quote:In a test that we wouldn't have believed had it not been documented, 100 Cornell students were shown 48 common nouns and given three seconds to observe and visualize each word. Then they were asked to type out as many words as they could remember. After that, a computer re-displayed half of those words, which the students then retyped.
You don't have to be psychic to know where we're going with this: It turns out that the students more likely recalled the words that they were later asked to retype.
I'd like to know how many iterations they did of this, and whether the results aren't just an odd sampling fluke.
Posts: 12266 | Registered: Jul 2005
| IP: Logged |
posted
Eh. The experiment as described sounds reasonable assuming no obvious mistakes like not selecting the words at random were made, but the interpretation of quantum mechanics given as explanation sets off my dingbat alarms. I want the study replicated, and I also want to know the size of the effect, remembering that in twenty experiments you're going to get one 95%-confidence refutation of the null hypothesis just by random chance.
Posts: 10645 | Registered: Jul 2004
| IP: Logged |
posted
This isn't the first time people have brought up evidence that they claimed pointed to precognition.
The other experiments, when they attempted replication under better controls, turned out to be showing nothing.
Posts: 15421 | Registered: Aug 2005
| IP: Logged |
quote:In a test that we wouldn't have believed had it not been documented, 100 Cornell students were shown 48 common nouns and given three seconds to observe and visualize each word. Then they were asked to type out as many words as they could remember. After that, a computer re-displayed half of those words, which the students then retyped.
You don't have to be psychic to know where we're going with this: It turns out that the students more likely recalled the words that they were later asked to retype.
I'd like to know how many iterations they did of this, and whether the results aren't just an odd sampling fluke.
Couldn't this just as easily be proof that the computers at Cornell are psychic in that they knew which words the students would recall?
Posts: 14316 | Registered: Jul 2005
| IP: Logged |
posted
Wow, this paper is hilarious. For instance, after detecting a "significant" result for people being "precognitive" of erotic images, he discusses various alternative explanations . . . but fails to list "a significant result happened by chance" (which most people would assert, at around a percent of the time, would be much more likely than a psychic phenomenon).
Then there are outright incorrect statements such as
quote:In particular, if the human participant is replaced by the same PRNG or RNG that selects the left/right target positions, this maximizes the possibility that any non-random patterns in the sequence of left/right target positions will be mirrored by similar patterns in the left/right responses of the virtual participant (the RNG itself), thereby producing an artifactual psi-like result.
Ignoring all the secondary analyses and focusing on the basics of the series of experiments, if this series of experiments were run properly and no experiments were omitted, they would constitute some evidence for a small psi effect.
I am, unsurprisingly, doubtful that the necessary conditions are met.
Posts: 15770 | Registered: Dec 2001
| IP: Logged |
posted
article's opening statement is wrong too. think you have precognitive powers and can divine the future through supernatural means? you should probably think about upping your meds.
Posts: 15421 | Registered: Aug 2005
| IP: Logged |
posted
KoM: regarding effect size, they're talking on the order of one to three percent of trials 'matching' (or whatever the effect signifier is) more often than chance. I find it interesting that it is the word studying experiment from the paper that is mentioned in the article, since it seems to be one of the weakest, statistically, and is harder to describe. Some of the other trials are more straightforward.
There are also secondary analyses where they assert people who seek out stimulus have an even greater effect in certain trials, but there's so much more capability for fudging results like that I mostly skipped over them.
Posts: 15770 | Registered: Dec 2001
| IP: Logged |
quote:KoM: regarding effect size, they're talking on the order of one to three percent of trials 'matching' (or whatever the effect signifier is) more often than chance.
Ok, but with how many trials? Otherwise you can't work out a significance.
Posts: 10645 | Registered: Jul 2004
| IP: Logged |
posted
Most of their experiments involve a hundred to two hundred students (from Cornell) performing some task a moderately large number of times (36, 50, et cetera). As I allude to earlier, significance for the main effect is often at around the .01 level, but sometimes more and sometimes less. In all but one experiment, I think, better than .05.
edit: you know, the paper is linked in the article
Posts: 15770 | Registered: Dec 2001
| IP: Logged |
posted
Well yes, but I do have actual work to do, you know. If I want to read borderline papers, I've got a perfectly good one right here purporting to show the compositeness of quarks.
Posts: 10645 | Registered: Jul 2004
| IP: Logged |