08-31-2006, 04:43 PM
|
#76
|
Guest
|
Quote:
Down21 wrote
Now you're really losing me. Do you really think that funding bodies are going to support research to find out what DIDNT happen in our evolutionary history?
|
Why should they fund something that won't tell is why it DID happen?
Quote:
Down21 wrote
You ask why a constructive scheme was not chosen? Back to neural development. A constructive scheme doesnt leave as much potential for correcting axon guidance errors. Neurogenesis occurs mainly embryonically but synapse pruning and fine-tuning continues in early postnatal life. A constructive mechanism would have to check itself after it wrecks itself and to build up neurons as they are required to make functional connections, meaning moving in and out of mitotic cycles continuosly rather than in one overproductive phase. An apoptotic scheme checks itself while it wrecks itself, making it more efficient. If natural selection has to "decide" there can only be one winner.
|
Now you're just making my arguments for me. This is an untestable claim, and is not science. It is merely a posteriori considerations. There are far more instances of constructive growth in the world that apoptotic. It takes an extraordinary explanation to make an extraordinary claim, such as "apoptotic development is superior to constructive development for neurological networks." In fact, this phenomenon isn't even necessary in simpler creatures.
Quote:
Down21 wrote
Aplysia is really well studied in memory research isnt it?
|
It's beginning to be. Since 2000, more work is being devoted, but it is still underfunded research.
Quote:
Down21 wrote
Physicists can study simple particles down to minute detail. Nobody can study the brain like that because it is not made of simple stuff. Physicists wont understand the brain any more than neurobiologists. We can only use the tools (including the brains designed for naked hunting) that are available. It would be great to study consciousness down to that level and unify the field....but isnt it too subjective an experience to be open to scientific investigation? Maybe , maybe not....never say never!
|
It may seem that we are studying an object incomparable to particle physics. But this, again makes my argument for me! Why don't we start as simple as possible? It is lunacy to do otherwise.
I re-iterate before taking a long weekend to think about things: we are simply asking the wrong questions. Please, take your time. And really show me that I'm wrong.
|
|
|
09-01-2006, 03:09 AM
|
#77
|
Guest
|
No mike, you are now making my argument for me. Of course what I said is speculation becuase it is impossible to test what DIDNT happen during evolutionary time. One can only speculate in light of what we know happens. We can compare the mechanism of what did happen ie neurotrophic theory to other potential mechanisms but since these mechanisms did not construct our brains it is obviously speculative. That was my point since we cannot test what "didnt , but couldve potentially happened" nobody can design experiments let alone get funding to test a speculative claim of that nature.
We are studying at the molecular level what happens in single cells. I dont see how much simpler we can get than that with the tools at our disposal.
By the way , just out of interest could you describe how the non-apoptotic mechanism of development in simpler creatures works to build a functional nervous system? Are you saying there is no overproduction of neurons at all. I have not heard of such a mechanism and would be interested in it.
|
|
|
09-01-2006, 03:49 AM
|
#78
|
Obsessed Member
Join Date: Oct 2005
Location: Oxford, UK.
Posts: 2,330
|
I have to argue with you about the legitimacy of evolutionary explanations. I was interested to see the idea that you'd like an explanation for why NS favoured a particular setup.
The thing is, any explanation of why something evolved that way will, by its very nature, be post hoc. Although it would be interesting, any explanation of any evolutionary pathway will have to take contingency into account. It may be that this mechanism has developed because the genetic mechanism was available, because it was useful for some ancestor in an environment that no longer exists. NS doesn't start from scratch but works with the mechanisms and tools that an organism already has. So I don't see how this is going to give you the kind of predictions you want, although there will be testable predictions generated along the way, so science's error checking mechanism is still operating.
Now it may be that you could get some kind of energy budget style explanation, but for it to be "true science" as you put it you would have to prove it superior to every other possible mechanism. Since we haven't thought of them all yet, I don't see how this is possible - you want to make a truly untestable hypothesis! What you're asking for is equivalent to searching for a law of history - it's the same kind of reasoning the IDiots use when they try to portray evolution as untestable. While NS does make predictions, they generally aren't about specific evolutionary pathways, because those pathways are subject to contingent circumstances that are unlikely to be releated.
If this smacks of the Gould approach to NS, I'm with Gould on this issue (though not on many others). It's also one of the few areas where Gould agrees with the neodarwinian consensus. I'm not sure the kind of explanation/theory you want is even appropriate to observational sciences.
What I don't understand is why you think this makes those sciences somehow less worthwhile.
"You care for nothing but shooting, dogs and rat-catching, and will be a disgrace to yourself and all your family"
|
|
|
09-05-2006, 07:51 AM
|
#79
|
Guest
|
When people engage in a debate, and neither side is receiving any impetus to move, it usually means that the debate is happening in two different arenas.
Thinking over the weekend, it occured to me that I was arguing about the goals of neuroscience, and you are arguing the methodology. Once I understood this, it became a lot easier to see how two presumably bright individuals could continue to disagree for so long.
Fundamentally, you are arguing that the goal of neuroscience, as the name explicitly implies, is simply "to understand the neuron". I, on the other hand, was arguing that the goal was "to understand neural systems" (leaving much of the details to molecular cell biology). Now, clearly, my goals for the science are just a subset of your goals, which explains why neither of us could understand why the other kept arguing, and why you had no problem asking the questions you wanted, and I thought they were too broad.
Also, after much thinking, I decided to concede that neuroscience really should be the more global study of everything neural, not just neural systems. There is not yet a name for the subfield that I really want to see developed -- cognitive neuroscience is pretty far off base, and what we usually think of as systems neuroscience is not quite what I'm looking for either. Maybe it will never come to fruition, but I hope it does.
So, with a very different concept of what the goals of neuroscience should be, I would concede this argument. (I am not so hard headed that I can't admit when I've made a mistake.)
Finally, to answer your questions from the last couple posts:
Quote:
Down21 wrote
By the way , just out of interest could you describe how the non-apoptotic mechanism of development in simpler creatures works to build a functional nervous system? Are you saying there is no overproduction of neurons at all. I have not heard of such a mechanism and would be interested in it.
|
You misunderstood. I was arguing that there was no a priori reason to assume that any nervous system should use apoptotic mechanisms for development, when most systems, like bone construction, muscle construction, exoskeleton construction, epithelial construction etc etc are all performed constructively. I would still hold this to be a valid question, and unresolved. (As for my simpler creatures comment - I am unaware of any apoptotic development mechanisms in any colonial or simple multicellular creatures. If you have counter examples, do share. Further, for creatures with ~20 neurons, apoptotic development is a ridiculous concept. Hence why I wondered why you felt it was an inevitable evolutionary path for more complex systems.)
Quote:
a different tim wrote
What I don't understand is why you think this makes those sciences somehow less worthwhile.
|
In light of my comments above, about what I felt the goals of neuroscience should be, it should be obvious that I thought that people were conducting experiments that were well outside the purview of the field. An organic chemist does not, typically, worry about the species they got their ADT from, by analogy. However, after updating my idea of the goals of neuroscience, these types of experiments seem permissible, and even acceptable.
Quote:
a different tim wrote
The thing is, any explanation of why something evolved that way will, by its very nature, be post hoc. Although it would be interesting, any explanation of any evolutionary pathway will have to take contingency into account. It may be that this mechanism has developed because the genetic mechanism was available, because it was useful for some ancestor in an environment that no longer exists. NS doesn't start from scratch but works with the mechanisms and tools that an organism already has. So I don't see how this is going to give you the kind of predictions you want, although there will be testable predictions generated along the way, so science's error checking mechanism is still operating.
|
I disagree. If you can hypothesize the selective pressures that caused evolution to proceed in a certain direction, then find a similar scenario where either the same or the opposite pressures occured (say, two related species which live in different salinities), you can then make predictions that are verifiable and/or falsifiable. I'm not sure, though, if this is strictly speaking the domain of neuroscience, however.
Hopefully I will have cleared up any confusions. If elaboration is required, however, I won't be shy.
|
|
|
09-05-2006, 07:53 AM
|
#80
|
Guest
|
Quote:
a different tim wrote
I have to argue with you about the legitimacy of evolutionary explanations.
|
Oh, one more thing.
I still think evolutaionary explanations are totally invalid. They are nice and comforting, but unless you can back them up, they're worthless (scientifically speaking).
There, that should fuel another fire or two :)
|
|
|
09-05-2006, 08:26 AM
|
#81
|
Obsessed Member
Join Date: Oct 2005
Location: Oxford, UK.
Posts: 2,330
|
Quote:
Mike the Kitty wrote
Quote:
a different tim wrote
I have to argue with you about the legitimacy of evolutionary explanations.
|
Oh, one more thing.
I still think evolutaionary explanations are totally invalid. They are nice and comforting, but unless you can back them up, they're worthless (scientifically speaking).
There, that should fuel another fire or two :)
|
Oh yeah. But if you want to fuel the fire answer the whole post, not one line. I made an argument to back up my point there, you know. ;)
However, I sort of know what you mean, evolutionary explanations can fall into the category of just so stories very easily. The trouble is, so can the kind of thing you're talking about here:
Quote:
If you can hypothesize the selective pressures that caused evolution to proceed in a certain direction, then find a similar scenario where either the same or the opposite pressures occured (say, two related species which live in different salinities), you can then make predictions that are verifiable and/or falsifiable.
|
Now, if you isolate factors in the way that you suggest, this is something that clearly only applies in a lab situation, unless you have some hitherto unknown method for isolating single factors in a fossil environment. So it may or may not bear the slightest resemblance to the way a given organism actually evolved in the wild. The kind of experiment you suggest is fine for testing the notion of evolution in broad, and is also OK if you want to look at engineering organisms to, say, produce insulin. What it won't tell you is the reasons a specific organism took a specific evolutionary pathway. So your experiment wouldn't tell us much about the natural evolution of actual organisms, which is why I felt that you were looking for a kind of solution that is inappropriate to this kind of science.
You argue that neuroscience should be looking more at neural systems, but what if the understanding of that system depends on an understanding of neurons and their connections? Neurons, it turns out from what I understand of Scatty's posts on the subject, are far from simple on/off switches or even the simulated neurons in most digital models - they vary from one second to the next in their signal processing depending on their chemical environment, which they in turn affect, they change their connections to other neurons on an ongoing basis, synapses are capable of sending multiple types of chemical signal rather than 1/0 etc. Now it's true that you could model, even digitally simulate, all this, but unless you have a pretty good idea of how neurons behave your model will be wrong, and probably in a non-useful way.
I think neuroscience is edging towards systems models (and simple models have been built by roboticists to simulate e.g. insect locomotion) but most neuroscientists admit that they don't know enough yet to build anything useful.
[edited to cut out a bunch of stuff that we've gone over before]
"You care for nothing but shooting, dogs and rat-catching, and will be a disgrace to yourself and all your family"
|
|
|
09-05-2006, 09:08 AM
|
#82
|
Guest
|
Quote:
a different tim wrote
If you isolate factors in the way that you suggest, this is something that clearly only applies in a lab situation, unless you have some hitherto unknown method for isolating single factors in a fossil environment. So it may or may not bear the slightest resemblance to the way a given organism actually evolved in the wild. The kind of experiment you suggest is fine for testing the notion of evolution in broad, and is also OK if you want to look at engineering organisms to, say, produce insulin. What it won't tell you is the reasons a specific organism took a specific evolutionary pathway. So your experiment wouldn't tell us much about the natural evolution of actual organisms, which is why I felt that you were looking for a kind of solution that is inappropriate to this kind of science.
|
I agree to some extent. But at least the lab scenario is telling with scientific evidence that we have one possible mechanism for that change! This is much more than simply falling back on natural selection.
Even more important would be a negative result -- you thereby eliminate one pathway entirely, saying expressly what did NOT happen in the genetic past of the object in question. If a given receptor binds acetylcholine, and you prove that it is NOT because of a change in salinity under certain other assumptions and conditions, that is very useful information.
Stopping your argument by relying on natural selection is, of course, always correct, but not always informative. That's my only point. I agree that the converse isn't always particularly useful, but at least you're asking the right questions and continuing the line of questioning.
Quote:
I made an argument to back up my point there, you know. ;)
|
There. I've done the same :)
|
|
|
09-05-2006, 11:03 AM
|
#83
|
Obsessed Member
Join Date: Oct 2005
Location: Oxford, UK.
Posts: 2,330
|
Quote:
Mike the Kitty wrote
Even more important would be a negative result -- you thereby eliminate one pathway entirely, saying expressly what did NOT happen in the genetic past of the object in question. If a given receptor binds acetylcholine, and you prove that it is NOT because of a change in salinity under certain other assumptions and conditions, that is very useful information.
|
But I'm not sure you could do this. Genes express depending on environment, especially in development (which, going back to neuroscience, is a pretty big part of what you're interested in). So you prove a gene does not change in expression depending on salinity, in a present day lab experiment, but with an ancestral aninal you're dealing with a different environment and most certainly a different suite of other genes that are influncing the developmental expression of that one gene. So you haven't closed the pathway.
You can show what happens in present day animals, though, which is useful in some cases but not when answering "why" questions. (I think in many cases "why" questions are misconcieved but that's a philosophy of science argument for another time, perhaps).
OK, perhaps it's not as hopeless as that, because some genes, we can show have similar effects across a range of organisms (hox genes and other regulatory genes come to mind) but generally speaking you aren't necessarily proving what you think you are with your kind of experiment. Which is why evo/devo people, and neurologists, are wary of the kind of conclusions you want to draw and get a bit hacked off when you describe this proper caution as stamp collecting.
(yeah, I know the Rutherford quote, but I think he was wrong).
"You care for nothing but shooting, dogs and rat-catching, and will be a disgrace to yourself and all your family"
|
|
|
09-05-2006, 11:16 AM
|
#84
|
Guest
|
Quote:
a different tim wrote
But I'm not sure you could do this. Genes express depending on environment, especially in development (which, going back to neuroscience, is a pretty big part of what you're interested in). So you prove a gene does not change in expression depending on salinity, in a present day lab experiment, but with an ancestral aninal you're dealing with a different environment and most certainly a different suite of other genes that are influncing the developmental expression of that one gene. So you haven't closed the pathway.
|
Yep. But as I mentioned, you've figured something out given certain assumptions and conditions. At least you're asking the right questions, which is probably over-abundantly-clearly my main thrust.
Quote:
a different tim wrote
You can show what happens in present day animals, though, which is useful in some cases but not when answering "why" questions. (I think in many cases "why" questions are misconcieved but that's a philosophy of science argument for another time, perhaps).
|
Hell no! This is one of the principle wastes of my mind's time! I am constantly wondering when it is or isn't appropriate to ask "why". I would love to hear some opinions. My mind isn't even close to made up yet, and the more ideas the merrier.
Quote:
a different tim wrote
OK, perhaps it's not as hopeless as that, because some genes, we can show have similar effects across a range of organisms (hox genes and other regulatory genes come to mind) but generally speaking you aren't necessarily proving what you think you are with your kind of experiment. Which is why evo/devo people, and neurologists, are wary of the kind of conclusions you want to draw and get a bit hacked off when you describe this proper caution as stamp collecting.
(yeah, I know the Rutherford quote, but I think he was wrong).
|
No, I was being very careful about saying what I was proving. It almost never transfers to the original conditions, which are usually an unknown, but it you have a question you want answered, you start by tweaking one variable at a time, until you figure out how the linearities in the system work, then you progress to the complexities. You are only proving what happens when you adjust one parameter.
And incidentally, I think almost all science is stamp collecting, including the science I personally engage in on a daily basis. I'm not saying it never has pragmatic value, I'm just saying that I wish more people would do other things. I don't personally practice what I preach. This is just a beautiful dream. (I also don't try to convert the religious -- I should probably start doing that, too.)
|
|
|
Thread Tools |
|
Display Modes |
Linear Mode
|
Posting Rules
|
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
HTML code is Off
|
|
|
All times are GMT -7. The time now is 12:54 PM.
|