying that consciousness is not what we think it is.
Joy's take on this reveals her misunderstanding of this point:
Comment by robin — June 20, 2008 @ 11:18 pm
June 20th, 2008 at 11:21 pm
Todd B,
Seems wrong. I'd suggest,
Ah. Better, for my money. A little eastern, but nothing wrong with that now and then.
As for the topic-at-hand, I did notice the paper qualified random once as "essentially random (unpredictable)". Whether or not they intend that qualification to hold throughout, I just hold that out as somewhat interesting.
Then again, I'm the guy who thinks it's all design either way, so hey.
Comment by nullasalus — June 20, 2008 @ 11:21 pm
June 20th, 2008 at 11:47 pm
robin:
I am misunderstanding none of it, my dear. I can argue for or against any of the models, 'real' or 'unreal', from JJ McFadden's EM Field to Penrose-Hameroff's Orch-OR, all the way through Matti Pitkaanen's TGD. But it's entirely unnecessary to go all the way to eight dimensions and multi-sheeted spacetimes and massless extremals that look a lot like monopoles (or gravity) using p-adic primes and a hierarchy of 'selves' to get around the singularities because that's not the subject of this thread. It's just an expansion (of some interest) of the terminology in this particular research.
Having established the fact that we don't have 'scientific' definitions for these conceptual representations of basal phenomena, the very question of ID vs. no ID becomes very blurry. As it is in reality, come to think of it (because I can).
How dumb does it seem to you that there's a whole ongoing 'Culture War' about whether or not evolution is random or designed? It sure seems mighty darned dumb to me. No matter what science tells us, you will believe about life what you already believe about life. So will I. That's because our beliefs about life are metaphysically based, not the products of any 'proof' biological science has ever offered or can ever defend against experientially-informed belief.
Mayhaps someone will start a new rabbit thread (open thread) soon, metaphysics is fair game on those. This thread is about the research reported in the OP and published in Science (accessible as pdf from Rock's link). And about what are or are not fair conclusions to be drawn from it on the issue of 'cellular intelligence' as seemingly displayed by the biological subjects.
As I said, prokaryotes do indeed have cytoskeletons analogous to eukaryotes, using protein homologous to all those involved in eukaryote information processing (etc.). Hameroff's model looks better and better the more information comes in from the fields. I presume because we are physical that there are physical correlates of consciousness [PCCs]. Identifying those is a large step toward understanding.
Given the large degree of controversy surrounding that quest at the present time, it is entirely premature to rule out anything in the realm of possible intelligent design in evolution of life here on planet earth. IMO.
I'm not a zombie. Are you?
Comment by Joy — June 20, 2008 @ 11:47 pm
June 21st, 2008 at 12:26 am
Joy,
What does any of that have to do with your misunderstanding of what it means to be an eliminativist about consciousness?
Comment by robin — June 21, 2008 @ 12:26 am
June 21st, 2008 at 12:52 am
nullasalus wrote:
The context makes it clear what they mean:
Their point is that even a deterministic perturbation may appear to be random if the cell can't sense or track the underlying pattern. In other words, a deterministic perturbation can be "essentially random (unpredictable)."
Comment by robin — June 21, 2008 @ 12:52 am
June 21st, 2008 at 1:02 am
robin,
I know. I believe Joy would call it FAPP random. It just means that 'random' loses a whole lot of its bite on a metaphysical level. One of the most subtle distinctions out there. Everyone should know about it.
Comment by nullasalus — June 21, 2008 @ 1:02 am
June 21st, 2008 at 1:30 am
Not really. Genuine randomness still comes into the evolutionary picture via the subset of genetic mutations that are precipitated by quantum events.
Anyway, even deterministic mutations can be legitimately modeled as random as long as the output of the mutagenic process isn't influenced by the fitness of the resulting mutations.
If they are "essentially random", that is enough. Darwinian evolution is possible in a perfectly deterministic universe.
Comment by robin — June 21, 2008 @ 1:30 am
June 21st, 2008 at 1:43 am
robin,
"Genuine" is the stuff of philosophy with regards to randomness, even when the quantum is in play - determinism/indeterminism (and the potentially in-between?) is no more settled there than anywhere. Back to FAPP random.
I'm no opponent of evolution. It's one more tool in the Designer's set for my money. I'm not arguing that they can't be legitimately modelled as FAPP random either - as I said, just important to realize one's metaphysics. I enjoyed seeing the brief distinguishing attempt in this article. It's the sort of thing that should be in textbooks.
Comment by nullasalus — June 21, 2008 @ 1:43 am
June 21st, 2008 at 9:38 am
Associative learning.
Pavlov's bacteria.
Comment by Zachriel — June 21, 2008 @ 9:38 am
June 21st, 2008 at 9:42 am
nullasalus wrote:
Classical randomness usually reflects our lack of (detailed) knowledge about a physical system: for example, in chaotic systems a small uncertainty about coordinates and momenta grows exponentially with time, but one requires that initial uncertainty or computational errors along the way to generate randomness.
Quantum randomness is entirely different. We can know everything there is to know about a physical system and yet quantum measurements will generate random outcomes under appropriate conditions. Furthermore, physicists have been able to rule out experimentally a large class of theories where this uncertainty of the outcome was generated through some hidden classical variables that make the observer ignorant of the situation. In other words, not only the observer cannot predict the outcome, nobody can, even in principle. That's the current understanding of quantum randomness. It seems to be as close to "genuine" randomness as it gets.
At any rate, it doesn't matter whether randomness is "genuine" or just comes from ignorance. We can simulate a physical system in thermal equilibrium using a generator of pseudorandom numbers and the thermodynamics comes out right. Likewise, it doesn't matter whether mutations are truly random or only practically random, unless of course you want to say that the creator supplies carefully calculated signal disguised as random noise to guide mutations where he wants. While such a possibility can never be ruled out, science shouldn't—and won't—waste its resources on doing that, we have theology and philosophy that can take care of such things.
Comment by olegt — June 21, 2008 @ 9:42 am
June 21st, 2008 at 12:03 pm
nullasalus wrote:
You can quibble over the meaning of "genuine" randomness, and whether quantum processes qualify in a metaphysical sense. That's what folks like Ken Miller depend on to argue that radiogenic mutations might be guided by God, even if the guidance is now and forever scientifically undetectable.
However, as olegt points out, we can draw a non-metaphysical distinction between apparent randomness and the randomness of quantum phenomena: outcomes that are only apparently random can, in principle, be predicted; quantum outcomes cannot.
Thus when the authors of the paper wrote of environmental perturbations that were "essentially random (unpredictable)", they were not making a metaphysical point, as you seem to be claiming with this comment:
Comment by robin — June 21, 2008 @ 12:03 pm
June 21st, 2008 at 12:18 pm
robin:
Excuse me? I don't 'misunderstand' eliminativism, though I consider it a waste of time, voice and ink. But eliminativism is not the subject of this thread. You said to Todd:
Which is horsehockey, since eliminative materialists make all sorts of positive assertions and claims which, if believed, means eliminative materialism is false. Philosophy has its uses, but this is one of the least useful of its pastimes.
The primary assertion is that mental states do not exist. No free will, no qualia, no pain, no visual perceptions. No beliefs, no desires, no intent… no 'self' for one to be aware of, no empirical evidence to obtain, no science to apply it to. These folks can get downright solipsistic, which can be semi-humorous in some settings but is mostly a self-refuting bore. My opinion, having encountered the Churchlands and not being impressed.
robin to nullasalus:
??? What "quantum events" are you referring to? The type of random mutations that occur due to radiation exposure and ionization by free radicals aren't particularly pertinent to evolution, sans those accumulated in sperm during life. And those are, if not repaired, primarily neutral or harmful. Damage, not adaptation. Only those genetic alterations present in germline cells that go on to be born count. If you get skin cancer from a lifetime of sunbathing, evolution is not affected at all.
Now, certain genetic polymorphisms that have accumulated through history do prove adaptive much later on, as is the case with the many (42 as of 2000) versions of the human prion gene. These are survivors of poor dietary choices in isolated cultures historically, and will determine the incubation time for development of beta prion (and possibly other amyloid plaque) diseases. Any incubation time of 20 or more years is unlikely to affect reproductive success and will not be eliminated from the pool. An incubation time of months will wipe out otherwise 'fit' young people quickly in a most horrible way.
Incoming evidence is showing that gene mutations are not the engine of evolution as much as gene expression is. Gene expression is very complex, involving chromatin dynamics and histone 'coding' that does involve quantum forces. Because expression changes dynamically (and can be caused by nothing more material than thought), information processing is involved as well. Chromatin dynamics look to be best described biophysically. Both histone 'coding' (the function suite exposure triggers to chromatin state, and write capability) and information processing appear to involve state-switching proteins. State-switching is a result of quantum molecular forces (polarity and bonds). Or, in the case of prions (useful to memory imprinting and long-term storage), mere exposure to its evil twin.
Thus not necessarily random at all, nor largely unpredictable. Notice oleg's qualification…
"…quantum measurements will generate random outcomes under appropriate conditions."
Comment by Joy — June 21, 2008 @ 12:18 pm
June 21st, 2008 at 1:25 pm
To those of you making the tired argument that the evolutionary simulation described in the Science paper is invalid because design is 'sneaking' into the simulation via the design of the program itself, consider these points:
The simulation has no foresight. The mutations are generated (pseudo)randomly, with no correlation to their effect on fitness (they could just as easily be generated from a true random source with no effect on the end results of the simulation). And so a blind process, driven by random mutations and subject to selection, leads to virtual biochemical networks that are capable of representing and predicting the environment.
The networks that evolve are so complicated that when the researchers look at them, as the Science Daily article says, they can at first "make little sense out of them." As one of the experimenters notes, they bear no resemblance to the designs an engineer would produce to solve the same logic problems. To understand the networks, the experimenters have to pare them down to their essentials using virtual knockout experiments, and then figure out how the pared-down networks work by "careful inspection."
This boils down to studying the networks as if they had just been discovered in nature.
All of this makes it obvious that actual designs are not being introduced into the simulation in some subtle way.
A fan of EAM (endogenous adaptive mutagenesis), such as Joy, might argue that it is a mistake to model the mutations as random because an organism will actively generate specific mutations to enhance its fitness. Unfortunately for Joy, there is little or no evidence to suggest that this happens. Further, the simulations show that random mutations suffice. Guidance, whether endogenous or exogenous, is not needed. So not only is there no evidence for EAM in this study; it's also explanatorily superfluous.
If the mutations are essentially random and the actual designs haven't 'sneaked' into the simulation inadvertently, and if EAM isn't part of the picture, then the critic's remaining option is to criticize the "NS" part of the simulation — natural selection.
The critic must argue not only that the simulation mismodels selection — he must argue that it does so in a way that entirely accounts for the designs produced by the simulation. So far no one in this thread has pointed to a specific defect in the way the simulation models selection, much less shown how the defect gives rise to the designs.
Commenter johnnyb complained earlier that
His complaint ignores the fact that the researchers are not attempting to model the origin of life or the long-term evolution of bacteria. They are starting with a model organism and simulating how its responses evolve as it is subjected to patterned environmental perturbations. To say this is illegitimate is like claiming that we can't trust aerodynamic simulations because they don't explain the origin of the atmosphere.
The simulations show how the model organisms can evolve solutions to logic problems via random mutations and selection. This alone is a serious blow to IDers and creationists who have heretofore argued that new information cannot be created by RM + NS.
Some critics have argued that the problem lies not in the way the simulation models selection, but in something more fundamental. They claim that any simulation of RM + NS is doomed to fail, because computer code is designed and therefore cannot accurately model an undirected process such as RM + NS.
But if you make that argument about evolutionary simulations, then why not make the same argument about weather simulations? Would you claim that weather simulations are inherently unreliable because they model an undirected process using designed software?
Funny how we don't hear ID supporters making this complaint. You do hear some global warming skeptics complaining that existing climate models are unreliable, but you don't hear them claiming that accurate models are impossible in principle. Yet that is exactly what these critics of evolutionary simulations do.
My challenge to critics of the simulation described in the Science paper:
1. If you hold that evolutionary simulations are inaccurate in principle because they use designed code (and designed computers) to simulate a undirected process, then explain to us why this objection does not apply to simulations of other undirected processes, such as the weather.
2. If you hold that the problem is with the way that the researchers are modeling selection, then show us specifically where the error lies.
3. If you can't do either of these, then sit down, ponder, and come to grips with the fact that RM + NS can fashion solutions that solve logic problems, model the environment, make predictions, and allow organisms to change preemptively.
Comment by robin — June 21, 2008 @ 1:25 pm
June 21st, 2008 at 2:37 pm
I wrote:
Joy responded:
Joy,
That is a strawman caricature of eliminative materialism.
Take the example of pain. From the Stanford Encylopedia of Philosophy's article on eliminative materialism:
P.S. For the record, I am not an eliminative materialist (at least with regard to consciousness), but I find it unfair and counterproductive to misrepresent EM as Joy does.
Comment by robin — June 21, 2008 @ 2:37 pm