Catch up on stories from the past week (and beyond) at the Slashdot story archive

 



Forgot your password?
typodupeerror

Comment Re:Hard problem of consciousness (Score 1) 351

Nothing in natural science can predict that such and such configuration of atoms and fields has a consciousness attribute.

That's only because you are not defining this "consciousness attribute." You complain of handwaving, yet engage in this most egregious form. Define it as "the subjective experience of a functioning neural network" and now you can test it(remove or disrupt the neural network and examine the effects). Now you can predict from it(Prediction: neural networks with varying complexity will still exhibit common behaviors to similar stimuli). Now, it has moved from speculation to a hypothesis.

This is the difference between science and groundless speculation. Speculate all you want, but if you can't define your speculation well enough to be testable, don't pretend that it's rising to the point of science. We know that neural networks function. We grow and train them in the lab all the time. We know that disrupting these networks has objective and subjective effects. We can examine the stratification and separation of function in our own NNs. We can often predict the type of damage that we will see when examining behavior. We can see that brains of similar configurations to ours(mammals, for instance) react in similar ways to us, and brains of highly similar configuration(primates, for instance) have VERY similar(in many cases identical) reactions. And so on. If you think that there is something special about consciousness which is unexplained by this theory, by all means, define it, test it and publish your findings. Don't hide behind handwaving and philosophical wankery and expect to garner respect for your ideas, though.

Comment Re:Hard problem of consciousness (Score 1) 351

You are simply offering one among conceivable conjectures consistent with observations, the matter-mind stuff identity hypothesis.

Yes. This is called "science." Invisible magic unicorns poking neurons with sub-atomic horns is also consistent with observations. There is a reason why no one would take it seriously, though, and you even brought it up in the next line...

The point of the "hard problem of consciousness" (which, from your "summary", you seem to thoroughly misunderstand) is that present natural science lacks any way to empirically distinguish between them i.e. currently they are merely unfalsifiable speculations.

No, I understand it just fine. The difference between those and the scientific or naturalistic explanation is that "unfalsifiable" bit. The naturalistic explanation is falsifiable, and has never been proven false. Everything else, on the other hand, has either been proven false or is mere unfalsifiable speculation.

Comment Re:Hmmm .... (Score 1) 351

Why does it have to be one or the other? Do you think that if God talked to you, there would be no evidence of it to be found in your brain?

Define the term "god," define the evidence you expect to find and why, and come up with a plausible method of measuring that evidence and proving that it is indicative of your "god" premise, and you might be on to something. Good luck with that.

Comment Re:Hard problem of consciousness (Score 1) 351

The natural science lacks explanation for consciousness

You are mistaken. The explanation is rather simple: your nervous sysem gives rise to consciousness. Your link simply asserts that physical explanations of consciousness are invalid if they don't include an element of teleology, without proving that this is necessary, of course.
Your post indicates that you are hyperskeptical when it comes to consciousness. Sure, perhaps physical states merely coincide with mental states. And perhaps the computer you are using doesn't actually use electricity to perform work. The flow of electricity through the system simply coincides with the work. If you can't explain WHY an electron behaves the way it does, then we lack any explanation for the function of digital computers.Right?

Comment Re:Hmmm .... (Score 1) 351

It's interesting that you're willing to draw a significantly stronger conclusion that the authors of the study.

Statement from poster(emphasis mine):

So at a time when you're not conscious, and random activity is spiking in your brain, you might experience something as the various bits turn on.

Statement from article(emphasis mine):

He does think that this surge in activity recorded in rat brains would probably be similar in humans.

Not a significant difference in conclusions.
The bit about supernatural experiences is a truism: some people interpret experiences religiously.

Comment Re:not evolution (Score 1) 387

So you can bring those to bear on the subject of finches and cliff swallows? Please, do cite away.

No, I bring those to bear on the larger issue of evolution.

So if a population oscillates between two dominant phenotypes (small beak/big beak; short wings/long wings), you are satisfied that this is an example of evolution even if there are no net changes to the organism's genome?

If that trait is being passed on to offspring, then yes, it is. In the case of Darwin's finches, you have distinct species(little to no interbreeding) which inhabit different niches and which display morphological differences which are readily apparent. You can find similar examples in ring species. The cliff swallows are indeed evolution(provided that the traits are indeed being passed on to offspring), but this is not a speciation event(at least, not yet).

And you find that this is strong evidence that over a time period of millions/billions of years, that this oscillation will result in an entirely different animal?

Since this is couched within the larger body of evidence regarding evolution, certainly. As for what type of animal the decendents of those birds will be, it all depends on the variety of environmental pressures which are exerted upon the species. In a few million years, those swallows may well give rise to flightless sea birds. In a few million years, invasive species in the Galapagos Islands may lead to the evolution of larger meat eating finches. Genetic drift may lead to something entirely unexpected. And they may all just go extinct, leaving some other species to adapt and fill their respective niches. It would take the ability to see the future to know how environmental pressures are going to affect their genomes in next few million years.

If there is no net change over hundreds of years, how do you that into non-zero net change over millions of years?

I said drastic evolutionary change. Those species are indeed ungoing evolution right now. Every species is. The process runs on geological time. We can chart genetic change, but you have already made it clear that you hold to the creationist idea of there being different "types" of evolution. You won't be satisfied by anything short of a major morphological change happening in your lifetime. Do you also dismiss plate tectonics because mountains don't spring up over the course of a human lifetime?

We can both accuse each other of obfuscation, but I haven't relied on accusations of dishonesty and ignorance.

Good for you. Having cornered many creationists into finally admitting their a priori assumptions, I no longer have the patience to assume that someone who is presenting the same old arguments is doing so in good faith.

I don't find it useful to slap "evolution" on any type of change, when the character of the change is entirely different.

The character is not different. That's the point.

Does a child "evolve" into an adult?

Is a child a population?

Comment Re:not evolution (Score 1) 387

When was that observed? Oh, it wasn't? You're extrapolating based on fossil evidence? That's nice.

Fossil evidence, DNA evidence, geological evidence. You know, converging lines of evidence.

You still aren't talking about the finches. Did you concede that as a point?

Nope. Why would I? You have simply rehashed an old creationist canard which is usually presented to assert that microeveolution occurs while macroevolution does not. It's also usually presented as an example of Biblical "kinds." It is certainly an example of evolution. Whether it is an example of complete speciation is another question, but it has no bearing on the larger issue.

If you're going to concede the finches, which of those other "numerous" examples are based on observation of live specimens, and which are extrapolations from fossil records?

We have exactly two categories of specimens to examine: live and dead. The expectation that we are going to see vast evolutionary leaps in real time is part of this distorted view of evolution that you are pushing. It doesn't correspond to reality or to the theory of evolution. You can choose to ignore evidence from fossils and DNA comparisons all you like. It's just dishonest.

Do you think it is useful for the sake of scientific observation to distinguish between "evolution" that results in no change, as opposed to "evolution" that results in brand new organisms?

All evolution, by definition, results in change. So, no, it's not a useful distinction. It's just an attempt to evade the obvious conclusions of evolutionary theory.

s it impossible to distinguish between the two, even though Darwin's finches are still finches after hundreds of years of "evolving"?

I'm tempted to rescind my prior apology. Are you really so obtuse that you think that we should expect some drastic evolutionary change in those finches in hundreds of years?

Do you prefer a scientific lexicon that deliberately obfuscates different activity by using the same word in different senses?

No, I prefer honesty in approaching science and evidence, not deliberate obfuscation to support creationism.

Comment Re:not evolution (Score 1) 387

"All the time" ... Observed how?

Examples include trisomy, insertions and gene duplication.

In short, no, human mutation rate was not measured. They compared human and chimp genomes, assumed the difference is due to evolution and then treated that difference as the human mutation rate using dates derived from evolutionary assumptions. Exactly what you just said before in summary, which I challenged for being circular when used as a rebuttal to my challenge.

Point taken. So, start here, then(a good overview of precisely the mistake I made). The lowest estimated mutation rate based simply upon human genomes is 1.0 x 10-8 per site per generation. Still the same order of magnitude, so it won't have a substantial effect on my point.

Read this and question for a moment the fallibility of human imagination.

And I suggest you read some of the comments by actual paleotologists on that page.

Then look at these two pictures and tell me why the concept art is "scientific" as opposed to "fantasy"

Because we also have fossils of the related species as well, which gives us a good idea of how the intermediate species will look. But, then, judging by your later comments, you've already decided that physiology can't be derived from fossil records, so I wouldn't expect it to matter.

See, if evolution is true, then there's no reason why we can't one day reverse engineer DNA completely (hey, it's random and we're semi-intelligent). At that point in time, we can create any arbitrary DNA sequence, and should be able to reconstruct the intermediate life forms from the DNA sequences. If it happened once by lucky circumstances, we can do it again, intentionally.

And why would you assume that? We could create a lookalike, but we will never know if we got all of the genes correct.

In short, there's nothing in natural selection that can select against specific base pair mutations. Natural selection can't select for the future, it only compares against now.

Of course.

Not by natural selection based on overall fitness, but by highly specific genetic "error-check" systems (against what reference?).

What are you talking about? I am talking about death. An organism with a broken metabolism won't survive past a single cell stage. An organism lacking cellular adhesion would not survive past that stage. No "reference" involved at all. Things that are broken just die.

Now how did natural selection work *before* that system evolved into existence, and where is your evidence that life works without it? (Even "basic" bacteria have this functionality)

Early life would have had more errors in transcription, which is exactly what we would expect. Later, as gene expression became more robust and complicated, selection pressure would increase for less error prone mechanisms.

It was a starting point to illustrate the enormity of the problem

And, as I pointed out, you ignored the reality of the theory in order to artificially inflate the problem.

This is part of why I don't consider evolutionary theory to be "solid scientific fact" as you do - it fails to do rigorous mathematical modeling. If it is so plausible, there should be math that puts my rough model to shame; and yet I haven't seen anything that attempts to capture the probabilities involved or how evolution comfortably meets the challenge.

That is because you are starting off from a bad theoretical foundation and expecting the theory to match up to that.

Take away the "evolution by natural selection" examples, and all you have left is a bunch of pseudo-history extrapolated from tea-leaves-esque fossil readings.

"Take away some of the proof, and all you have left is evidence that I personally disagree with."

Comment Re:not evolution (Score 1) 387

You ignored the grandparent's discussion of Australopithecus in favor of focusing on finches. Let's look at the parts of their questions that you cut out:

Right. What is the alternative, add up a lot of small changes and then not see any change? Australopithecus came down from the trees, got longer legs, shorter arms, better vision, upright posture, wider pelvis, smaller teeth and a larger brain. Is it still an Australopithecus, or is it now a Homo Habilis? I fail to see any logic in your statement, these small changes have added up to a new animal.

Why did you ignore this? Is it because it doesn't fit your "all changes are minimal," sine wave theory? Australopithecus is not just a variation of humans. We don't vacillate between those extremes. Why are you ignoring the thousands of examples out there of biological change over time in favor of your static vision of species?

This is why I bandied the term "ignorant" around. Either you are not aware of the numerous examples of cumulative changes resulting in new species, or you are. One is a form of ignorance, which is blameless and can be easily remedied. The other is a rejection of the evidence, and, given the fact that you are a literate human with access to the Internet for research, is far more problematic. I apologize for being dismissive, but I am really trying to get you to justify your position that changes over time cannot add up to speciation events.

Comment Re:not evolution (Score 1) 387

Is natural selection an algorithm that is "looking for" humans?

Nope. We are just a happy accident in the history of life.

That inheritance improves the odds slightly was not in question. The point was that there are other unaccounted forces (survival, environmental changes, luck) working against the unaccounted filtering effect, such that you cannot assume progress just because a filtering effect exists.

Given that the evidence supports the fact that this is in fact what happened, that assumption is more than justified.

Do you believe that 3.1 Gbs of human genome is filler? If true, you should be able to irradiate and randomly mutate somewhere around 99% of a human's DNA sequence with no ill effect.

See, this is where your simplified version gets you into trouble. Genes are added to genomes all the time, from gene copying mistakes to interspecies breeding to bacterial gene interchange to endogenous retroviruses. Entire chromosomes are sometimes duplicated. Random events are adding genetic material all the time, and natural selection is winnowing out the non-working combinations.

You assumed that humans evolved from chimpanzees, extrapolated that rate to the past, and found it reasonable that humans evolved from chimpanzees and fish. That's circular reasoning.

Nope. I demonstrated that, if evolutionary theory is true, the rate of change as demonstrated by our closest living relatives is consistent with the amount of time that change had to have happened, which is precisely what you are denying.

Can you support the assumed human mutation rate with a measured human mutation rate?

Sure. Start here:

The average mutation rate was estimated to be ~2.5 × 108 mutations per nucleotide site or 175 mutations per diploid genome per generation

6 million years divided by an average generation time(20 years for humans, 15 for chimpanzees, so 17.5 average for both species) gives us 342,857 generations. This yields 59,999,975 mutations.

Can we reverse engineer all those intermediate organisms? (Can't do that now, but should be possible if evolution is true!)

Can we extrapolate the likely form and location of the intermediate organisms? Sure. That's how we discovered Tiktaalik, for example. The researchers worked from an understanding of the development of the tetrapods and the population distribution in the fossil records and predicted that they would find a proto-tetrapod in Devonian strata in Canada. And they did. As for whether they can be recreated exactly, then no, and there is no reason to expect it to be possible at all. We can compare related organisms to determine which genes were likely conserved or developed independently. That can get us in the ballpark(which is why we can infer a lot about intermediate species), but recreating an entire extinct genome is likely never going to be possible.

Filters always remove information. At best, with an analog signal carrying digital encoded information, you can remove the noise without harming the information.

Hey, you are almost there! Congrats!

Filters can be a part of an overall creative process

Bingo! That's what I was saying!

A random string generator has the potential to "write" a novel.

Indeed. And I have generated the first paragraph of "Romeo & Julliette" using one that starts with a small string of random gibberish and uses random mutations and selection to generate it. Of course, the couplet is just a model used to represent a theoretical genome which is fit to survive in a particular environment.

Having solved that simpler problem, does natural selection really behave like the filter you built?

Natural selection works on a higher level, so to speak. Genes live or die in the organisms which they build. So, while the filters represent a high level view of how natural selection works, it would take a few more steps to make it more realistic: interpreting the "genome," building an "organism" from that genome and allowing it to compete in a simulated environment. This is, coincidentally, one of my current personal projects.

Every organism has a sequence: be born, mature, survive, reproduce. That is a meaning that natural selection has to find, even though it doesn't know how to look for it.

OK, let's go with that, noting that we are ignoring a large part of the organisms on earth. Most catastrophic genetic combinations are selecting against before birth. If an organism has a problem with low level protein synthesis, it probably won't make it through gestation. This process continues throughout the organism's lifetime. If they can't compete for resources, they are less likely to live long enough to pass on their genes. If they can't escape predation, they are less likely to live long enough to pass on their genes. If they can't compete in mating, they are less likely to live long enough to pass on their genes. And so on. Natural selection isn't "looking" for something. There is no "correct" genome which is the goal. There is no goal. There is no intelligence. There is just life, death and adaptation.

Yes, I get that you're saying it. The math you have offered is insufficient to support it though. Using simplified models that assume things in your favor is not how you "proof of concept" the numbers.

Funny, coming from someone who started with a ludicrous "tornado in a junkyard" style argument to pick apart a straw man representation of evolutionary theory.

"It works in theory" does not mean "it works in practice".

You have it backwards. We saw that it worked in practice, and constructed a theory to fit the evidence.

On the flip side, I can "disproof of concept" with a model that assumes in your favor; if you can't do it in favorable conditions, you obviously can't do it in disfavorable conditions.

No, your model disproves a model of evolution which exists only in your head.

All barbs aside, I would say that you should seriously consider taking some courses in population biology and evolutionary biology.

Comment Re:not evolution (Score 1) 387

Faced with an inability to defend your assertion, you have switched topics. Keep digging.

You are the one switching topics. The fossil record clearly demonstrates the changes in lifeforms over billions of years. You refute that by appealing to a sine wave, with no justification for ignoring the evidence. Saying that you are ignorant is the kindest assumption.

Comment Re:not evolution (Score 1) 387

Your model is incapable of failure (extinction), which makes it rigged.

And your model ignores everything we know about the development of life, which makes it at best an example of a complete misunderstanding of evolution.

It assumes that every single "correct" base pair you hit is "hard-saved" from ever being lost. Your model only takes 25-30 independent rolls to "hit" the target, because every dice is rolled in parallel and is utterly independent of the dice around it.

And you completely miss the point. My example is a very crude example of how filters turn random input into non-random output. Of course it doesn't take into account mutations, extinctions, etc. It's a model which can be expressed in a short Perl script, after all. But it's still closer to what is actually happening than the mess you posted.

Certainly there's no real world guarantee that every "correct" mutation to the genome gets saved and never gets corrupted.

Of course not. That's a legacy of my model. And I've made more complex models which do not guaranteee survival of "correct" genes and model sexual reproduction, random deletions, random additions and various mutations. The same filtering process is apparent there, too.

(Natural selection is going to determine survival or death with 100% certainty based on a single base pair mutation in a 10 million/1 billion base pair sequence?)

If the mutation is sufficiently detrimental, certainly. Most of the time, it is not. You and I both have novel mutations in our genes which are not present in our parents. Yet, we've survived long enough to type out these messages, and I've survived long enough to reproduce, so my novel combinations have a chance to perpetuate.

The models we've used also ignore that code doesn't work incrementally. Half of a protein is more likely going to be a non-effective protein rather than a half-effective protein. The dice are interdependent. "THE" is a word and has meaning. "TNE" is not.

Proteins do not have to be perfectly effective to work. And non-functional or semi-functional proteins are not necessarily detrimental to survival, thus having no effect on survivability. For example, suppose a primitive proto-bacteria had a duplication event and now two proto-genes are generating the same protein. This very well may have no effect on its ability to reproduce(and it very well could, too; there are no guarantees). However, now one of those copies can be subject to mutation without affecting the original protein. It may end up non-functional(so called "junk DNA"), it may end mutating and functioning slightly different from the original gene(like the many related proteins responsible for blood clotting) or it may end up deleted at some point. In fact, if you want a good example of how proteins can evolve, look up the evolutionary history of blood clotting mechanisms.

So let's try a different analysis. How many generations do you need to get from "nothing" to human? If we assume each generation can add one "correct" base pair, that takes 3 billion generations.

Given that we share a large part of our genes with bacteria, roughly along the lines of 40%, then a lot of that took place rather rapidly. At a generation rate of 20 minutes and shooting for 40%(1.2 billion generations), that takes 937,000 years, roughly. Subtract that from 3.6 billion years, and you have a 40% human genome with 3.599 billion years left to generate the remaining 60%. The generations do get longer for multicellular organisms, but that is counter balanced by the development of sexual reproduction which speeds up the exchange of novel and complimentary genetic changes.

Let's look at it from the opposite side. We differ from chimpanzees by roughly 2% of our DNA, or 60 million base pairs. We diverged from their line about 6 million years ago. That's a rate of 10 base pair changes per year. Using that rate, we have a possible 5 billion base changes in the past 500 million years, which is roughly the age of fish. We only differ from fish by about 25% of our DNA, or 750 million base pairs. At that rate, there was enough time to evolve a human from a fish 6.6 times.

Of course, in reality, generation times and mutation rates vary, as does the environment. The point here is that your numbers don't reflect reality. You apply math to a straw man version of evolution.

If natural selection only removes individuals from the population, it cannot increase genetic diversity, which means it cannot add the information needed to get from life Zero to humanity.

I fail to see what you are missing here. Natural selection doesn't generate new genetic material. Sure. It acts as a filter on genetic material, new or otherwise, and that's what turns the random introduction of genetical material and reconfigurations into non-random genetic diversity. Call it additive, call it creative, call it destructive, whatever. The semantics don't change the function.

Perfect natural selection does improve your chances if it only lets the "most correct" individuals reproduce, but that assumes a winner take all system where every tiny reproductive advantage is propagated across the entire population in a short period of time. That's not how large populations work.

"Perfect natural selection" is not required. Nor is it a part of actual evolutionary biology.

I don't believe in guided evolution. I'm saying you need to believe it if you want the numbers to work.

And I'm saying you are dead wrong.

Slashdot Top Deals

1: No code table for op: ++post

Working...