SCIENTIFIC JOURNAL TO PUBLISH ARTICLE ON THEORETICAL PRECOGNITIVE ABILITIES IN HUMANS.

i always told you guys i was psychic.

Well, all I can say is that Journal of Personality and Social Psychology is a very badass journal, and Daryl Bem is an experient respected psychologist. Of course that doesn't eliminate the possibility that he would falsify the data, but the scientific community is not likely to make that kind of judgement towards his findings.

I always believed in the possibility of premonition/precognition, though I thought it was some sort of gift that some people had and some didn't, which would be a big problem in terms of scientific research, since scientific data must be replicable. But Bem picked random subjects, and he got a significant (even though very small) evidence of premonition.

Let's see if other researches will try to replicate that. Because if they do, and they get positive results... welp. We're in big trouble. Big, big trouble.

I knew you were going to post this.

it's like deja vu all over again.

post=thedjt
You can't explain that crap like when you think of someone and they phone or when your dream ends around your alarm clock going off.

Those are easily explained.

Deja vu is just a memory anomaly, but I don't think it's generally thought of to be a precognitive thing.

The effects he recorded were small but statistically significant. In another test, for instance, volunteers were told that an erotic image was going to appear on a computer screen in one of two positions, and asked to guess in advance which position that would be. The image's eventual position was selected at random, but volunteers guessed correctly 53.1 per cent of the time.

That may sound unimpressive â€" truly random guesses would have been right 50 per cent of the time, after all.

My...My god. Who could have possibly predicted this! And it would only ever be fifty percent if you did around a million trial experiments. Call me when this whole thing is replicated.

post=calunio
MMmmm... all these findings may be just a sign of misused statistic analyses after all.

Misused statistic analyses is indeed the culprit. http://www.ruudwetzels.com/articles/Wagenmakersetal_subm.pdf

Bem's experiment has two errors that I'm able to understand. There's a third listed, but I could not understand what it actually meant.

The first error Bem made was to make multiple attempts and only report those who yielded significant result. Let's for simplicity's sake say that the probability of something to happen is 5% if left entirely to chance. Let's assume you do five experiments, each with that probability. Now, the chance that at least one of the experiments will yield that 5% result is 1-0,95^5 which is about 23%. If one of your experiment does happen to produce that 5% result, you should report a 23% result, not a 5% result. Bem did an error similar to reporting a 5% result in my example.

The second error was to not weight his evidence for precognition against the evidence <i>against</i> precognition. We have no evidence that proves beyond any doubt that precognition doesn't exist and I do not think it's even possible to have such evidence. We do however have evidence suggesting that precognition is very unlikely. One example given is casinos being profitable. It's not hard to imagine how people with precognitive abilities could ruin a casino. Casinos being profitable therefore speaks against people with precognitive abilities existing. Obviously, if you find evidence for precognition, it has to be stronger than the evidence against precognition before you can draw any conclusion.

Now for something fun. Bem followed the rules that guide academic publishing. There may at this point be a lot of published findings by psychologists that came to be due to flawed statistic analyses.

All I need now is a blue and yellow striped shirt Ooooo Yeeeh

Welp, I'm not sure the case of the middle variable (like in the study of nightmares) applies here.

Also, I don't think Bem's study should even be called parapsychology. He's dealing with concrete psychological variables... of course the effect is interpreted in terms of something that can't be explained without the use of parapsychological terms, but still... it sounds like valid psychological science. What is to question is the analysis of data, and maybe methodological issues.

The thing about the statistics is... if you have 10 subjects performing with a 0.53 average accuracy, it will probably not be very significant. If you have 100 subjects, it will be more significant. If you have 10000 subjects, it will definitely be significant, even with the same average score. So, yeah, 53% accuracy is not so great, but what matters for the scientific conclusion is the p.

author=calunio
Welp, I'm not sure the case of the middle variable (like in the study of nightmares) applies here.

It applies to what you posted, but not to Bem's experiment. My example was provided to explain why per review has to catch flawed experimental design, not to explain anything about Bem's experiment.

Also, I don't think Bem's study should even be called parapsychology. He's dealing with concrete psychological variables... of course the effect is interpreted in terms of something that can't be explained without the use of parapsychological terms, but still... it sounds like valid psychological science. What is to question is the analysis of data, and maybe methodological issues.

It's not so that some fields of science are valid while others are not. You can make perfectly valid studies of parapsychology. Such studies have also been made. However, those studies have not showed any significant result. So, just because something is valid doesn't mean it can't be parapsychology. As a general rule, if it concerns our minds and you're trying to observe something we can't explain as opposed to explain something we observed, it's parapsychology. Precognition fits nicely.

The thing about the statistics is... if you have 10 subjects performing with a 0.53 average accuracy, it will probably not be very significant. If you have 100 subjects, it will be more significant. If you have 10000 subjects, it will definitely be significant, even with the same average score. So, yeah, 53% accuracy is not so great, but what matters for the scientific conclusion is the p.

Bem did multiple experiments and only one yielded that 53% accuracy. If you do multiple experiments, the chance that at least one of them will yield those 53% is obviously higher than if you only do one experiment. You need to correct the p-value accordingly, something Bem did not do. Also, more than the p-value matters. There are evidence against precognition, such as the example I mentioned of casinos being profitable. The evidence for precognition has to overcome the evidence against.