There’s a fascinating story in the New York Times about a controversial paper about to appear in the Journal of Personality and Social Psychology. The paper’s author is Daryl Bem, a respectable social psychologist famous enough to have made my undergraduate textbooks. He has conducted what appear to be several carefully designed studies that seem to show the existence of “precognition” – that is, the conscious awareness of a future event “that could not otherwise be anticipated through any known inferential process”.
One of the studies involves participants having to guess which “curtain” (displayed on a computer screen) has a picture behind it. Some fraction of the pictures were “erotic”; depending on the exact trial there were images that fell in other categories, such as “romantic but non-erotic” (for instance, the classic wedding kiss), “neutral”, or “negative”. For each category of images other than “erotic”, the participants guessed right about 50% of the time. However, for the erotic images, they guessed right about 53% of the time. The twist in the tale here is that whether the photo was actually “behind the curtain” was chosen by an electronic coin-toss (using a hardware random number generator that relies on quantum effects), only after the participant has made their choice of curtain.
Now, sometimes, when you toss a coin multiple times in a row you’ll end up with more heads than tails. What scientists do in this situation is define a “null hypothesis” (in this case, that participants will guess right 50% of the time) and calculate the probability of getting the results they did, assuming the “null hypothesis” is correct). The average success rate on the porn pics was such that the probability that the “null hypothesis” was true was about 1%. In the sciences, if the chance that the null hypothesis is true is less than 5%, you “reject” the null hypothesis, and accept what’s called the “alternative hypothesis”. In this case, the alternative hypothesis is a rather extraordinary one – that Cornell undergraduates can see into the future, but only when it involves pr0n.
The paper’s acceptance has, of course, provoked a great deal of debate. Few scientists are taking the possibility that ESP might be real seriously. That the paper has made it through the normal review processes of a good scientific journal has provoked a good deal of discussion about the scientific method as currently practiced in psychology, and how non-experts should interpret scientific literature.
The NYT article links to a rebuttal of the paper by Eric-Jan Wagenmakers, who makes a number of important points (it contains a little mathematics, but it’s relatively readable and accessible). Of these, perhaps the most interesting is the propensity of top journals to publish papers that contradict currently-accepted wisdom in a particular field – something akin to “man bites dog” in the news media. However, while these journals may apply the same rigorous peer review to, and require the same level of statistical significance, in such papers as they do to more routine matters, Wagenmakers argues that this is insufficient. In a nutshell, these journals are not applying the dictum that “extraordinary claims require extraordinary evidence”.
The discussion in the rebuttal – and, also, the “room for debate” series at the times – ranges considerably more widely, with a lot of good points being made.
For what it’s worth, while I accept the idea that “extraordinary claims require extraordinary evidence”, there are some problems with it – how do you decide just how extraordinary a claim is, and therefore how extraordinary the evidence is that is required to support it? And “extraordinary claims” might be considered extraordinary because insufficient studies showing them don’t presently exist in the literature – but if studies showing an effect are continually rejected on this basis, how will they ever establish themselves as less extraordinary?
While this study might give heart to a few fans of pseudoscience, I think that the harm on that score will be minor. If it provokes a good hard think about research methodology around the sciences, both physical and social, considerable good will have come out of it. Which, just possibly, might have been Bem’s real intention in the first place.
Postscript: For what it’s worth, I’ve spotted a methodological flaw of my own in Bem’s study, (though other computer scientists will probably already have pointed it out) – and this gets rather technical, so feel free to skip this paragraph as it’s kind of an aside to the rest of the post. As part of the study he uses a software-based “pseudorandom number generator” – in essence, to shuffle the photos – he uses the hardware generator to decide which curtain they appear behind for reasons which are discussed at length and reasonably well in his paper, and his reasoning to use the software PRNG for the shuffling appears sound. However, his choice of PRNG is flawed; while the Marsaglia PRNG is a good general-purpose number generator, it is straightforward for a computer (and, thus, theoretically possible for a human) to figure out the next number in the sequence based on the numbers already seen. As such, he should have chosen a cryptographically-secure PRNG, to rule out the possibility that humans might somehow be able to pick up the patterns in the Marsaglia generator. This can’t explain how his participants were able to precogitate where the pr0n pictures were located, however; merely that it might be possible to identify which picture in a sequence was pr0n without “psi” being involved.
Seems to me the “man bites dog” effect is the key thing here. The experiment showed an effect at the 1% significance level, meaning that this should be observed once in every hundred experiments by chance (and notice that the paper published is really four experiments in parallel).
The question is, how many experiments have been performed by this team (or, more likely) other teams? Since a negative result (being ‘obvious’) is very unlikely to have been published, it’s impossible to know, but it seems quite possible that at least 100 have been performed.
What this really argues for is a contract to make negative results available (as I believe exists in some pharma publications). The idea is that an experiment will not be published unless it was registered before it is begun, with a guarantee to make the results available whether or not they are published). That way the negative results are available for meta-analysis.
Based on reading only the NY Times article, the problem here seems to be publication bias. Bem has presented 9 studies which disprove the null hypothesis, but we have no register of trials so we don’t know how many studies he did that failed, which he isn’t reporting. Even amongst those nine he reports, the probability of falsely rejecting the null is getting quite high, so even if he had just done nine studies and, say, 7 of them retained the null the other two could still be just random chance. And that’s independent of the sample sizes, which seem to be quite small in all of them.
Furthermore, there is no reputable meta-analysis of ESP studies, is there? If these 9 successful papers – even if they were all Bem had ever done – were incorporated into a meta-analysis of the literature on ESP, they’d make zero difference to the finding of its overall lack of existence.
But as the end of the article points out, this really is a case where bayesian statistics are essential. You have to privilege the null hypothesis in studies with no prior theoretical justification!
I haven’t had the chance yet to read the article or rebuttals, but one potential problem springs to mind.
There are actually more than a few academics worldwide who conduct experiments in ‘parapsychology’. There used to be one at Adelaide Uni. I was a test subject for him (about 20 years ago).
Assume that these characters conduct 100 experiments. You’ll get one experiment, on average, in which the null hypothesis is rejected at the 1% significance level. Now if the results of the other 99 experiments are filed in the waste-paper bin, and the 1 unusual one is published …
I had a friend of mine who did some psychological research into brain-waves and tarot readong in the late 70s/early 80s. Can’t remember if it was a third year psych research paper or an hons. thesis, or if in fact the research was ever completed.
Paulus, that’s one of the points made at length by Wegenmakers.
There was an excellent documentary para-psychology released in the 1980′s which I suggest everybody watch before engaging in this debate. The documentary was called Ghostbusters.
“Based on reading only the NY Times article, the problem here seems to be publication bias.”
Publication bias — that’s the term I was trying to remember!
But perhaps we shouldn’t be too sceptical. I am more than willing to concede a special mental connection between undergraduates (particularly of the male variety) and p0rnography. Indeed, I can use my own precognition to foresee that the NBN will be used to bring pr0n to undergrads across Australia!
Incidentally, I recall that the Adelaide Uni guy was very bitter about the marginalisation of para-psychology within the general framework of academic psych.
If anyone wants to do that kind of stuff and still keep their job, they have to make themselves extremely useful to their department as a tutor and marker of 1st year subjects. And they won’t be making it to Professor in a hurry!
On the other hand, anyone who does manage to prove the existence of this stuff will go down in history, so maybe it’s worth the gamble.
That particular article or study may have some issues with it. I haven’t had a good hard look at it but on the surface it just seems … well it doesn’t really seem to say anything at all. Nothing meaningful in the way of establishing a scientific anything.
I’m right into this stuff, but I’m not gonna make a dogmatic case for it here, and honestly I couldn’t care less what other people think about it. Its their business.
I’m interested in what you think sure, and I’d love to have a friendly discussion about it. Provided people understand what I am talking about is to be viewed as speculation from their pov. And perhaps a window into a different way of thinking about the world.
In some ways this goes to the heart of western culture, and scientific materialism.
There is a lot of reason to be both grateful for and sus on the western worldview wrt wyrd shit, and given the context of western history the rise of science is fundamental in the historical vectors (things that had magnitude and direction,) that led to us living in Australia in whats trying to be a free secular democracy.
Anyway i’ll leave it at that for now cos if we’re looking at that study itself I don’t think the discussion has a future. From a quick glance it doesn’t actually say anything meaningful in scientific terms, and a more serious analysis by someone with statistical skillz would probably show that.
However if you want something to ponder think about this:
One thing that kind of kicked me out of my atheist phase was the fact that awesome erotic experiences and wyrd stuff seemed to go together. There were other things too, and weird shit has happened to me all my life. So maybe my heart was never in the atheism thing. But anyway.
I’ve been involved with several “experiments” one me and someone else did that we abandoned cos it got too weird. Others that would be laughed out of here in seconds. One with a guy, a martial arts teacher, trying to test for “Chi”. Its was his honours thesis in a sports science degree.
The epistemological with studying something as nebulous as the idea of “chi” are enormous tho, and they reflect the whole debate.
One more thing. Anything “paranormal” will always raise some serious issues wrt a persons credibility.
This is a “serious political blog”. And fair enough too.
I dunno how far people would want to take a discussion like this could as a society this sort of thing has little or no serious cred.
“epistemological issues…”
“cos not could..”
Edit don’t watch telly.
I should have added, publication bias is a good explanation for why we have no decent scientific evidence for ESP but it’s not actually any argument (within the current psychology publishing framework) for not publishing these papers.
That argument I think rests with the Bayesian stuff, but there’s no consensus in the psych literature (as far as I know) on requiring the validity of the null hypothesis to be incorporated into analyses.
Until such a consensus emerges, I think these experiments need to be published and the psych lit need to take a good long hard look at the issue of publication bias.
Paulus, I was hoping that the study report would be used as a justification for publishing porn in the MSM. Which is like the NBN, but more traditional.
Right now the scientific establishment doesn’t have any idea how to go about measuring claims of or even the presence of “ESP”. (imo anyway.)
It basically has no understanding of what its talking about. And I’m talking about both sides of the scientific argument. For and against.
sg I concede you might be right about publication bias, but you might also have totally missed the point.
What are the papers looking for? What actually is the null hypothesis?
In that spirit here’s an attempted thought experiment.
You and someone else are now (for the purposes of this experiment,) aliens from somewhere or other, studying humans.
Someone else says, “… and their physical dexterity is remarkable. I’ve seen them propel one these over twenty five of their metres with astounding accuracy, sometimes through something this big, time and again.”
As they hand you a sherrin and indicate a shape approximately the size of a car window.
You go down to earth, take a random sample of 300 humans, abduct them, 1 car and a couple of footies (anyone know if this is the plural of footy?).
These rather confused humans are then expected to kick said pills thru said car windows. Cos you’re an all powerful alien with godlike abilities you can actually remove the detrimental effects of the confusion and shock.
You could take random samples of 300 till the sky turned orange and still not find one person capable of doing it. Naturally you conclude someone else is pulling your leg or an idiot or liar.
Perhaps someone more statistically minded than me could point out exactly how big a random sample size of the human population would have to be before you could be reasonably certain of finding one person who could actually kick a sherrin through a car window if you just walked up to them and handed them one.
I really have no idea, nor even how to begin working out how to approach it. (Its quite possible people that play other ball sports could do it, and some people might naturally be able to the first time someone hands them a footy, tho I have never seen it.)
Jules, the experiment quoted there is actually a pretty solid attempt at ESP testing. Unless someone out there who claims to have ESP can do a better suggestion (and they should know how it will turn out, so it shouldn’t be hard for them) then getting someone to predict a completely random event in the immediate future and testing the results should be pretty easy to do.
And the null hypothesis in this case was “none of the test subjects has ESP.” And in fact they rejected that null hypothesis with a well-constructed experiment, so there’s not really much to complain about there is there?
The claim that “scientists are unable to test this” is a really infuriating cop-out. It gives the right to claim anything you want about the physical world without proof or contention. Defenders of anything need to be able to do better than that.
Putting skepticism in brackets for a moment, let’s assume the experiment is “valid”. Even so, it doesn’t seem to me that it’s clear ESP is the only possible cause. It may be, for instance, that reality, or space-time, is structured differently than we think it is. Or it may be that “will” is capable of interacting with the physical world in ways we don’t understand yet. Or it could indicate the existence of phenomena we don’t have a conceptual vocabulary for.
Or it could be that Liam has figured out a very clever way to prank these guys.
What is the point of that random allocation after the choice is made?
Were those pr0n-sniffing UGs as acute when the randomiser was removed from the experiment?
If not, does the randomising program like pr0n?
Katz, I think that’s to ensure that the future was uncertain when the prediction was made. Strictly speaking, if the randomizer was set before the subjects made their guess then they would be intuiting the present, not the future. Presumably that would be a test of a different sort of hokey-pokey.
It doesn’t matter if the randomizing program liked pr0n, because for the UGs to guess the pr0n ones correctly the randomizer had to read their minds before it flipped its pr0n onto the screen. Which is a frankly much scarier thought.
Actually put like that I wonder if there’s a Bayesian rule that can be applied within the framework of the experiment without considering validity of the null hypothesis.
If in this experiment the randomizer produced more pr0n results randomly than average; and if the UGs, being UGs, predicted pr0n more often than average, then surely those effects need to be considered in adjusting the probability of success?
By way of analogy, I’m pretty confident that Paul the Octopus was trained to pick the German flag. Given that Germany win most of the time, it’s very easy to produce the impression that Paul is psychic with this technique. Chances are that Germany will win every game in the group, and then after that they can only lose once. So in the long-run, it looks like Paul is psychic.
But then, as Cthulhu said, in the long run we’re all calamari…
It is a judgment call, but it needs to be recognised that the 5% (or 1%, or whatever) level of significance is also a semi-arbitrary judgment call. One that has become conventionalized in various fields of study, but a judgment call nonetheless.
Robert: The not quite random number comment in your postscript was interesting. Maybe the porn photo was the only one that people were interested enough in to subconsciously see some pattern.
True, but the experiment hasn’t challenged the null hypothesis that the UGs are no more capable of intuiting the present than foretelling the future, in other words that these two skills are independent of each other.
Katz, one should walk before one crawls. Who cares if UGs can intuit the presence of already-existing pr0n? It’s potential future pr0n where the profits lie.
Only if you control all potential pr0n. Otherwise UG libidinousness spirals out into regions that only daleks can control.
(Mmmm … dalek-pr0n).
Strictly speaking if the randomiser is set after they make the guess the make be influencing the outcome through some process we can’t understand. Of course I suggest this specifically wrt the pRon.
the make be… sorry that should read “they may be”.
Sceptics long ago noted that the strength of psi results seemed to be systematically inverse to the degree of rigour of the experiment. It’s an informal version of a more formal technique used in meta-analyses to look for publication bias.
I think that if you do enough well-designed frequentist experiments you’ll eventually get one that gets the result you’re looking for. As others have said, Bayes Law rules. And also when yoyu’re looking for weak effects (as this is) it takes only a small flaw in the experimental design to invalidate the whole thing.
derrida derider, in addition to that most of the tests seem to have small sample size, which will weaken any positive conclusions from a meta-analysis. And I think that it’s well-established now that inclusion of lots of small negative results in meta-analysis, even if they’re weighted for poor quality, can swing it away from a positive conclusion (I could be wrong though).
I still think though that until the psych literature can come up with a good position on how to handle frequentist statistics, publication bias and unusual hypotheses, they shouldn’t refuse to publish on the basis of the hypothesis itself, no matter how wild. It just opens them up to accusations of bias, if not against the wierdo paranormals, just on slippery slope grounds.
Which perhaps implies something about the quality of the psychology literature – if your editorial procedures allow you to publish woo as genuine science, you need to look at your procedures a little closer…
Looking forward to this experiment being reproduced at a research institsute near you! Coming in 2011!!
jaspers @ 14, none of those answers are mutually exclusive either.
But they are all good points.
One of which I was getting at (half jokingly) @22.
Chaos Magic would say (if it was a person with an opinion): “Obviously the subjects determined what was generated “randomly”. The interaction of the random and their will/strength of gnostic trance before each individual event determines those individuals particular magical potency at that moment. That is all this test can measure.”
“Obviously if you can’t see this I have nothing more to say to you” and would then wander off to do something interesting.
sg @ 13.
Is it?
The test is inconclusive for me because there seems to be no way to tell whether the 53% is actually significant. (see Martin B @ 17) And I’m not sure how well they have defined or isolated the relationship the are testing. And what Chaos Magic said too.
Whats ESP?
Ok the test itself is possibly a good one to see if there may be a motivational relationship between a particular form of precognition and the subject of that precognition. Thats it. John D suggested this @ 18 It does suggest there may be a relationship.
It didn’t test for “ESP”. If such a thing exists.
Extra sensory perception.
How many senses do you have? How well do you know them? How do you know knew ones won’t be found tomorrow?
The cultural baggage in the term is ESP is significant, but its effectively meaningless in terms of conveying actual information.
By your definition of ESP I can do it. I wasn’t gonna bother saying that but since I probably have no cred here anyway it doesn’t matter.
Whats actually infuriating is that you didn’t take into account the words “Right now”.
Testing fur ESPz? ur doing it rong.
Is not the same as claiming the right to say what whatever I want about the physical world. At some point I will make that claim, but it’ll be qualified as a thought experiment or something.
I have actually addressed this too, before you brought it up:
I’ve already tried to start doing this. I suggested a thought experiment. I know you wouldn’t have done me the discourtesy of ignoring it, but I do think you are expecting me to say something I’m not saying, and looking for that instead of listening/reading.
Jules, depending on teh background information you have on getting balls through car windows, the sample size for your alien experiment (if they have a beam-me-up thingy with which they can snatch experimental subjects) is between 5000 and 20,000, well within their capabilities. Your sample 300/resample 300 approach is very bad because it increases the risk of spurious significance, and it would require an adjusted analytical method because of the issue of sampling with replacement.
Also the abduction method of sampling raises some ethical issues, but I presume it got through the aliens’ ethics committees, so maybe that part is okay.
I’m not sure what you mean when you say they had no way to test if the 53% was significant. They did, and they found it was significant. I’m not sure what else you mean in that statement.
Regarding Chaos Magic… I think if you read my blog you’ll find I’m intimately acquainted with that evil and those of our sisters who fight it.
Sg cthulu said lots of things…
It did make some nice calamari tho.
The 300 number isn’t important. Its the population of Australia as a proportion of the world if you round the numbers and do primary school level maths, I chose it deliberately becasue it was so low and so potentially useless as a sample, but any number will do.
Those pesky aliens are always up to no good apparently. So their dodgy ethics doesn’t surprise me, my point is … when do you think you or the alien would conclude that humans are capable of kicking a football through a car window at 25 m.
I can do it, well not right now, but give me a month of training. There’s thousands of people who could in Australia who could do it tho, with a 50 to 100% success rate right now.
I reckon I could teach anyone on this website how to kick one through a car do with in a month or two (if you can’t do it already) and after another month of two of practice you’d be able to do it more than 50% of the time. Thats no great feat either. There are probably thousands of people round Australia who could teach that, they do it to kids every year.
The point of that thought experiment is simply this. That alien is the science that’s unable to test ESP. Not cos of some fundamental property of ESP, but cos of how its going about testing. It doesn’t really know what its testing for.
I might add that in the aliens case and the case of science its the same thing that is being ignored. (In the aliens favour it knows nothing of human culture or behaviour and doesn’t regularly devote a huge chunk of effort to developing better ways to achieve what its missing here, in other fields. Unlike science.)
And yes I’m being obtuse cos I’m hoping someone will figure out my point before I spell it out. But if its annoying I’ll spell it out soon.
Re the 53% Is that enough to be significant?
Jules I think what you’re suggesting is that maybe this paper being published now is the paper that finally found someone who could do ESP.
Which is why it’s being published – because it found a result.
So what’s the problem? From the perspective of a chaos magician, science has finally proven the existence of their powers. This article suggests that actually with the right experiment and null / alternative hypotheses, science can investigate even the twilight zone.
If ESP is only possible in some, trained people then science can detect that easily – ask the trained people to submit to experiments. And when they do that, they are always shown to be frauds. So the better scientific approach is to find the talented but untrained people and then burn them alive at the… I mean, test them extensively until we know how it works. Which is I think what this paper could be said to have done?
Unless it was just a lucky result, and doesn’t mean anything at all…
Lets assume that some people really do have ESP (or whatever) Research may fail to demonstrate this for a number of reasons:
The people are well aware of their skills and don’t want others to know about it. (Why get yourself banned from casinos?)
The skill comes and goes – something like hearing something at the extreme range of your hearing.
Emotions may interfere. The lab environment, the researcher or simply becoming wary with the first signs of success.
Only works when there is a motivator. A number of successes may reduce the motivation.
We are talking about a skill that is very very rare.
Finding someone who clearly does have some special skill may provide proof. Not finding anyone doesn’t mean the hypothesis is wrong.
Honestly the perspective of a chaos magician doesn’t give a flying figroot about what science thinks. I just think this may be a case of people who want to “prove ESP” overreaching. Tho the aspect of motivation is an interesting one and may need further investigation.
I mean that about the chaos magician tho, a proper one that is, they couldn’t care less. if something gets shown that is useful it’ll be used thats about it for science tho.
As far as a method of precise investigation of repeatable phenomena so far its unequaled in human history and by the looks of it never will be. As for the limits of what science can say about reality. Thats up to someone else to decide.
How many trained people do you know? I mean properly trained?
(And how would you judge?)
But yeah. I think training effect is something that maybe would be a better way to address the issue. You have to know what to train and how to train it tho. And assuming all this stuff has something to do with brains, and their structure, seeing as everything else we experience does, then perhaps the idea of neuro-plasticity and concept of cultural conditioning. (ie we learn to read at a young age, a fisherman several thousand years ago would have learned to read currents the same way.. etc etc)
All speculation at the moment tho.
Under those conditions it’s a pretty crappy power isn’t it? Less “chaos magic” and more “barely influential magic.” Hardly worth funding scientific studies to find…
Robert,
As I understand it the PRNG was being used to decide which curtain to put the picture behind after the subject had made their choice. I don’t understand why the choice of PRNG would make a big difference? In simple terms you just what the computer to choose 1, 2 or 3 or what ever number of curtains you have.
It’s rather interesting but I suspect that there is an experimental artifact influencing the outcome. I mean, precognition but only on porn?
There’s already a clue in the article. Same design of test failed when the subject wasn’t porn, it’s not mentioned how many of the test results came u with null hypothesis, but it is clearly more than one.
“For each category of images other than “erotic”, the participants guessed right about 50% of the time.”
So this is one test out of how many exactly?
As for the precognition claim- absolute tosh.
The correct alternate hypothesis is that the selections will vary from the statistical average, nothing about mode of action or causality.
“I mean, precognition but only on porn?”
Why not tho?
Precognition on porn given the drive to mate … makes sense if there was gonna be a place where evolutionary stresses would allow that it’d be to do with sex.
@37
Yes but how did they no it was porn? Oh hang on …
Anyway I remain highly unconvinced. Let’s just have 50 people correctly predict the Lotto numbers.
sg @34 I don’t quite follow.
Lets say for hypothetical’s sake that I discovered the power to influence future events. Why would I want anyone else to know that to begin with?
Its a clear advantage in every possible way. If someone else knows I have given away the most powerful advantage possible. What military wouldn’t kill for that power, for example. What gambler wouldn’t love to have it for those times when the money ran out.
People who have a hard on for proving magic by science have some serious issues with their egos imo. They want to be the first, and I wouldn’t be surprised if they spent as much time thinking about the adulation they’ll get when they finally do it as they do about the problems themselves.
But given that may be a grossly unfair generalisation, if there is some value in proving or disproving this stuff, for humanity as a whole, then perhaps a more accurate assessment of what it is we are talking about, and how we go about measuring it might be in order.
Thats why I think looking at “ESP” – or specific “paranormal” processes through a model based on training effects, neuro-plasticity and cultural conditioning might be a starting point. Something to do before people start saying this does or doesn’t prove that.
BTW What I do or don’t believe about this isn’t the issue, it doesn’t matter and isn’t anyone’s business.
I doubt anyone here can come up with anything that would make me doubt “the evidence of my own senses” anyway.
Patrickb at what time during the evolution of older creatures into homo sapiens sapiens would there be an evolutionary stress that gave survival advantages to organisms that could “correctly predict the Lotto numbers.”
I mean go all the back to when the aliens built the moon and cthulu went and hid under the ocean (from me most likely) and I’ll bet you won’t anything remotely resembling a biological necessity to “correctly predict the lotto numbers.”
I’ve never understood the logic behind that argument.
Patrickb: the decision to put it behind a specific curtain was done by a hardware RNG, not a software PRNG. The order of the pictures was chosen using the software PRNG.
If the software PRNG was used to select which curtain was used, this would be a big problem, because PRNGs are by definition not random. Furthermore, the particular PRNG used is not really an appropriate one for the purpose, as computers can look at a stream of numbers produced by it and determine the internal state of the generator (and thus the next number produced). It’s theoretically possible that humans might be able to learn this also (though unlikely, given the relatively short sequences).
Hence, people (without resorting to supernatural explanations) might know whether the next picture is pr0n or not, even if they can’t tell which curtain it’s behind.
@40
Indeed, a Lovecraftian cosmology would be of assistance here. Your point abut evolution is an interesting one. It’s too simple to presume that the precognitive would have had an advantage over their slow-witted cousins (bogans). More likely that the bogans would have reacted with “this prick’s always right, he’s pissin’ me off, glass (or insert prehistoric equivalent) the c*nt”. Thus the precognitive gene was eliminated.
As to the “predict the Lotto numbers” argument, I propose a similar experiment.
Instead of a hardware random number generator, have the curtain which the pr0n appears behind chosen by the results of the next card drawn on an online poker game. Online poker sites use sophisticated hardware RNGs very similar to the one Bem used in his experiments.
As the rebuttal I’ve linked to explains, a 3% edge could be magnified further by letting multiple people “vote”, to the point that you’d have a massive edge over other players at the virtual table.
Microtraders could do something similar – choose the curtain based on whether a stock rises or falls in the half-second between a participants’ choice and the “curtain” opening.
As such, even this seemingly trivial form of precognition would have enormous financial consequence, were it real.
That’s why I remain skeptical.
Not if you knew what they were gonna do and pulled one of those awesome drunken master yang style tai chi chaun moves. And having precog abilities, you’d have known to learn that years earlier. And were a bogan too. You can take the boy out of the west, but you can’t take the bogan out of the boy. And I don’t want to anymore anyway.
Seriously tho it only has to provide a slight but useful advantage over time. think about cats and how jumpy they are for a moment. That seems useless in a domestic context, but all the non jumpy cat genes were eliminated cos there were times over a long time when that hyper nervousness enabled those jumpy cats to just get out of the way a predator.
And I know that cos one of those cats from Saturn told me. You know the ones? The ones that don’t like the Ulthar cats? They find it extremely annoying now cos they can’t sneak up on earth cats and kill them. (I personally prefer the earth variety to those unwholesome Saturnian moggies. Cept for dinner. I’d try not to eat an earth cat.)
But yeah wrt to the actual study itself I had a feeling Robert would say something like that.
So, like this is all about how crap modern science is these days… Say it ain’t so? What, people are falsifying results?! Just to try and get their hands on the Money?!!
I don’t believe it…
I can’t believe it: If we can’t trust the scientists, who can we trust?
(I mean, even if this experiment is formally correct and deserves to be reproduced independently, it’s a complete load of crap. B-but he’s getting publicity and he’s publishing. And, more importantly, I think he’s being deliberately provocative (I don’t know, just guessing) by showing how many of the so-called academic publications these day are just noise. What about those academic contracts huh? Certain number of publications per year or you’re phupped… And don’t get me started on the agenda setting behind the private/ industry-backed research grants system…)
We are going to sh!t in a hand basket. That’s right. It’s a doing word.
Joe, as I indicated, Daryl Bem is an emeritus (retired) professor of considerable reputation. He has nothing left to prove, and presumably doesn’t apply for grants any more (and would have no trouble getting them if he did). None of the paper’s many critics have accused him of falsifying his results, or conducting his experiments in anything other than a careful manner.
The problems which you mention are very real, but I don’t think they’re applicable here.
Robert,
yes, but what a great time in his career to pull a fast-one. I take this as being a critic on the whole edifice. May it come crashing down. And soon.
Cats from fucking Saturn? You can’t be serious.
Yeah, Paul, with you there. This thread has rapidly become a trainwreck, albeit quite an amusing one.
However, any thread which starts from the premise that social psychlogy is a science, and that “respectable social psychologist” is anything other than an oxymoron, is bound to go off the rails and down the ravine sooner or later. Whatever it is that a couple of commenters seem to be on is a sufficient but not necessary condition to achieve that.
Wozza,
Its not that I don’t believe in precognition or telepathy. I mean, for reasons that are inexplicable to me, tarot cards can foretell the future, but one doesn’t realise with what accuracy until the event is actually occurring. Some people are precognitive enough not to board a plane/car/train etc if its going to crash; if you’re really, really close to a person you can have a telepathic connection with them; etc, etc. Whether any of these phenomena can be reproduced in a rigorous scientific environment for testing purposes, I don’t know,
But there’s a difference between being open or susceptible to psychic phenomena in a grounded way and being – well, more than a little out of touch with reality.
Paul @ 49 of course I can’t be serious. I was just taking the piss out of the whole Lovecraftian metaphor thing, cos I use them all the time in meatworld and at a couple of other internet sites. The cats from Saturn are part of the Lovecraft Mythos. They’re bad cats tho.
I am serious about the different methodology and the need to approach psychic phenomena from clearly defined grounds if you want to build reasonably serious attempts to try and quantify and measure what they involve. If they exist.
And one of those grounds would be to examine whether there might indeed be an evolutionary place for psychic phenomena and if so what is it?
jules @ 52,
I don’t know if there is an evolutionary place for pyschic phenomena. In my limited experience, unless you get into the serious business of fortune telling (which I would not recommend, actually)it seems to be an intensely personal experience, appropriate for a particular moment,eg clairvoyance) or with a very close relationship with a special person (eg telepathy). It is perhaps because of its personal nature – I’m sure heaps of people have got anecdotes – that it may be difficult to subject to rigorous scientific analysis.
OTOH, I don’t want to suggest intuition can’t be tested – in the sense, if you start doing tests etc, the intuition suddenly disappears. Thats rubbish, an excuse used by frauds conning a gullible pub;ic. I’m sure if it is there it can be found. The question is how.
Further to that any serious examination of cultures other than this one would reveal that they have a cultural conditioning program that would be similar to the sort that young athletic males in Melbourne recieve.
Except for the content.
I’m making the generalisation that the vast majority of non mestern cultures have some strongly held “supernatural” or “paranormal” or other non rational belief system. If someone can show me 50 that don’t, then obviously its not the vast majority (at least as vast as I meant) and I’ll withdraw the statement.
Both situations (footy in melb and most non western cultures) involve a culture that accepts talks about and examines the content. There are traditions and techniques, and a body of knowledge that are usually passed on person to person and by observation. Both involve developing schemata revolving ritual (football is a ritual, especially if you play it). Both involve training. Both have certain moral codes, (but lets not judge any right now.)
Etc etc.
Everything we do involves our brains so its reasonable to assume that if there is a physical basis in the brain for psychic activity, (and lets face it why wouldn’t there be – its how we experience everything else,) then its subject to some form of neuro plasticity or other learning related change.
So there is room to really examine if any of those things contribute to psychic phenomena, by measuring stuff associated with them and psychic phenomena.
The only effective way to do this scientifically is to find an effective training method, train people, compare the results and if there are any of note repeat the process in every possible way till you have put the idea through the wringer.
If anything is left then there might be a scientific basis for psychic or paranormal things. Tho by that stage they’d just be normal.
Missed that @53. Ok obiviously I’m kind of right into this.
But one of the reasons is wyrd stuff kept happening I couldn’t ignore.
It might be a family trait.
Turns out my dad was a hindu priest as a kid. One of the things Hindu priests do, or did in his particular case was go into mildly trances epileptic states and channel people’s ancestors or hindu deities. I used to think it was solely trance based. I’ve seen this stuff too and I’ll tell you something is going on, but I don’t know what. He never mentioned anything till he realised I’d spent years studying all sorts of other traditions of weirdness.
Epilepsy is weird. It refers to a series of syndromes with a similarity – unusual brain activity and associated lesions. Beyond that its hard to say alot cos different forms have different freqquencies and intensities of electrical interruption.
My wife has it. Tho she didn’t exhibit noticeable seizures till her 20s. Once she learned what auras were she recognised them and said they dated back a long way. Long before the seizures started. This is the sort of the thing that would probably help with early intervention if it ever gets to the point where thats possible and useful.
Her form is serious. Infrequent seizures that seem to be controlled but have of high proportion of status events – seizures lasting over 30 minutes. This raises the issues of all sorts of complications. Sometimes the only way to stop it in her case is to induce a coma. Once this happened at the botton of the track at Natural Arch on the Qld/NSW border at the top of the Numinbah valley. At the glow worm cave. I thought she was gonna fit in my arms till she died. We don’t know lucky we are in Australia.
So understand I take her condition very seriously. When someone told me vaccines caused it and she’d be better off if she stopped the medication and did breathing exercises I had to laugh. Or else.
Cos funnily enough…
Before either of us suspected anything about her condition I took her along to someone I was learning “Tai Chi” and the related stuff from. This person was good too. They understood chi and could communicate what they were on about. They also knew their stuff wrt forms, Tai chi is a dangerous martial art if you learn all the stuff that gets stripped.
We were doing a particular chi kung exercise involving a stance and a hand position, sometimes called a mudra in other eastern traditions. The point of this exercise was to balance the chi flow on between both sides of the brain.
Within seconds of beginning the exercise my wife got sick and exhibited what we’d later recognise as a severe aura and associated nausea. MY teacher at the time said: “There’s something serious in her head. You need to get that checked out soon.”
Naturally her response was to freak out and ignore it and mine was to try to forget or at least try not to think about it.
Five years later she started having seizures.
Most people look at chi and scoff. I don’t care cos it works for me. You can think what you like.
I mentioned all this to her specialist once. He was polite and not that dismissive. I wasn’t adamant either – raised it cos it really is something that intrigues me. He didn’t know what to say tho. Fair enough, its far from his field and what he does. Which he is damn good at too. Because western medicine works in this case. He’s head of dept of the school of med at some Gold Coast or Brissie Uni.
That anecdote of mine is functionally useless to him. However much he might be prepared to accept my word it happened what can he do with that info?
If someone were able to find a basis for whether or not “chi flow” was a measurable thing then there is a potential to use that exercise that made my wife ill to check for the same reaction in others, and then make meaningful statements that could lead to early intervention. Well that could be useful potentially. Perhaps her neurologist could use that info for other people’s benefit.
I know how fraught this in light of “quackery”.
Just so its clear where i stand on the complicated issue of scientific method and the paranormal.
Jules @37
you don’t get to reproduce more sucessfully by looking at porn. Therefore no Darwinian advantage.
Hate to break it to you, it takes two to tango.
Well obviously, when humans were evolving there was no porn.
Are you saying that there is no connection between humans watching porn and what motivates them to have sex and try to breed?
Jules, the powers you describe are too weak to be worth studying, hence why bother applying for grants or trying to get scientific respect for this field of study?
Wozza’s Ghost, as usual showing your ignorance. I’m happy to agree that lots of social psychology is a crock, but there’s some very useful stuff done in that field – e.g. everything we now know about addiction. Not that it’s worth taking pointers on “respectable” science from a ghost with the kind of views you hold about climate science.
I mentioned this “issue” to my partner, a journal editor, last night, and she said “They should just tell him it’s beyond the scope of their journal and to publish it somewhere else.” Problem solved… how much of this “issue” has arisen because the researcher in question can get himself considered for publication in a journal just through his name?
i’m just hoping this is some awesome troll to show the problems in the scientific publishing process that creates data mined results. how many different people have run different pre-cog experiments and received a null result that we haven’t heard of? it’s not surprising that one eventually lucked it.
Things get a bit fuzzy when you are near the confidence limits, more so than far too many researchers realise. This issue plagues medical research as well
Issues such as:
Outlier treatment. Now removing outliers is a well worn technique, but if overused can end up changing the distribution, either variance or even type. I saw one example where a (statistician no less) removed so many outliers that they changed the TYPE of the distribution.
What is the underlying distribution? Far too many people automatically assume a gaussian and treat it as such. But what if it not? The poster children for this are the various financial risk models and things like Scholes-Black, all assume a normal dist, sadly the data is not so compliant and is definitely NOT normal. So risk was understated by orders of magnitude, hence the GFC and Long Term Capital went under.
Sample size, too many studies with far too small sample sizes. ‘I got a result with 10 people’, for example, is meaningless, especially if you remove outliers as well.
Unconscious bias, combined with outlier removal. This leads to target tracking, where (often without realising it) you tend towards an expected, or hoped for, result. Read about the search for the value of an electron charge, how people only slowly moved from Milikan’s original result to the more accurate one we have today (no criticism of Milikan, with his equipment he did a fantastic job, many subsequent researchers don’t deserve such plaudits though).
Subtle experiment biases, where small errors in your design (each independently insignificant) can combine to a create a bias.
Lack of a real understanding of statistics, which is not as well developed as most people think*. Too many don’t really understand the maths and conscious/unconscious/hidden assumptions that underlie a particular technique. Sadly far too many just push the data into a stats package and read off the result.
In ESP research ‘leading’ is a real problem. People are incredibly good at picking up even incredibly subtle clues about how they should respond. I remember (sort of, there was a lot of wine) Susan Blackmore regaling me with stories about various studies she had debunked because of this reason.
So I remain very sceptical and expect (as has happened many times in the past in this area) that future repeated studies find contrary evidence.
*Look up the distribution of a ratio for a frightening example.
sg, do you have any debating tools other than accusing me of “as usual showing your ignorance” on any and all subjects? You know, minor stuff like facts, intellectual rigour, knowledge of the subject?
This thread is done and dusted, and I should leave it alone, but just for future reference, I have a PhD in a “hard” science discipline (physical chemistry). I do have some idea of the scientific method. That is why I am convinced that social psychology is not a science.
I would interested to know what your own qualifications are for being so dogmatic about your superior understanding of what is and is not science. Not very many I would venture to guess, if it is your practice to wheel out the views of a “journal editor” as evidence for your science dogma.
That is not to say that it is not possible to do “useful stuff” in the field – and if you ever bothered to read what I actually write, instead of mentally substituting what your own prejudices suggest are what someone with political views to the right of Lenin might write, you would have noticed that I did not say that – in the some way as it is possible to do useful work in other non-scientific disciplines, or even in stamp-collecting or educating sg (the last of course would present just as many difficulties as proving the para-normal).
As for climate science – and it is your chosen comparison, not mine – it is a subject for another day, but since you raise it, yes, there are regrettably a number of unscientific aspects about the way it is currently carried out. Basically, of course, it is about physics, and what it has managed to firmly root in physics I generally support. Unfortunately, it increasingly strays out of that area, and far too many “climate scientists” have insufficient expertise in it.
He has furnished the details on how to do these experiments to dozens of interested people, so a flood of test results should eventually be available. If a large number of them show an above chance outcome, then it reinforces what he has shown. If not, PSI gets knocked down once again. And it’s incorrect to say the experiments only involved erotic images, the ‘post priming’ test involved pleasant/unpleasant images, not erotic images.
Hmm,
Second Law Violation Procog not possible:
However.
http://www.hplusmagazine.com/editors-blog/precognition-real-cornell-university-lab-releases-powerful-new-evidence-human-mind-can-
Huggy