You saw this post coming

November 11, 2010 10:50 PM

An eight-year, extremely large study (p = 1.34 × 10^-11) has found statistically significant results that point towards a human capability for precognition. Reviewers for the Journal of Personality and Social Psychology are puzzled by the paper but cannot find any flaws in its methodology. Is this confirmation of the fluid nature of time? Or is it simply another candidate for the Journal of Irreproducible Results?
posted by shii (100 comments total)

I knew it!
posted by BitterOldPunk at 10:51 PM on November 11, 2010 [32 favorites]

I know how this thread will go.
posted by vidur at 10:56 PM on November 11, 2010 [1 favorite]

I dreamed about this when I was eight.
posted by Halloween Jack at 10:58 PM on November 11, 2010 [3 favorites]

I've always kind of wondered how well proof any sort of paranormal phenomenon would go down if it was done correctly by tenured researchers. I guess now we'll find out (or, apparently, witness facts of which we already had some extremely dim perception?).
posted by Ryvar at 10:59 PM on November 11, 2010

Meh. What of the myriad years of study dedicated to parapsychology (a now debunked pseudoscience)? Parapsychologists tried for years to proves that humans had some kind of psychic ability, to my knowledge they proved nothing, only that parapsychology was a complete waste of time. I don't think this proves anything definitively, interesting to be sure though.
posted by IvoShandor at 11:00 PM on November 11, 2010

Maybe, delmoi, but I guess we'll have to wait for follow-up studies to find out.
posted by No-sword at 11:01 PM on November 11, 2010 [79 favorites]

vidur: "I know how this thread will go"

I saw that comment coming a mile away.
posted by Blazecock Pileon at 11:01 PM on November 11, 2010

If you rearrange the letters in "Daryl Bem", you get "Hari Seldon".
posted by vidur at 11:01 PM on November 11, 2010 [5 favorites]

I wish I could say I saw this coming, but it's right out of the blue for me..
posted by Ahab at 11:04 PM on November 11, 2010 [4 favorites]

Blazecock Pileon: I saw that comment coming a mile away.

Exactly.
posted by vidur at 11:07 PM on November 11, 2010 [2 favorites]

Extraordinary claims.........
posted by Long Way To Go at 11:08 PM on November 11, 2010 [2 favorites]

As a skeptic that's had a few inexplicable experiences that would hint at something like precognition I've come to at least allow for the possibility that some part of our psyche is not constrained to the regular flow of time. I mean my understanding of science leaves room for these experiences not being magical.
posted by mhjb at 11:09 PM on November 11, 2010

There is a shout out to this organization in the footnotes. Not a great sign, given it's history.

That said, I hope this is taken seriously and investigated further. It's about as radical a claim as I can imagine though, so it's going to take a lot to convince me there is anything significant here. If had to guess I'd say that individuals were somehow conventionally tipped off to the upcoming practice session.
posted by phrontist at 11:12 PM on November 11, 2010 [2 favorites]

well... it's a first step, but seriously I won't get all that interested until it's replicated. Essentially it's not true unless someone else can do it too.
posted by edgeways at 11:17 PM on November 11, 2010 [1 favorite]

My understanding of science leaves room for these experiences not being magical.

What kind of experience could ever qualify as magical? It's conceptually confused... a holdover from days when we had a worldview split in to two parts "natural" and "supernatural". As "natural" is used by scientists today, "supernatural" is meaningless by definition.
posted by phrontist at 11:19 PM on November 11, 2010 [6 favorites]

vidur: "I know how this thread will go."

And I know how it will end:


« Older A charity auction whose grand prize was a busines... | Race to Nowhere (trailer) is a... Newer »

This thread has been archived and is closed to new comments

Related Posts
Study: Lesbian Parents Raise Better-Behaved Kids June 7, 2010
he of the weird al hair and santa claus beard December 20, 2009
Behind Door Number One... April 8, 2008
The Men Who Stare At Goats November 26, 2007
Is God An Accident ? - Long Version November 24, 2005

posted by Rhaomi at 11:24 PM on November 11, 2010 [41 favorites]

There is the flip side to precognition here - the subjects may be influencing the results so that the future random selections are not truly random. This would also be considered a "psychic" ability. My point here is that the events are correlated, but causation - precognition versus manipulation - has not been established.

I have a pet theory that I trot out quite frequently. Murphy's law says that if it can go wrong, it will. My pet theory I call "inverse causality" is an extension of that. If you do something, or neglect to do it, it actually causes things to go wrong. For example, if you need to go get the mail, but you leave your keys, thinking "ah, the front gate is unlocked, I don't need them" you actually cause events to occur that lock the gate. You get out to get the mail, and the damn gate is locked. On the other hand, if you say "ah, the gate is unlocked...but I will take my keys anyway", you'll find that the gate is indeed unlocked, and you can get to your mailbox without trouble.

Once I embraced this theory, I have found my life is somewhat less troublesome. And I never go anywhere without my keys.
posted by Xoebe at 11:28 PM on November 11, 2010 [26 favorites]

"The sequencing of the pictures and their left/right positions were randomly determined by the programming language’s internal random function"

Bwahahahahahahaha.

Well there's your problem!
posted by phrontist at 11:34 PM on November 11, 2010 [17 favorites]

Does this mean it's okay to have ESP in science fiction again? 'cause I miss some of that stuff.
posted by Zed at 11:35 PM on November 11, 2010

Aaaaand, I'm a sleep-deprived idiot for posting so quickly. They used one of these hardware RNGs as well, so nevermind.
posted by phrontist at 11:37 PM on November 11, 2010 [2 favorites]

well... it's a first step, but seriously I won't get all that interested until it's replicated. Essentially it's not true unless someone else can do it too.

Sure, but we already know that it is going to be successfully replicated.
posted by b1tr0t at 11:38 PM on November 11, 2010 [1 favorite]

"I don’t believe I have any particularly strong psi abilities myself, but in my everyday life, I’ve witnessed some rather striking examples of psi phenomena involving others. For instance, one day a few years ago, a friend and I were walking in the forest with her beloved dog, and the dog (as was common) ran far away from us, exploring the woods and chasing animals. Then, all of the sudden, my friend said, “She [the dog] is looking at a turtle. I can see it right now as if it were in front of me.”

I was understandably skeptical: “Yeah right. How could you know?” Turtles were not that commonly seen in those woods.

I was going to call the dog, but my friend asked me not to. Instead we quietly looked for the dog, and she was about 100 feet away staring intently at a turtle, which was sitting there peacefully by a stream."

It never occurred to the author that the friend knew the dog had seen the turtle before and was interested in it, and was playing a joke? If you look at the graph of who believes in precognition, you'll note that psychologists do not buy it, in stark contrast to everyone else. That's because they know that believing you're psychic is one symptom of schizophrenia.
posted by BuddhaInABucket at 11:39 PM on November 11, 2010 [3 favorites]

J.W. Dunne's An Experiment With Time is a great book for anyone with a bit more than a passing interest in this subject. Dunne's thesis is that in the same way dreams can be influenced by the recent past, they can also be influenced by the the immediate future. He uses years of personal dream journal entries as his data set.

Another neat thing is the Princeton study on Retropsychokinesis: the idea that not only can the mind influence remote events, it can violate the order of causality to do so. Their results were also statistically significant.

If that tickles your fancy you can try some RPK experiments of your own.
posted by clarknova at 11:39 PM on November 11, 2010 [8 favorites]

In one experiment, students were shown a list of words and then asked to recall words from it, after which they were told to type words that were randomly selected from the same list. Spookily, the students were better at recalling words that they would later type.

I wonder if you could actually put this to practical effect in the real world. Say you are a student studying for an exam. Would your recall in the exam room be improved if, in addition to studying before the exam, you went over the material one more time after the exam (analogous to typing the words after the recall test in the study)? Does the effect stop working if you know about it? Does it only affect which items you get correct, instead of increasing the overall total? What if this works, and you take the exam intending to do the post-exam crazy-time-paradox retro-active swotting, but then you decide to blow it off to go drinking? But your answers are already determined by then!
posted by L.P. Hatecraft at 11:43 PM on November 11, 2010 [9 favorites]

Does this mean it's okay to have ESP in science fiction again? 'cause I miss some of that stuff.

Man, I hope so. Back when I was an awkward, weirdly intense adolescent girl, there was nothing I liked better than reading about awkward, weirdly intense girls from the future with GIANT HAWKING SUPERPOWERS O'THE MIND.

That stuff still gives me a delightfully cozy kind of thrill.
posted by palmcorder_yajna at 11:47 PM on November 11, 2010 [6 favorites]

From the paper:

In the experiment just reported, for example, there are several possible interpretations of the significant correspondence between the participants’ left/right responses and the computer’s left/right placements of the erotic target pictures:
1. Precognition or retroactive influence: The participant is, in fact, accessing information yet to be determined in the future, implying that the direction of the causal arrow has been reversed.
2. Clairvoyance/remote viewing: The participant is accessing already-determined information in real time, information that is stored in the computer.
3. Psychokinesis: The participant is actually influencing the RNG’s placements of the targets.
4. Artifactual correlation: The output from the RNG is inadequately randomized, containing patterns that fortuitously match participants’ response biases. This produces a spurious correlation between the participant’s guesses and the computer’s placements of the target picture.

My fucking god. These are the 4 explanations you could come up with, Dr. Bem? People get paid for think up shit like this? Fucking pseudoscience. Fuck it.


As for the two experiments outlined in the Psychology Today (sigh): it's not that hard to think of neurobiological mechanisms, e.g. learning interference, for how it works. If you instead bypass those and favor explanations based on time-travel, there is something seriously incomplete with your thinking style and educational background, sorry.
posted by polymodus at 11:53 PM on November 11, 2010 [10 favorites]

New Scientist link says subjects guessed a fifty/fifty result correct 53% of the time. This is all I need to know the results won't be repeatable or every casino in the world would be bankrupt.
posted by bystander at 11:58 PM on November 11, 2010 [1 favorite]

For those who aren't going to RTFA, the paper is arguing (among other things) that humans are able to perceive the presence of hidden sexual stimuli slightly more often (about 3%) than other stimuli.

So, if the study is found to be true, then it will be revealed that we humans have psychic powers, but they are very, very weak (almost imperceptible) and can only sense sexual feelings.

That would be a very interesting finding, but I don't think it would Shake The Foundations of Knowledge As We Know It.
posted by Avenger at 12:01 AM on November 12, 2010 [2 favorites]

An eight-year, extremely large study (p = 1.34 × 10-11)

a p-value is a measure of statistical significance, not sample size.
posted by ChasFile at 12:11 AM on November 12, 2010 [16 favorites]

See here for a nice paper by some Dutch psychologists critiquing the Bem paper's statistics from a Bayesian point of view. Though some of their argument boils down to the assertion that a p-value threshold should be very low indeed for such extraordinary claims, they also make good points about exploratory data analysis and have an interesting discussion of Bayesian t-tests. For my money, while these are valuable points to make, I'm content to wait for the many follow-up studies which will definitively validate the results or not without need for fancy statistics.
posted by chortly at 12:13 AM on November 12, 2010 [14 favorites]

More than 50% of natural scientists believe in precognition???????
posted by scose at 12:16 AM on November 12, 2010 [2 favorites]

The problem with prophecy is that your initial prophetic glimpse of your future contains a "you" who was not aware in advance of what would happen. Having made your prophecy, you would then be able to anticipate the events, presumably to your advantage, and invalidate the very prophecy that advantaged you. Unless of course there is no such thing as free will.

This is true to a lesser extent in more mundane prediction theory: if you were able to successfully model, say, the movement of a particular stock in the market, or a horse in a race, you could profit from that. But by doing so, you have moved the stock or odds a tiny bit (or perhaps a really big bit, depending on who you are and how much you spend) away from your predicted result. In theory you're fine with that, because in theory you've made your profit and you can just run your model again to account for your own transaction. But in practice, any known successful predictor of anything would immediately acquire a crowd of followers seeking to emulate the predictor's profits, and thereby destroy them.

In other words: successful prophecy drives itself out.
posted by aeschenkarnos at 12:21 AM on November 12, 2010

In the history of the human race, we've seen a total of maybe two things that fundamentally alter our understanding of how reality actually works. We've also seen countless millions of experimental defects and just plain fraudsters. A claim that the arrow of time can point backwards at macroscopic scales through an unknown mechanism specific to humans is going to need so much evidence in its favor it's hard to imagine even caring about this paper yet unless you're actually in the field and have a grant to try to replicate it.

If you show me a guy with a shaggy beard, long fingernails, and a penchant for howling at the moon, he needs to actually turn into a wolf before you'll convince me he's a werewolf rather than another crazy street wanderer.
posted by 0xFCAF at 12:30 AM on November 12, 2010 [12 favorites]

1. Whenever someone says something like "there are several possible interpretations" and then gives a list, be on your guard: the list is almost guaranteed to be non-exhaustive.
2. Were we to find an effect (which I so-far-beyond-highly doubt), it would not be a "supernatural phenomenon", it would be a "natural phenomenon that has previously not been formally described".
3. How many other researchers have done similar work, and we haven't heard about it, because they've been unsuccessful? Journal publication is rather self-selecting: just try submitting a paper with an abstract of "I did extensive work on time-traveling thoughts with a thousand students, and found no evidence of it" and see if it gets published. 53.1% is impressive, but as you increase the number of samples, it becomes less and less likely that something of a given unlikeliness would not happen.

posted by quarantine at 12:45 AM on November 12, 2010 [8 favorites]

I wrote this on a Polaroid once, but I burned it.
posted by iamkimiam at 12:56 AM on November 12, 2010

I wrote this on a Polaroid once, but I burned it.

I think the "once" is "tomorrow", so you're in the clear.
posted by quarantine at 12:59 AM on November 12, 2010

I predict that lots of people here will make "I knew it!" posts.

What? Of course I haven't already looked at the previous comments!
posted by Decani at 1:05 AM on November 12, 2010

They've done some weird shit with entangled quantum photons, where a photon in the present is able to react to events that have not yet happened to its entangled twin. So we already know, from hard experimental evidence, that time is not linear in the way that we perceive it. An extrapolation that the brain is somehow getting a weak link into future photon state collapses isn't even particularly farfetched. Assuming that the experimental evidence continues to hold up, this could be a fascinating area of research.

Some immediate variations on the experiment that come to mind.

First, what changes when the random number generator changes? If the RNG is derived from a single quantum state, do the results change? How about if it's decided with, say, the lottery generators that use ping pong balls? That's apparently random, but very much a macro effect -- it would be very interesting to see if the results were any different. Does anything change if the numbers are generated before the human's choice is made? How about if they're written down and manually pre-keyed in by a human?

As a corollary, it would be very interesting to see how well humans do at manually picking lottery numbers -- I imagine the lotteries store every ticket, probably for years, so that would be a rich data source from multiple angles. One example; does being closer to the lottery draw make predictions more accurate?

Second: what changes if the humans are encouraged versus discouraged? That is, if the humans are told, "We're seeing very interesting results so far, people are doing noticeably better than average in our study," does that change the outcome? How about a neutral or a negative pronouncement? I'd be very interested to see if positive or negative expectations improve outcomes.

Another suggestion would be to maybe try for the occasional fear reflex as well, maybe a loud sound from one side or the other at the same time as the picture. If humans actually do have any kind of precognitive power, I'd expect it to be closely tied into fear, maybe more so than sex.
posted by Malor at 1:08 AM on November 12, 2010 [4 favorites]

An extrapolation that the brain is somehow getting a weak link into future photon state collapses isn't even particularly farfetched.

Of course it is farfetched. It's not completely impossible, I guess, but "farfetched" doesn't mean "absolutely impossible". But it's certainly extremely farfetched and almost certainly not true.
posted by Justinian at 1:16 AM on November 12, 2010 [6 favorites]

An eight-year, extremely large study (p = 1.34 × 10-11)

a p-value is a measure of statistical significance, not sample size.

Oh thank god. I was sitting there going "I thought it was 'n'...I thought it was 'n'...I'm losing brain information...oh wait!
posted by hal_c_on at 1:17 AM on November 12, 2010 [1 favorite]

An extrapolation that the brain is somehow getting a weak link into future photon state collapses isn't even particularly farfetched.

It's ... not? Brains aren't photonic, and it's a massive jump from waveform collapse of a single photon to effects on even one neuron. It would be surprising to me were there amplification of that scale in place.
posted by quarantine at 1:17 AM on November 12, 2010 [2 favorites]

Of course they could determine which curtain hid the erotic image. They listened closely, then picked the picture corresponding to which of the computer's stereo speakers had the "bocka-chicka-wow" music coming from it.
posted by JHarris at 1:24 AM on November 12, 2010

Sorry for conflating p with n. In my defense I never took a statistics class. Maybe that's why I'm making a post about psi research! Ha ha!
posted by shii at 1:25 AM on November 12, 2010 [2 favorites]

One more thought, on precognition being tied to fear vs. sex.... much would depend on when the ability was developed. If we evolved it very early in our history, tying it to fear would seem the most likely outcome, but if it was a late development (perhaps part of what made us human?) then tying it into reproduction might have offered more of an advantage than linking it to survival.

Another mostly-unrelated thought that comes to mind -- taken on a broad scale, humanity is an incredibly lucky species. Somehow, again and again, we keep surviving crises that had the possibility of wiping us out. If this does actually turn out to be solid, instead of a bizarre statistical accident, our close scrapes might not just be luck. SF authors have sort of touched on this on occasion (Larry Niven's Ringworld comes to mind), but wouldn't it be strange if it were a real effect?

On preview, quarantine: It's ... not? Brains aren't photonic, and it's a massive jump from waveform collapse of a single photon to effects on even one neuron.

I'm pretty sure we've already shown that at least some of the neuron to neuron signaling has to be happening on a quantum level, but it's been at least ten years since I last read anything on the subject. The assertion of whatever articles I was reading, a decade ago, was that much of the reason that neural simulators weren't giving us the results they expected was because they were using a mechanistic model of neuron interaction. It was claimed there was a lot more going on than just electron transfer.

There's no way I'll be able to find a link for you on that, though. It's been way, way too long. Maybe someone else who read on the subject more recently can chime in.
posted by Malor at 1:27 AM on November 12, 2010

On preview, ...

Heh.
posted by No-sword at 1:33 AM on November 12, 2010 [1 favorite]

@Malor: There's no way I'll be able to find a link for you on that, though.

Damn. Would love to read that.

@shii: In my defense I never took a statistics class

Well, I guess we've established that you won't take one tomorrow, either. :-D
posted by quarantine at 1:36 AM on November 12, 2010 [2 favorites]

Avenger: So, if the study is found to be true, then it will be revealed that we humans have psychic powers, but they are very, very weak (almost imperceptible)

Well, imperceptible may be a matter of perspective. As pointed out in the paper that chortly linked to, a 3% edge is sufficient to destroy the 1.35% house edge in a red-black bet in European roulette.
posted by mhum at 1:40 AM on November 12, 2010 [1 favorite]

phrontist: They used one of these hardware RNGs as well, so nevermind.

Actually, they ran a bunch of different experiments and used somewhat different RNG setups each time. Sometimes they used the native PRNG in REALBasic (which, according to this, was linear-congruential in version 4.5 and additive congruential in 5.0, for what it's worth). Sometimes, they used Marsaglia's multiply-with-carry PRNG. Sometimes, they used the hardware RNG and the Marsaglia generator together.
posted by mhum at 1:46 AM on November 12, 2010

So in the paper, they talked about a substantial correlation between extroversion and having psi powers (pp. 10-11) in this test and in prior tests done by others. I've always wondered if that was due to errors in the testing methodologies for PSI-related experiments. I imagined that those who were extroverted would go, "oh man these are bad results, let's go one more time and see if I do better", and they'd keep going until the results were favorable. If enough extroverted people had this attitude, that would really mess up the results. It doesn't look like they did that based on their methodologies, but it does make me wonder.

On the other hand, it'd also be really funny if psi powers really did exist and the people having the most effect on controlling their results with the mind is the experimenters themselves, not the participants. I.e. they started the experiment at the exact correct moment and clouded the random number generator, and therefore have difficult to reproduce results because their desire for good results made themselves unintended participants.
posted by amuseDetachment at 1:58 AM on November 12, 2010

vidur: "I know how this thread will go."

And I know how it will end:

Well?
posted by The Light Fantastic at 2:45 AM on November 12, 2010

The first step here is to reproduce the results. For a finding of this magnitude, that's it's going to require a lot of work.

Assuming others confirm the results, then comes the tricky part. With a finding of correlation but no hypothesis as to why, figuring out how this mechanism works will be difficult, not the least reason of which being that it somehow bypasses causation. We could debate why it happens, but how on earth could we hope to determine how? Is it only erotic images that produce the result, and if so, how the heck could that be? Why would the universe care if we get our rocks off? It would be rather strange if sexy sexy sex turned out to be fundamental to the laws of causation.

One thing's for sure: the woo-woo crowd will have a field day with this, regardless if it's disproven.
posted by JHarris at 3:30 AM on November 12, 2010

I came here to make one of the exact points those Dutch scientists chortly pointed out made (although I didn't know about their paper before this).

Essentially, the statistical test used has certain properties that make it quite likely that the result is spurious, even though it has a low p-value. Their paper is way more thorough than my comment is going to be and it points out multiple problems... but I think the critical one, which I came here to say, is point three in their paper -- that p-values overstate the evidence against the null in certain circumstances. In particular, in cases where the null is a point mass (as it is here, i.e., is "chance" or 50%), the test is one-sided and the effect size is small, it is likely that you will reject the null hypothesis with a "highly" statistically significant p-value regardless of the data.

** In fact, in these circumstances the probability that you will reject the null approaches 1 as the sample size becomes arbitrarily large (Lindley, 1957). This has been proven (see, e.g., here (pdf) for an explanation).

Why is this? In a nutshell, the intuition driving it is that "chance" is a point hypothesis (i.e., it means precisely 50%)... but no data are ever going to have a mean of precisely 50%, and as the sample size gets larger, those deviations will become "significant" even though the difference from 50% might be quite small. I'm glossing over a bunch of the math, but that's the intuition. This is one of the reasons that some scientists are so suspicious of p-values and advocate Bayesian statistical methods, for which these sorts of problems do not arise.

Take-home message: I'm pretty sure that dodgy stats are the explanation for Bem's results (well, as sure as I can be without analysing the data myself). We're not psychic, or, if we are, this paper hasn't shown it.
posted by forza at 3:49 AM on November 12, 2010 [28 favorites]

I'm pretty sure we've already shown that at least some of the neuron to neuron signaling has to be happening on a quantum level.

That's not my impression; apart from Penrose and Hameroff, I'd have said most people think neurons operate classically, with any quantum effects being marginal noise at most. I haven't got a good, up-to-date link either, though.
posted by Segundus at 4:25 AM on November 12, 2010 [2 favorites]

Results like this have been showing up for quite a while. Again, no one knows why -- it may or may not be "Psi" in a variety of speculated flavors, whatever that is. But these statistical anomalies are real.

http://www.amazon.com/Conscious-Universe-Scientific-Psychic-Phenomena/dp/0062515020
http://www.amazon.com/Entangled-Minds-Extrasensory-Experiences-Quantum/dp/1416516778/
posted by zeek321 at 4:56 AM on November 12, 2010

Say you are a student studying for an exam. Would your recall in the exam room be improved if, in addition to studying before the exam, you went over the material one more time after the exam (analogous to typing the words after the recall test in the study)?

I've done that before, many times, trying to figure out how I did on a tough test or cert exam before the results were returned. I'll be damned if my scores weren't higher than I thought they'd be. I put it down to cynicism bordering on paranoid superstition - if I think I aced it, I flunked, and if I thought I flunked, I aced it.

Ater reading these studies, I still put it down to cynicism bordering on paranoid superstition. Yes, the universe is out to get me.
posted by Slap*Happy at 5:04 AM on November 12, 2010

What? Of course I haven't already looked at the previous comments!
posted by Decani at 4:05 AM on November 12 [+] [!]

Don't try to frighten us with your sorceror's ways, Decani. Your sad devotion that ancient religion has not helped you conjure up the stolen data tapes, or given you clairvoyance enough to find the rebels' hidden fortress.
posted by Comrade_robot at 5:18 AM on November 12, 2010

Some time ago, I kept a dream journal as a method for improving my lucid dreaming. I am a boring person, so whenever dreams are about me (as opposed to the dreams where I am someone else), they're very pedestrian and even a little dull. It was prompted by a horrifying dream in which my cat got a hairball, but was unable to successfully expel it, so she died in front of me having nearly ripped herself open after walking out of my mother's bedroom and staring at me while she kept ... spasming.

No more dreams like that, I decided, because I had spent a few days absolutely stunned from the intensity. I purchased a book on lucid dreaming and began my dippy little dream journals, right in time for my cat to have a hairball and have trouble ejecting it until I whacked her on her little kitty back.

I started reading my dreams, about a week back from current times, just to see if the dream had recurred. And every so often ...

So I have written down that I nearly get in an accident at this particular intersection, on my birthday, because of a white van. I did not think about it too much until, when tooling around on my birthday, a white van blows through the red light and nearly creams me as I was making a turn. I go back and read my journal. Yeah, it's there.

About a year later, I write down my dream that my father calls me and I am sitting on my bed and he tells me he has lost his job. This is, in addition to being a boring dream, a silly dream because my father's job was very stable. And, well, I am a phone pacer. No sitting while talking. About two weeks later, my father calls me to tell me he has lost his job. And I'm sitting on my bed.

Had I noted at some level below immediate awareness that my cat had troubles with hairballs and simply incorporated that into dream imagery? Was I picking up on unvocalized cues of fear from my father? Had I spotted the van out of the corner of my eye and unconsciously decreased my usual wait time to replicate the situation in my dream?

I did not know, and would never know, but I stopped writing down my dreams because I decided that this is how The Crazy gets started.
posted by adipocere at 5:24 AM on November 12, 2010 [7 favorites]

He uses years of personal dream journal entries as his data set.

I keep trying to parse this isn't some sort of logical construct and failing.
posted by JaredSeth at 5:32 AM on November 12, 2010 [1 favorite]

I think the simplest explanation for this is that the placement of the erotic pictures was not truly random. True randomness is difficult to achieve. Even a slight bias for repeated event placement could achieve a 53% success rate.
posted by dances_with_sneetches at 5:43 AM on November 12, 2010

An interdisciplinary approach to certain fundamental issues in the fields of physics and biology: towards a Unified Theory

Closed timelike curves, superluminal signals, and “free will” in universal quantum
mechanics
posted by AElfwine Evenstar at 6:03 AM on November 12, 2010

You're all under arrest for committing a future MURDER!
posted by briank at 6:08 AM on November 12, 2010 [3 favorites]

Daryl Bem is a departmental colleague of James Maas. I took James Maas' psych intro class and he felt like a showman, I would even call him the Ron Popeil of my professors. He's a pop psychologist too, with a best selling book and appearances on Oprah, and the 700 Club. James Maas' parting comment at the end of the class was a vague assurance that ESP existed.

My point is that I suspect that there is a bent of greed for fame in this department, I think that they are smart, but that they lack the normal self deprecation of good scientists. Because of this character judgement, combined with the continued non-existence of perpetual motion machines* I am confident that these results are not trustworthy.

(*'cause, you know, people are continuously making them, they continuously get minor media coverage, and they continuously work out to be frauds, kind of like ESP research.)
posted by nmr8 at 6:15 AM on November 12, 2010 [2 favorites]

Though I'm not religious, I totally believe precognition and Clairvoyance/remote viewing is possible, based on nothing but a vague feeling and general wish. The little I've read about quantum physics sort of backs it up or it least, IMO, puts in the realm of maaaaaaaybe, maybe not, but damn it sure would be nice.
posted by nomadicink at 6:21 AM on November 12, 2010

That's fascinating stuff, adipocere.

What alternative is there for those taking a strictly skeptical view but to accuse adipocere and every single other person who has ever experienced a precognitive dream of being a straight up liar?
posted by motty at 7:16 AM on November 12, 2010

i have to read this study because i've often dreamt of situations in which i find myself sometimes years later. i've often wondered if pre-cognition has more to do with being really good at logistics and trend watching than with having some sort of random relationship with the space-time continuum aka magical powers.
posted by liza at 7:33 AM on November 12, 2010

What alternative is there for those taking a strictly skeptical view but to accuse adipocere and every single other person who has ever experienced a precognitive dream of being a straight up liar?

Plenty. Coincidences, frequency illusions (white vans probably run red lights all the time, and there's certainly no shortage of them on the roads), post hoc selection bias (adipocere's not reporting all those other dreams that didn't come true). All without denying the reporting of either the dream or the event, or the sincerity of the report.
posted by Electric Dragon at 7:45 AM on November 12, 2010 [7 favorites]

All you need to know is that the most wondrous, exciting days of your life are still ahead of you, my friend. Yet you will know tragedy that surpasses anything you have yet experienced. All of these things have yet to be - but they are tumbling down the pipe. This is the way of the world and if you leave your house today please know that your are deeply and truly loved.
posted by Baby_Balrog at 7:51 AM on November 12, 2010 [5 favorites]

I knew you were going to post this, so I compiled the following links:

http://www.youtube.com/watch?v=fn7-JZq0Yxs&t=0m20s
http://angryflower.com/timelo.gif

... but I overslept.
posted by Reverend John at 7:52 AM on November 12, 2010 [1 favorite]

I think the simplest explanation for this is that the placement of the erotic pictures was not truly random. True randomness is difficult to achieve. Even a slight bias for repeated event placement could achieve a 53% success rate.

No, the simplest explanation is that they misanalysed their statistics. That paper posted by chortly is pretty damning of their analysis techniques. The experiment may well have been carried out perfectly, but they might still be wildly overstating any effect:

For instance, Bem’s Experiment 1 tested not just erotic pictures, but also neutral pictures, negative pictures, positive pictures, and pictures that were romantic but non-erotic. Only the erotic pictures showed any evidence for precognition. But now suppose that the data would have turned out differently and instead of the erotic pictures, the positive pictures would have been the only ones to result in performance higher than chance. Or suppose the negative pictures would have resulted in performance lower than chance. The Bem Exploration Method holds that a new and different story would then have been constructed around these other results. This means that Bem’s Experiment 1 was to some extent a fishing expedition, an expedition that should have resulted in a correction of the reported p-value.

posted by Electric Dragon at 7:55 AM on November 12, 2010 [1 favorite]

Exactly, Electric Dragon! That's why I stopped looking at things like that, because I thought, "You know, I could look for patterns like that all the time." Sure, white vans do not normally try to hit me at the exact same intersection (Manchester Road, the road of madness) as was in my dream, on my birthday, as in my dream, but it is still faintly probable.
posted by adipocere at 8:01 AM on November 12, 2010 [1 favorite]

What alternative is there for those taking a strictly skeptical view but to accuse adipocere and every single other person who has ever experienced a precognitive dream of being a straight up liar?

Confirmation bias. Nobody makes note of the thousands of dreams that don't happen to match up with some incident in the future; they only remember the ones that do. Any of us could sit down right now and make up a list of perfectly plausible things that might happen to us someday -- cat getting a hairball, near miss on the highway, someone close to us losing a job, etc -- and then wait for them to happen; odds are at least some of them will, eventually.
posted by ook at 8:11 AM on November 12, 2010 [2 favorites]

We could debate why it happens, but how on earth could we hope to determine how?

It would be even weirder than some sort of black-box psychic power, wouldn't it? I mean, if you flip a coin, the way you know which side landed upwards is to take a look at it. The sense experience is processed into knowledge, and there's no other way to know the past of that coin. So it seems like a non-woo-woo precognition would have to take into account sense experience, too--not just that you knew it would be tails, but that your senses were picking up on it being tails. You'd have to see it first, then know it.

Unless...unless...we are being bombarded by sense experiences from the future all the time, and it's only our stubborn consciousness that filters them out that keeps us safe!
posted by mittens at 8:17 AM on November 12, 2010

Some students appeared to have greater premonitory power than others . . .
Wait, so, this time it really IS Ghostbusters II?
posted by The Bellman at 8:23 AM on November 12, 2010

FATHER! THE SLEEPER HAS AWAKENED!
posted by "Elbows" O'Donoghue at 8:43 AM on November 12, 2010

That was fast:
A Replication of the Procedures from Bem (2010, Study 8) and a Failure to Replicate the Same Results
posted by talos at 8:47 AM on November 12, 2010 [3 favorites]

Does this mean it's okay to have ESP in science fiction again?

As long as it's accompanied by unconditional hairy pre-utopian group sex.

Paging Spider Robinson.....
posted by CynicalKnight at 8:51 AM on November 12, 2010 [2 favorites]

I once won $12 in a lottery based on numbers I woke up with after a dream. It was the first and only lottery ticket I've ever bought. Nice co-incidence, though fairly the dream didn't give me enough numbers to complete the sequence required for a full ticket. Also the numbers were close to my birthdate and it was a ticket bought on my first legal to gamble birthday.

I kinda wish I'd dream of more numbers so I can get more data points as that alone was kinda fun as far as ways to waste a toonie.
posted by Phalene at 9:04 AM on November 12, 2010

parapsychology was a complete waste of time
posted by IvoShandor

So a lot of your surgery was unnecessary?
posted by obiwanwasabi at 9:11 AM on November 12, 2010 [4 favorites]

Step 1: Assign a set of stocks a word each
Step 2: Show a person a list of these words
Step 3: Purchase stocks for the ones they recall successfully
Step 4: Have person type the stock's word if it increases in price

INFINITE MONEY
posted by Jon Mitchell at 9:11 AM on November 12, 2010

Another mostly-unrelated thought that comes to mind -- taken on a broad scale, humanity is an incredibly lucky species. Somehow, again and again, we keep surviving crises that had the possibility of wiping us out. If this does actually turn out to be solid, instead of a bizarre statistical accident, our close scrapes might not just be luck. SF authors have sort of touched on this on occasion (Larry Niven's Ringworld comes to mind), but wouldn't it be strange if it were a real effect?

Humanity could be said to be lucky, seeing as how we're still alive, but I don't think we're particularly lucky. Surviving crises that had the possibility of wiping us out is otherwise known as the game of existence -- the fact that we're still at the table is no more lucky or special than the fact that the badgers, ants, and ravens also haven't run out of chips.

Actually, considering that they and their immediate ancestors have been playing for millions of years longer than we have, maybe we should ask them for a little luck...
posted by vorfeed at 9:21 AM on November 12, 2010

You guys, I totally know what all the comments on this thread are. I totally postdicted it.
posted by Galaxor Nebulon at 9:25 AM on November 12, 2010

What alternative is there for those taking a strictly skeptical view but to accuse adipocere and every single other person who has ever experienced a precognitive dream of being a straight up liar?

I don't know how you'd calculate the exact odds, but I imagine that with around 7 billion people having some number of dreams every night, runs of prophetic or seemingly prophetic dreams are inevitable.
posted by Lentrohamsanin at 9:35 AM on November 12, 2010

Sexual ESP is a real thing, man. Like, I totally knew I was going to end up in bed with that girl! Even before I gave her the money.
posted by Mister_A at 10:03 AM on November 12, 2010 [1 favorite]

...a nice paper by some Dutch psychologists critiquing the Bem paper's statistics from a Bayesian point of view...

Oh, it's on! Bayesian Extropianists vs. Psi-positive New-Agers in a hippie-fight! This shit just got real. Gonna grab me some popcorn...
posted by Slap*Happy at 10:06 AM on November 12, 2010 [2 favorites]

I am concerned that he waited until he had a significant effect. As your N increases, small changes in your dataset may become significant through sheer random chance. Continuing to gather data until such a time occurs, means that many, many tests were done each with a probability of say 5% of being noise. Eventually, you'll hit paydirt with one of these tests.
posted by scrutiny at 10:13 AM on November 12, 2010 [5 favorites]

This study explains my uncanny ability to find pornography on the internet.
posted by gamera at 11:19 AM on November 12, 2010 [2 favorites]

Ok since everybody else is airing their own crazy theories, here's mine; slight psychokinetic effects are what allow free will. Each human being is, very slightly but persistently, influencing the (otherwise entirely causal) machinery of their own brain. Where the PK energy and the underlying volition come from is the real question though. The brain, complicated as it is, is just there to (aside from making everything run) be a target of influence.
posted by newdaddy at 12:50 PM on November 12, 2010 [1 favorite]

Pfft. I'll believe it when James Randi says he believes it.

(And I have a feeling he's not gonna.)
posted by erniepan at 2:43 PM on November 12, 2010 [1 favorite]

scose> More than 50% of natural scientists believe in precognition???????

Back in 1979, apparently.
posted by UrineSoakedRube at 4:00 PM on November 12, 2010

Bem and Honorton also carefully evaluated the possibility of a “file drawer effect.” This is the name used to illustrate a tendency to publish studies with positive results, while studies with negative results don’t get reported. Their analysis shows that, to explain the obtained results using a file drawer effect, would require around 50 unpublished negative studies for each published positive one. Given the large amount of cost and effort required to run a serious psi experiment, this really doesn’t seem plausible. (It should be noted that many results in other areas of science would have to be thrown out, if one were to adopt significance criteria so strict as to rule out the ganzfeld data due to the possibility of an extreme file drawer effect like this.)

I would have drawn a different conclusion from that parenthetical than Goertzel did.
posted by UrineSoakedRube at 4:02 PM on November 12, 2010

Bem and Honorton also carefully evaluated the possibility of a “file drawer effect.” This is the name used to illustrate a tendency to publish studies with positive results, while studies with negative results don’t get reported. Their analysis shows that, to explain the obtained results using a file drawer effect, would require around 50 unpublished negative studies for each published positive one. Given the large amount of cost and effort required to run a serious psi experiment, this really doesn’t seem plausible. (It should be noted that many results in other areas of science would have to be thrown out, if one were to adopt significance criteria so strict as to rule out the ganzfeld data due to the possibility of an extreme file drawer effect like this.)

I would have drawn a different conclusion from that parenthetical than Goertzel did.

There's a pretty good article in the current Atlantic about the other side of this (among other topics). A huge share of published medical science reaches false conclusions, in part because of the publication bias (or "file drawer effect").
posted by grobstein at 4:13 PM on November 12, 2010 [1 favorite]

In PhD school I once got handed a 50-page packet on research writing techniques. The author's tone was extremely cocky, as if all scientists are basically illiterate cavemen. I got curious and looked up his only research publication: a study published in a minor psychology journal that asserted ESP exists based on shining red lights near (but not at) undergraduate students with their eyes closed. Sample size: 12.

Argh.
posted by miyabo at 4:23 PM on November 12, 2010

Anyone else ever journal his or her failed predictions? I did for a while, to see what kind of confirmation bias I witnessed. My "strong feelings" that my awaited check would be waiting in my mailbox, or that there would be a big earthquake soon, or that a particular person would comment on my blog post, &c.

Guess what? Not only was my batting average abysmal, but I had forgotten I made most of the predictions at all. Try this exercise before you decide you're psychic.
posted by quarantine at 5:47 PM on November 12, 2010 [5 favorites]

Yes but WHERE IS CAPTAIN HINDSIGHT?
posted by yoHighness at 8:21 PM on November 12, 2010 [1 favorite]

Whoa. Total déjà vu.
posted by susanbeeswax at 10:46 PM on November 12, 2010

Related. GotPsi, to test one's psychic powers online.
posted by nickyskye at 11:22 PM on November 12, 2010

The Bem Replication Registry
posted by edd at 4:32 AM on November 17, 2010

Whoa, it's totally the same guy!
posted by miyabo at 6:58 AM on November 17, 2010

And Wiseman points out a problem with one of the tests.
posted by edd at 12:24 AM on November 19, 2010

"Daryl Bem is a departmental colleague of James Maas. I took James Maas' psych intro class and he felt like a showman, I would even call him the Ron Popeil of my professors. He's a pop psychologist too, with a best selling book and appearances on Oprah, and the 700 Club. James Maas' parting comment at the end of the class was a vague assurance that ESP existed.

My point is that I suspect that there is a bent of greed for fame in this department, I think that they are smart, but that they lack the normal self deprecation of good scientists. Because of this character judgement, combined with the continued non-existence of perpetual motion machines* I am confident that these results are not trustworthy."

Prof. Bem was my undergraduate advisor. He could not be further from Maas if he tried. He's a very nice, soft-spoken guy. And definitely not a "woo-woo" cuckoo.

I knew about this research while I was still at Cornell, and it always made my mind boggle. But I checked it out and it seemed legit (as well as dispassionate). The experimental design may have flaws (most do), but I always had the sense that the experiment was just as much to disprove as prove the existence of any parapsychological phenomena. It was more about performing a really rigorous study than validating some assumption.

I wish I could say more, but I never participated in or assisted with the study (on any intentional level... I'm probably one of the n somewhere in there) as it was not my area of expertise. I spent most of my time studying lucid dreaming (OMG CORNELL IS A LOONY FACTORY!!!).
posted by Eideteker at 6:32 AM on November 20, 2010