Sensitivity to initial conditions is one feature of a chaotic system, but that alone does not make a system chaotic. Chaotic systems are ones that are capable of generating behavior that looks random, even though the system itself is not. Even from what you're describing, it sounds like cancer cells generate random, unpredictable behavior because the patterns of mutation and aneuploidy ARE random and unpredictable, not because there are chaotic dynamics at work. A more precise definition of chaos is here.
I think you're right that there are anti-science elements on both the left and the right. You say that Christians are targeted more than radical environmentalists, and I agree. However, you seem to be arguing that this is because scientists are all pinkos, and the attacks are politically motivated. In truth, it's because anti-science Christianity is a much larger threat to science in the US right now. If anyone actually listened to radical environmentalists, I'd be out in the street picketing them too.
A second point is that many of the "ethical" objections you're talking about are actually based on misunderstanding, or occasionally deliberate misrepresentation, of the facts. For example, somatic cell nuclear transfer is not the same thing as cloning. Yet some political operative managed to get it branded with the term "therapeutic cloning" or "research cloning" or whatever, which is simply incorrect, and by the time the scientific establishment realized what was going it was too late. How many fewer people would be ethically opposed to SCNT if they ACCURATELY understood what it was? It's the same thing with the "life begins at conception" meme that caught on years ago, in spite of being demonstrably false. The point is, it's not just ethics.
For those of us who work with femtosecond pulse lasers, the cliche is that the ratio of one femtosecond to a second is smaller than the ratio of a second to the age of the universe.
They might actually have some advantage when it comes to early research with using GAs to reproduce NNs. Is the Drosophila nervous system consistent, or is there a certain amount of randomness to it?
The Drosophila CNS is not like C. elegans, where every single neuron is specified, but in some brain regions it's close to that. Also, there are not very many neurons--I think something like 250,000 in the entire brain, and often as few as tens to hundreds of a certain class of neurons in a particular structure, such as the antennal lobe. The other advantages include the huge array of genetic and anatomical tools that can be readily borrowed from the fruit fly community, short generation times, and small brains that facilitate imaging (which is what I do). The one big drawback is the difficulty of doing traditional neurophysiology in the fly. Extracellular recordings are apparently very difficult, if not impossible, by traditional methods, since the neurons are so small. Doing intracellular recordings is also very difficult largely because the organism is so small--it takes skill even to open up the head and take the brain out. Behavioral assays are also a nightmare. So it depends what kind of work one is trying to do.
Right now, the Drosophila antennal lobe, which is their version of the olfactory bulb. To be honest, most of what I've done for the past few years is more like biophysics than neuroscience (the perils of technology development), but I'll be moving back towards actual neurophysiology after I defend my thesis next week.
So, the Izhikevich-type neurons...I actually used those once, a few years ago, when I was attempting to make a model of oscillations in the locust antennal lobe. I actually started with a Wilson-Cowan-type model, then moved on to the Izhikevich model when I couldn't get it to work. It was just a rotation project, so I never had time to make it pan out, but still, interesting stuff.
That sounds like a smart strategy. As a (theoretically-inclined) experimentalist, I'm working on developing new techniques for fast membrane potential measurement using light. Sounds like the same fundamental problem, but different approaches.
Exactly. There is also plenty of evidence for memory consolidation during sleep, and more recently, there was a study showing that sleep is when your brain processes what you learned and draws abstractions from it.
one major downside with almost all computational models, however, is that they rely on assumptions that the designers can't prove.
I think this is the key. Laypeople think that the reason we don't understand the brain is because it's too complicated. It is complicated, but the main difficulty is the inconvenience of the brain as an experimental system. It's very hard to see what's going on inside a brain without damaging it so it doesn't work any more, so we're stuck using experimental tools that answer the questions we can answer instead of the questions we want to answer. So basically what I think you're saying (and I agree) is that the problem with modeling approaches is that the data isn't there to back them up. I would argue that means the real problem is not with the modeling, but with the experimental side (and I say this as an experimentalist, so it isn't meant in any derogatory way).
We absolutely can model neurons in silicon, and people do, but that's not the hard part. The problem is that there are so many free parameters, and there's no easy way to know what the correct values are. The rate limiting step right now is experimental, not theoretical (of course, this is just my opinion as an experimentalist--lots of reasonable people probably disagree).
Also, just because you can reduce neurons to a simplified model, it doesn't mean that these models, or the thing they're modeling, are simple phenomena. Condensed matter physics (for example) is full of simple toy models that become incredibly complex when they're scaled up, and neuroscience is very much the same way.
I think that's close, but not quite correct. The main theories I've seen in the computational neuroscience literature are that sleep is required for memory consolidation. Since memories are stored by changing the synaptic weights between neurons, you need to be able to send a signal for which synaptic weights to change by how much. However, if you try to do that when the organism is awake, the signal for synaptic changes can't be separated from normal activity, so the system has to be "taken offline" for that to occur. A better analogy would be that your brain is doing a hardware upgrade every night, while you sleep.
You can't base long-term potential solely on what the technology has produced so far. Of course adult stem cell technology has worked better so far--it's easier, with less long-term potential. However, there is an overwhelming consensus among scientists (and yes, I am a working scientist) that ES cells have the potential to be much more powerful in the long term (I'm talking twenty years from now). Science takes time and money, and there's ALWAYS risk involved. If there weren't any risk involved, we wouldn't even need government involvement. The private sector would simply take care of it. However, in order to move things forward, we need to make investments in our future--this is not the same thing as wasting tax dollars. If you want to complain about your tax dollars being wasted, forget about the tiny sliver of the US budget being spent on technology that might save millions of lives and focus on military aircraft designed for cold war-style conflicts where each individual plane costs more than what the entire annual budget of stem cell research in the US would be.
There are a lot of metal ions in your body...sodium, potassium, calcium. Those are all metal ions, and their movement through ion channels is what underlies mental activity, including your perception of feeling anything. Cognitive effects of exposure to extremely strong magnetic fields (as in TMS or extremely strong MR scanners) are most likely due to effects on movements of those ions, while dizziness from the "weak" magnetic fields of fMRI is mediated by effects on mineral crystals in the inner ear. In no case is the effect on water likely to be directly involved. That said, I wouldn't want to put my head in there.
The same thing happened to me once, when I was working to prepare an MRI scanner for an experiment. There was a radiologist there, so I asked him what the mechanism was, and he said it was believed that the magnetic fields affect metal ions in your otoliths, which are the organs in the inner ear responsible for sensing motion. Apparently it's known that some fish and birds have magnetic materials in their otoliths, but I'm not sure if it's ever been demonstrated directly in humans.
You may be correct in your evaluation of the bulbs you mention, but the scientific basis of your argument is incorrect. There are no photoreceptors in the top of your head--those studies were done on birds, who do have photoreceptors inside the head. In humans and other mammals, the only photoreceptors are in the retina. There are several lines of evidence supporting this, including studies on blind people, some of whom have circadian and mood disorders, depending on where in the visual system the damage is located.
The recently-discovered class of photoreceptor cells mediating circadian rhythms, seasonal mood disorders, etc. express a photopigment called melanopsin. The absorption spectrum of melanopsin has been measured, and its peak is somewhere in the blue region of the spectrum, not the near UV. So it really is "more bluish, less yellowish light" that you are looking for.
The defense department does fund a lot of research, and they do have their own research centers. However, it sounds like you're comparing work done at military centers vs. government non-military centers. I'm talking about funding. Remember that the vast majority of all research going on in US universities, both public and private, is directly funded by the federal government through the NIH and NSF. And for your "rest of the government combined" comment, don't forget NASA. I don't remember the exact numbers, but the annual budgets of the NIH, NSF, and NASA are each somewhere in the ballpark of $15-20 billion. DARPA, the research branch of the defense department, has a budget of around $3 billion, which is less than 1 percent of the defense department total.
I'm so utterly baffled at how so many people are complaining about paying taxes, how the government shouldn't fund the NIH, etc. What the NIH gets is somewhere in the low $20 billion per year range (I can't remember the latest numbers...), but the defense department gets around $420 billion, IIRC. The NIH is an investment in the US maintaining its position of economic dominance in the future, and it's a smart investment to make. And to say that private industry could play a similar role is simply not correct. Why isn't anyone complaining that we need to stop wasting money on the military instead?
We are sure of its therapeutic value from animal models. Basically, I think the promise of stem cell research is not being exaggerated. The timescale is, however. ES cell research would not have cured Christopher Reeve, and it probably won't cure anyone for at least 5-10 years after the ban is lifted. I think there is a virtual consensus in the scientific community on those two facts (promise not exaggerated, timescale exaggerated).
Exactly. The President who declared war on science and scientific funding finally sees it in his political interests to start some serious pandering. I wish we had a responsible press that would report on this hypocrisy.
I thought I'd add this, a review published in the BMJ that does a meta-analysis of 214 existing public health studies on public water fluoridation. It basically finds that it is safe and effective.
A friend of mine is going to dental school, and I've talked with him about this before. He said that there's a lot fewer cavities than there used to be, but that's offset by the fact that there's less dentists than there used to be. Also, he said that doing fillings is one of the less profitable things that a dentist can do, and despite the fact that fluoridated water has decreased cavities dramatically, most people will still lose all their teeth to periodontal disease because they don't floss. Interestingly, that's apparently the big unsolved problem in dentistry, to find a high-tech method to replace flossing, which works well, but people simply won't do it, and those that don't do it will eventually lose all their teeth.
Something I see coming up over and over again whenever there's a discussion about evolution is silly semantic argument about the nature of the term "proof." People keep saying that scientists can never "prove" anything, only disprove things. I'm a scientist, and I would argue that this is simply a silly oversimplification without any significant value. The problem is that it implies that absolute proof is somehow attainable in the real world. It's not. The only place something can be proved is in mathematics, and as useful as math is, it's all made up. The inability of science to "prove" things is not a limitation of science, it's a demonstration of the fictitious nature of the concept of "proof" with respect to the real world.
More importantly, people make life-or-death decisions every day of their lives that are based on things they can't "prove." You can't "prove" that a twinkie isn't going to explode, but you eat it anyway. You can't prove that atoms exist, or that smoking causes cancer. By any reasonable standard, those things are considered proven, so one could argue that they're simply "proven beyond reasonable doubt." Likewise, evolution is proven beyond any reasonable doubt. To believe in atoms, but not in evolution, because it's "unproven" or "unprovable" is inconsistent.
Unfortunately, they often have more intelligence than they do humility.
Humility is ultimately what this is all about. When it all comes down to it, lack of humility is the reason people don't believe in evolution. They believe there's something so special about humans that they couldn't have descended from lower primates (or bacteria/whatever), in spite of overwhelming scientific evidence to the contrary. Lack of humility is the fundamental basis of religious thought. Science teaches that we're made out of atoms, and then we die. A lot of people have a hard time accepting that. To claim that this perspective lacks humility is as far from the truth as it is possible to get.
Instead of telling people what to do when they use a computer, give them a list of things NOT to do. The point is, most people are so afraid to break their computer that they don't explore it. The list of things not to do is actually pretty short, if you think about it: don't mess with your system files, don't change your network settings, etc. When people realize that there actually aren't that many actions they need to avoid, they will be less afraid to explore and learn on their own.
Slashdot Mirror
Sensitivity to initial conditions is one feature of a chaotic system, but that alone does not make a system chaotic. Chaotic systems are ones that are capable of generating behavior that looks random, even though the system itself is not. Even from what you're describing, it sounds like cancer cells generate random, unpredictable behavior because the patterns of mutation and aneuploidy ARE random and unpredictable, not because there are chaotic dynamics at work. A more precise definition of chaos is here.
I think you're right that there are anti-science elements on both the left and the right. You say that Christians are targeted more than radical environmentalists, and I agree. However, you seem to be arguing that this is because scientists are all pinkos, and the attacks are politically motivated. In truth, it's because anti-science Christianity is a much larger threat to science in the US right now. If anyone actually listened to radical environmentalists, I'd be out in the street picketing them too.
A second point is that many of the "ethical" objections you're talking about are actually based on misunderstanding, or occasionally deliberate misrepresentation, of the facts. For example, somatic cell nuclear transfer is not the same thing as cloning. Yet some political operative managed to get it branded with the term "therapeutic cloning" or "research cloning" or whatever, which is simply incorrect, and by the time the scientific establishment realized what was going it was too late. How many fewer people would be ethically opposed to SCNT if they ACCURATELY understood what it was? It's the same thing with the "life begins at conception" meme that caught on years ago, in spite of being demonstrably false. The point is, it's not just ethics.
For those of us who work with femtosecond pulse lasers, the cliche is that the ratio of one femtosecond to a second is smaller than the ratio of a second to the age of the universe.
They might actually have some advantage when it comes to early research with using GAs to reproduce NNs. Is the Drosophila nervous system consistent, or is there a certain amount of randomness to it?
The Drosophila CNS is not like C. elegans, where every single neuron is specified, but in some brain regions it's close to that. Also, there are not very many neurons--I think something like 250,000 in the entire brain, and often as few as tens to hundreds of a certain class of neurons in a particular structure, such as the antennal lobe. The other advantages include the huge array of genetic and anatomical tools that can be readily borrowed from the fruit fly community, short generation times, and small brains that facilitate imaging (which is what I do). The one big drawback is the difficulty of doing traditional neurophysiology in the fly. Extracellular recordings are apparently very difficult, if not impossible, by traditional methods, since the neurons are so small. Doing intracellular recordings is also very difficult largely because the organism is so small--it takes skill even to open up the head and take the brain out. Behavioral assays are also a nightmare. So it depends what kind of work one is trying to do.
Right now, the Drosophila antennal lobe, which is their version of the olfactory bulb. To be honest, most of what I've done for the past few years is more like biophysics than neuroscience (the perils of technology development), but I'll be moving back towards actual neurophysiology after I defend my thesis next week.
So, the Izhikevich-type neurons...I actually used those once, a few years ago, when I was attempting to make a model of oscillations in the locust antennal lobe. I actually started with a Wilson-Cowan-type model, then moved on to the Izhikevich model when I couldn't get it to work. It was just a rotation project, so I never had time to make it pan out, but still, interesting stuff.
That sounds like a smart strategy. As a (theoretically-inclined) experimentalist, I'm working on developing new techniques for fast membrane potential measurement using light. Sounds like the same fundamental problem, but different approaches.
Exactly. There is also plenty of evidence for memory consolidation during sleep, and more recently, there was a study showing that sleep is when your brain processes what you learned and draws abstractions from it.
one major downside with almost all computational models, however, is that they rely on assumptions that the designers can't prove.
I think this is the key. Laypeople think that the reason we don't understand the brain is because it's too complicated. It is complicated, but the main difficulty is the inconvenience of the brain as an experimental system. It's very hard to see what's going on inside a brain without damaging it so it doesn't work any more, so we're stuck using experimental tools that answer the questions we can answer instead of the questions we want to answer. So basically what I think you're saying (and I agree) is that the problem with modeling approaches is that the data isn't there to back them up. I would argue that means the real problem is not with the modeling, but with the experimental side (and I say this as an experimentalist, so it isn't meant in any derogatory way).
We absolutely can model neurons in silicon, and people do, but that's not the hard part. The problem is that there are so many free parameters, and there's no easy way to know what the correct values are. The rate limiting step right now is experimental, not theoretical (of course, this is just my opinion as an experimentalist--lots of reasonable people probably disagree).
Also, just because you can reduce neurons to a simplified model, it doesn't mean that these models, or the thing they're modeling, are simple phenomena. Condensed matter physics (for example) is full of simple toy models that become incredibly complex when they're scaled up, and neuroscience is very much the same way.
I think that's close, but not quite correct. The main theories I've seen in the computational neuroscience literature are that sleep is required for memory consolidation. Since memories are stored by changing the synaptic weights between neurons, you need to be able to send a signal for which synaptic weights to change by how much. However, if you try to do that when the organism is awake, the signal for synaptic changes can't be separated from normal activity, so the system has to be "taken offline" for that to occur. A better analogy would be that your brain is doing a hardware upgrade every night, while you sleep.
You can't base long-term potential solely on what the technology has produced so far. Of course adult stem cell technology has worked better so far--it's easier, with less long-term potential. However, there is an overwhelming consensus among scientists (and yes, I am a working scientist) that ES cells have the potential to be much more powerful in the long term (I'm talking twenty years from now). Science takes time and money, and there's ALWAYS risk involved. If there weren't any risk involved, we wouldn't even need government involvement. The private sector would simply take care of it. However, in order to move things forward, we need to make investments in our future--this is not the same thing as wasting tax dollars. If you want to complain about your tax dollars being wasted, forget about the tiny sliver of the US budget being spent on technology that might save millions of lives and focus on military aircraft designed for cold war-style conflicts where each individual plane costs more than what the entire annual budget of stem cell research in the US would be.
There are a lot of metal ions in your body...sodium, potassium, calcium. Those are all metal ions, and their movement through ion channels is what underlies mental activity, including your perception of feeling anything. Cognitive effects of exposure to extremely strong magnetic fields (as in TMS or extremely strong MR scanners) are most likely due to effects on movements of those ions, while dizziness from the "weak" magnetic fields of fMRI is mediated by effects on mineral crystals in the inner ear. In no case is the effect on water likely to be directly involved. That said, I wouldn't want to put my head in there.
The same thing happened to me once, when I was working to prepare an MRI scanner for an experiment. There was a radiologist there, so I asked him what the mechanism was, and he said it was believed that the magnetic fields affect metal ions in your otoliths, which are the organs in the inner ear responsible for sensing motion. Apparently it's known that some fish and birds have magnetic materials in their otoliths, but I'm not sure if it's ever been demonstrated directly in humans.
Also, it's known that the brain can be directly stimulated with strong magnetic fields, as in transcranial magnetic stimulation.
You may be correct in your evaluation of the bulbs you mention, but the scientific basis of your argument is incorrect. There are no photoreceptors in the top of your head--those studies were done on birds, who do have photoreceptors inside the head. In humans and other mammals, the only photoreceptors are in the retina. There are several lines of evidence supporting this, including studies on blind people, some of whom have circadian and mood disorders, depending on where in the visual system the damage is located.
The recently-discovered class of photoreceptor cells mediating circadian rhythms, seasonal mood disorders, etc. express a photopigment called melanopsin. The absorption spectrum of melanopsin has been measured, and its peak is somewhere in the blue region of the spectrum, not the near UV. So it really is "more bluish, less yellowish light" that you are looking for.
http://en.wikipedia.org/wiki/Melanopsin
Of course, I'm not arguing with your evaluation of the bulbs. I have no quantitative data on them, and you may very well be right.
Monoclonal antibodies have to be injected. Generally speaking, they're not effective if ingested.
The defense department does fund a lot of research, and they do have their own research centers. However, it sounds like you're comparing work done at military centers vs. government non-military centers. I'm talking about funding. Remember that the vast majority of all research going on in US universities, both public and private, is directly funded by the federal government through the NIH and NSF. And for your "rest of the government combined" comment, don't forget NASA. I don't remember the exact numbers, but the annual budgets of the NIH, NSF, and NASA are each somewhere in the ballpark of $15-20 billion. DARPA, the research branch of the defense department, has a budget of around $3 billion, which is less than 1 percent of the defense department total.
I'm so utterly baffled at how so many people are complaining about paying taxes, how the government shouldn't fund the NIH, etc. What the NIH gets is somewhere in the low $20 billion per year range (I can't remember the latest numbers...), but the defense department gets around $420 billion, IIRC. The NIH is an investment in the US maintaining its position of economic dominance in the future, and it's a smart investment to make. And to say that private industry could play a similar role is simply not correct. Why isn't anyone complaining that we need to stop wasting money on the military instead?
We are sure of its therapeutic value from animal models. Basically, I think the promise of stem cell research is not being exaggerated. The timescale is, however. ES cell research would not have cured Christopher Reeve, and it probably won't cure anyone for at least 5-10 years after the ban is lifted. I think there is a virtual consensus in the scientific community on those two facts (promise not exaggerated, timescale exaggerated).
Exactly. The President who declared war on science and scientific funding finally sees it in his political interests to start some serious pandering. I wish we had a responsible press that would report on this hypocrisy.
Were the Super-K results also an indirect observation that invoked the Standard Model to imply neutrino mass?
I thought I'd add this, a review published in the BMJ that does a meta-analysis of 214 existing public health studies on public water fluoridation. It basically finds that it is safe and effective.
A friend of mine is going to dental school, and I've talked with him about this before. He said that there's a lot fewer cavities than there used to be, but that's offset by the fact that there's less dentists than there used to be. Also, he said that doing fillings is one of the less profitable things that a dentist can do, and despite the fact that fluoridated water has decreased cavities dramatically, most people will still lose all their teeth to periodontal disease because they don't floss. Interestingly, that's apparently the big unsolved problem in dentistry, to find a high-tech method to replace flossing, which works well, but people simply won't do it, and those that don't do it will eventually lose all their teeth.
Something I see coming up over and over again whenever there's a discussion about evolution is silly semantic argument about the nature of the term "proof." People keep saying that scientists can never "prove" anything, only disprove things. I'm a scientist, and I would argue that this is simply a silly oversimplification without any significant value. The problem is that it implies that absolute proof is somehow attainable in the real world. It's not. The only place something can be proved is in mathematics, and as useful as math is, it's all made up. The inability of science to "prove" things is not a limitation of science, it's a demonstration of the fictitious nature of the concept of "proof" with respect to the real world.
More importantly, people make life-or-death decisions every day of their lives that are based on things they can't "prove." You can't "prove" that a twinkie isn't going to explode, but you eat it anyway. You can't prove that atoms exist, or that smoking causes cancer. By any reasonable standard, those things are considered proven, so one could argue that they're simply "proven beyond reasonable doubt." Likewise, evolution is proven beyond any reasonable doubt. To believe in atoms, but not in evolution, because it's "unproven" or "unprovable" is inconsistent.
Unfortunately, they often have more intelligence than they do humility.
Humility is ultimately what this is all about. When it all comes down to it, lack of humility is the reason people don't believe in evolution. They believe there's something so special about humans that they couldn't have descended from lower primates (or bacteria/whatever), in spite of overwhelming scientific evidence to the contrary. Lack of humility is the fundamental basis of religious thought. Science teaches that we're made out of atoms, and then we die. A lot of people have a hard time accepting that. To claim that this perspective lacks humility is as far from the truth as it is possible to get.
Instead of telling people what to do when they use a computer, give them a list of things NOT to do. The point is, most people are so afraid to break their computer that they don't explore it. The list of things not to do is actually pretty short, if you think about it: don't mess with your system files, don't change your network settings, etc. When people realize that there actually aren't that many actions they need to avoid, they will be less afraid to explore and learn on their own.