Large Hardon Collider

[Kaylia]

^
Kinda boring. It's more fun when they get unexpected result!


Has anyone here modelized a MOF (metal organic framework) before? Our chemist is on a maternity leave, and I'm trying to figure out how to build the cells (x,y,z position of atoms and lennard jone potential) for my simulation program. I'm not even sure where to start lol.

[Psion]

I know, I know, lolpopsci, but I saw this a few days ago and no one posted it, so figured I might as well:

http://www.popsci.com/science/articl...erceive-future

Study Shows Some Evidence Of Human Precognitive Powers
In particular, "participants correctly identified the future position of erotic pictures"
By Clay Dillow Posted 11.08.2010 at 3:00 pm 22 Comments

Paul the Predicting Octopus Perhaps Paul was simply very in tune with his cognitive abilities. Mbz1 via Wikimedia

It’s long been regarded as pseudo-science or simple lore, but precognition – that is, the ability to not just predict but to actually perceive the future – is getting a fair shake in some scientific circles lately. A research paper titled Feeling the Future from Cornell Professor Daryl Bem shows some statistically significant results coming from a series of experiments empirically testing the human mind powers of premonition and precognition. If his results are replicated elsewhere, it may change the way researchers look at the brain, its perception of time, and exactly what its limitations are.

That’s not to say that storefront psychics really can read your palm, or that one can see the future simply by thinking hard about it. But Bem’s empirical, straightforward science suggests the brain does have some ability to perceive what’s coming. The science is sound enough that Bem’s paper found a home in the prestigious Journal of Personality and Social Psychology, which will publish the piece shortly. It also received a fairly lengthy write-up in Psychology Today.

Bem’s research on what he calls psi – meaning “anomalous processes of information or energy transfer that are currently unexplained in terms of known physical or biological mechanisms” – attempts to explore and explain precognition (conscious awareness of a future event) and premonition (affective apprehension of a future event). To do so he conducted nine experiments on more than 1,000 Cornell students.

For an in-depth break down of Bem’s methodologies, you can access a preview of his paper (PDF). But his methodology is consistent throughout: Take an established psychological response to a certain stimuli, then flip it around so the stimulus comes after the response and see if the response is still the same. The results weren’t overwhelming, but they were statistically significant.

For instance, in one experiment Bem gathered 100 subjects, half male and half female. Using a computerized system, they then played a game in which two curtains were displayed on the screen and the subjects had to choose which one had a picture hiding behind it. Some of these pictures were neutral in content. Others were chosen at random by the computer from a database of semi-erotic and erotic photos (hey, looks like science isn’t boring after all).

The result: In cases where an erotic photo was lurking behind the curtain the subjects were able to accurately identify which curtain it was behind with 53.4 percent accuracy – not a huge statistical spike but significantly better than the 50 percent accuracy rate that could be expected by chance. The accuracy rates were not as high for non-stimulating images, which fell more or less in line with raw statistical chance. This suggests that the subjects could somehow sense the erotic stimuli that awaited them before it happened.

In another experiment Bem reversed the priming effect wherein subjects are subliminally tipped off before identifying a photo as positive or negative. Bem found that by subliminally tipping off subjects after – rather than before – showing the image, they still were able to categorize the pictures more quickly as if the brain knew that the subliminal hint was coming even though it hadn’t happened yet.

We won't declare the Earth shattered just yet, but other cognitive researchers are taking notice of Bem's work. Apparently there have been hundreds of requests for replication packages so other scientists can re-test Bem’s experiments and see if the results come back the same. If they do, we may have to reconsider how we perceive our own cognition. But you already knew I was going to say that, didn’t you?

[h+ magazine]

Of course, there's the whole "not big enough sample size, blah blah, but the reality is, sample sizes of greater amounts by far have been done, and get similar results.

Just thought it was interesting it got mentioned at all, since generally science types refuse to touch the topic with a 100 foot telepathic tendril, much less actually properly examine all the evidence for it and research. Like the fiasco with using brain scanning to see if ESP exists, because it "would show up differently on the scans" than normal activity, and finding that it did not. In actually, it did not for all but one married couple, who had an extremely high correlation, but they got excluded from the results completely. Not only that, it was a really small sample size to boot.

But did anyone actually hear about these flaws? No, because the news of it got ignored, everyone just assumed it was a credible study that showed ESP wasn't real when in reality it was a flawed one.

Anyways, there's links in the article that lead to other ones if anyone wants to look at em. <.<b the chances that this study came by chance and not by psi phenomena however, are about 1 in 100 if I remember right. Certainly not as low as some would like, but it's still higher than what most studies require in other fields of science, barring drug testing on humans.

[Max™]

The interesting paper was something where subjects were shown a list of words, then asked to recall them. Later, part of the group was shown the list while the others were not, and the group that was shown the list AFTER they did the recall test got somewhat higher scores than the others.

Still, ehhhh... and this is from a guy who favors adjusted causality rather than non-local interpretations of QM.

[Restrat]

Could be interesting, but actually reading the paper it seems a lot more likely to deal with chance. For example, in the test mentioned in the article, 53% of the time people managed to find the erotic picture, but 48% of the time people failed to find the picture if it wasn't erotic (the article mentions the first number as significant and claims the second is close enough not to matter). That would seem to suggest that we have nearly as strong psychic un-powers, that cause us to predict WRONG if we are looking into the future to find anything but boobs. Seems more like it's sneaky use of statistics

[Psion]

Actually, there are people who score significantly lower than expected in a lot of psi tests, so called "goats", probably because said people tend to be DISbelievers of psi, and yet they score significant... just in the wrong direction. Those who are on the fence tend to score average, and believers score higher than normal. This is going by other experimental data though, not this case. Those who score significantly higher than normal are sometimes called "sheep". So far, from what I remember, psi seems to:

-Ignore time. That is, it can be data from an RNG collected weeks earlier, or after the study, and it doesn't seem to affect results, still getting significant deviations from the norm compared to a RNG that isn't put to the psi test.

-Ignore distance. Regardless of where in the world the test is based on, and subjects are testing from, the results don't seem to vary. It may be that distance does affect it, but it's just not something that diminishes over distance that much.

-Be tied to personality and beliefs. Those who disbelieve psi, and those who are more data loving, number crunching types tend to score lower than expected results. Those on the fence tend to score average, and those who believe in psi/magic/whathaveyou and those who are artsy types, especially musicians, score higher than chance. Also, extroverted individuals tend to do better than introverted individuals. Maybe it's tied to self-confidence or something.

-Be tied to practice. Those who practice meditation, or practice in magical rituals or the like, score a lot higher than chance, even higher than the artsy types. The more practice they have, the higher they tend to score, with some individuals getting outrageous scores. I think in one zener card type test, some fortune tellers who had been practicing for decades got the right card over 80% of the time out of 4 different choices, if I remember right.

-Be increased by number of participants working together, though only if they are compatible. Strangers tend to increase the score slightly, a man and woman put tegether especially seems to do well (though perhaps that's because the majority of people are straight. I'd love to see if they accounted for sexual orientation), and close friends improve scores even more. Married couples tend to score the highest when working together to influence chance.

-Be tied to emotions. A common trend in these tests is people tend to start out scoring really high numbers, then it shoots down afterward to reduce it to near chance or slightly below it. At this point you think it's just averaging out, but when the researchers kept testing past this point, the numbers would then slowly start to climb back above chance, sort of like people are getting a second "mental wind" so to speak. People who are interested in the test, engaged, and positive score much higher, and negative, bored people score lower than chance. So this isn't unexpected, as when a person first tries the test, it's new, exciting, then excitement fades to boredom, then repetition leads to knowledge.

-Be the bane of cynics everywhere. Despite how much they say more data is needed, or ever increasing demands for stricter controls, this slight but significant effect refuses to go away, or even lessen. Sure, some studies will find at chance results, or even lower than normal, but all rigorous studies added together still produce a significant result. This is normal, because different studies will have a different pool of participants, sometimes you get the crappy wimps of the psychic world in one area, sometimes you base a study in Salem and get a bunch of New Age types that demolish chance results. (Note: I'm not sure anyone's actually done a study based in Salem. lol)

It's pretty much like testing for physical capability, then basing your study by testing a bunch of obese nerds for physical strength. Sure, once in a while one of said nerds actually exercises and is fairly strong, but the result isn't going to be very strong. Likewise, most studies are random populations who, for the most part, don't practice psi at all consciously. Of course the results are going to be weak. Studies based on those who practice meditation and the like though, who've been practicing for years, get much higher than normal results however, like in one case where participants went from 25% expected chance and got 32% instead. The fact remains, however, that a significant deviation from norm occurs even in the populous at large, which to me seems promising how powerful it really could be if people figured out an actual best method to strengthening said phenomena. Who knows how strong the max potential actually is, if we can alter chance with our atrophied mental muscles?

[Eliseos]

subjects were able to accurately identify which curtain it was behind with 53.4 percent accuracy – not a huge statistical spike but significantly better than the 50 percent accuracy rate that could be expected by chance.

So 3.4% is significantly better in a study with 100 people? Get the fuck out.

[Ferion]

Seems more like it's sneaky use of statistics

This. A sample size of 100 is no where near large enough to state that a 3.4% deviation from the population mean is significant. The fact that they don't discuss variances at all makes me even more skeptical. I'm willing to bet the confidence intervals for these tests are huge.

Edit: I was reading through the methods and apparently they were getting significant results, but wow do their tests look shady. Especially with their erotic pictures method. They basically were measuring arousal while their test subjects viewed pictures. They stated that when the person viewed the erotic picture that they were aroused slightly before the picture showed up. I have a very hard time believing the statistics behind that.

[Ferion]

So anyone have some tips for epsilon-delta style proofs. I have an exam tomorrow and I'm finding that I'm struggling with the more complicated proofs that require very complicated deltas. Also if anyone has a link to a site that proves limits of functions are unique I would really appreciate it. My professor gave a proof in class but it makes almost no sense to me the way she did it.

[Woozie]

Did your professor provide a study guide with any specific epsilon/delta problems?

Edit: The only real way to get better at those is to do lots of them or to see them done. I'm assuming the ones on your test will be fairly simple since there's a time limit. So there's probably not too much to worry about. If you can do the ones that were assigned as homework, you can probably do the ones on the test too.

If you've already gone through all the ones in your book, you're probably fine. But if you feel like you need more practice, you can check out other books at the library that may have different examples.

And if you're having troubles with anything specific, just post them here.

[Woozie]

SCIENNNNNNNNCEEE

http://www.bbc.co.uk/news/science-environment-11773791

One step closer to world domination.....

This is awesome that we can actually do stuff right now. There's so much we can learn about the asymmetries of our universe with this type of research.

[Max™]

Oooh, now to make anti-matter power plants... and blow up every chem/bio lab with anti-matter warheads!

[Psion]

So 3.4% is significantly better in a study with 100 people? Get the fuck out.

Wait, wut? According to the article, he conducted 9 experiments on more than 1000 people, not 100. That's quite a big difference there. Perhaps you mistook where I said odds of 1 in 100 it was chance results for how many people they tested? Because you don't achieve 1 in 100 odds with 100 sample sizes when it's only 3.4% difference. <_<

[samalis]

Figured I'd ask this here, does anyone have any recommendations on archaeology/anthropology books or videos? Just finished watching The Journey of Man - A Genetic Odyssey. I asked my professor but he hasn't responded yet.

[Ferion]

Did your professor provide a study guide with any specific epsilon/delta problems?

If you've already gone through all the ones in your book, you're probably fine. But if you feel like you need more practice, you can check out other books at the library that may have different examples.

And if you're having troubles with anything specific, just post them here.

She didn't give us anything, just basically said to look over workshop and homework problems. I'm not too worried about epsilon-delta proofs where they give you a function, those are generally easy unless you have a piecewise function. I'm more worried about the heavy theoretical epsilon-delta proofs which she seems to love. The only problem I'm having some trouble with at the moment is this:

Define f by f(x) = 8x if x is rational and f(x) = 2(x^2)+8 if x is irrational. Let p be an accumulation point and p=2. Prove that the limit of f(x) is 16 at p=2.

The proof she gives (I'm just gonna summarize it since it's long, excuse the sloppy language): |8x - 16| = 8|x-2|. |2(x^2)+8 - 16|= 2|x-2||x+2|
So |f(x) - 16| = 8|x-2| if x is rational and 2|x-2||x+2| if x is irrational. So she says set d1 (delta 1) equal to 1. (This is where I start to lose her.) So if d1=1, then (2-1)<= x <=(2+1) and (2-1+2)<= x+2<=(2+1+2). So she sets x+2 = 5. So now we have |f(x) - 16| = 8|x-2| if x is rational and 2*5|x-2| if x is irrational. Now she sets d2= e/10 since 10|x-2| < e. Then she says take min(d1,d2) and basically concludes it the usual way from there.

What I really don't understand is the d1=1 part. I understand how that sets up d2=(e/10), but I have no idea how that shows that the rational part of the function converges. My biggest problem with these types of proofs is knowing when to pick real number values for epsilon or delta instead of having delta depends on epsilon. I summarized her proof pretty heavily, so if something is unclear I'll try to be a little more specific.


Edit: Nevermind I think I figured out why I couldn't make sense of it, I'm just over thinking it. But if anyone has any comments feel free to make them.

[Eliseos]

Wait, wut? According to the article, he conducted 9 experiments on more than 1000 people, not 100. That's quite a big difference there. Perhaps you mistook where I said odds of 1 in 100 it was chance results for how many people they tested? Because you don't achieve 1 in 100 odds with 100 sample sizes when it's only 3.4% difference. <_<

For instance, in one experiment Bem gathered 100 subjects, half male and half female. Using a computerized system, they then played a game in which two curtains were displayed on the screen and the subjects had to choose which one had a picture hiding behind it. Some of these pictures were neutral in content. Others were chosen at random by the computer from a database of semi-erotic and erotic photos (hey, looks like science isn’t boring after all).

The result: In cases where an erotic photo was lurking behind the curtain the subjects were able to accurately identify which curtain it was behind with 53.4 percent accuracy – not a huge statistical spike but significantly better than the 50 percent accuracy rate that could be expected by chance. The accuracy rates were not as high for non-stimulating images, which fell more or less in line with raw statistical chance. This suggests that the subjects could somehow sense the erotic stimuli that awaited them before it happened.

From the same article you linked. And still, 3.4% of 1000 isn't that big either, only 34 people. That's still well within the realm of random chance.

Edit: And if you want your silly Facebook-esque experiments to have more weight, use more than two curtains in this case. Why not use 5, so you have some more convincing evidence that people can somehow sense titties when they are hidden. Oh, that's right. They wouldn't get enough positive data to support their claims.

[BerenTebogo]

Lol pseudo-science.

[Psion]

You're going off the assumption that each person only did this trial once, for a total of 100 trials. I have a feeling this isn't the case, but I'm too busy studying for my trig test tomorrow to look into it. <.<

Also, some studies have been done with 4 possible choices, not 2, and got significant results still. The problem isn't getting enough positive data to support the claims of psi's existence. The combined odds of all credible trials, both those that scored higher than chance, lower than chance, and at chance results, was done by 3 independent math cruncher professionals, whatever they're called <_<; (brains fried by cosines and arctangents and the like atm) and 2 of the 3 found the odds of it being due to chance was 1 in 100 trillion billions, or some absurdly high number. The third one, if i remember correctly, didn't agree, but analysis of his methods revealed it to be flawed.

The main problem isn't proving something's there. The problem is getting the other fields to accept the fact. Scientists who show too much interest in the field are often ostracized by their peers and their reputation ruined, even when there's lots of credible evidence for the fact of psi's existence. No one wants to admit that it exists, not because there's no evidence or that the methods are flawed, but because it would completely screw with everything we've learned and accepted, and require a complete reworking of models in all fields, not to mention throwing a wrench in the "scientist as outside observer of an experiment" idea, if the intentions, expectations, and hopes of a researcher can affect the data they collect.

The lack of funding certainly doesn't help, as again, regardless of the amount of evidence it has, the reputation it has from the early experiments with terrible methodology and famous frauds curses it even to this day, despite the fact that its field overall has higher standards than most other fields of science these days.

No one wants to admit the fact that maybe some of that magical mumbojumbo new age crap might be based on truth, but there's lots of evidence showing that that's just the case, if people just bother to look.

and:

Lol pseudo-science.

This is another common accusation, despite the fact that parapsychology is an actual registered science, and uses proper and rigorous scientific testing. It is a science. In it's early days you certainly could have accused it of that, and even to this day you get tons of new age bullshitters selling spells and the like guaranteed to work, but there's a real field hidden under all the piles of shit that scammers have put on it's reputation. Like it or not, it's a science, and a lot of the methods they use for testing have actually gone on to benefit other sciences as well.

[BerenTebogo]

lol rigorous testing of pseudo-science.

[Eliseos]

From wikipedia because I don't give enough of a shit to look into it further:

Parapsychologists describe parapsychology as an emerging science,[14][15][16] however the status of parapsychology as a science is disputed.[17] While the results of parapsychologists' experiments are regarded by the Parapsychological Association as having demonstrated the existence of some forms of psychic abilities,[18] critics argue that methodological flaws can explain the apparent experimental successes.[19] External scientists and skeptics have criticized the discipline as being a pseudoscience because, as they see it, parapsychologists continue investigation despite not having demonstrated conclusive evidence of psychic abilities in more than a century of research.

http://en.wikipedia.org/wiki/Parapsychology

Also, in actual science, the burden of proof is on the theory. If a study produces conclusive (in science that means just about irrefutable) evidence, then it might be taken more seriously. Until then, it's just an episode of the x-files.

[Psion]

Wikipedia is terrible for conclusive anything, any teacher will tell you it's a starting point, not a conclusive answer to all of life's problems. For example, the citation of people disputing it as a science is cited as coming from an article in 1982. Not to mention, wiki has opposing opinions in practically every article in it, that's hardly something special to parapsychology. The burden of proof may be on the psychologists, but the burden of listening and providing actual feedback instead of "go away, you're a pseudoscience, lalala, we're not gonna bother looking at your data" is on the fellow scientists. And this is where the scientists are failing in their duty.

Almost no one actually bothers to look extensively into the data on both sides with an open mind, and instead they just check wikipedia, go to SCICOP's site, and assume they've done their duty, despite neither being a credible source to judge on. It's pretty pathetic, really. You'd think scientists would be more... scientific.