Merged Bem's latest experiments / Researcher says we can see the future

[squig]

Basically, QM is about elementary particle (and molecules & atoms) being modellable as a wave , and their property described through a wave function. An observation in the physical term, is an interraction with the wave which provocate a collapse and enable one to read eigen value of the particle observed, for example energy level.

I am really summarizing here, but all woo pretending that the observer effect enable us to dostuff as human, are dead wrong, as QM has a specific definition of observing (interraction), and when we "observe" as human we don't interract with what we observe in any meaningful way.

I recommend to look on wiki on QM to get the basics.

Thanks, the wiki seems a bit too complicated for me to follow through but if I understand correctly, the gist is that every form of observation in the qm-sense is an interaction with the particle observed unlike human "observation", am I correct?

Okay. Parapsychologists generally seem to consider that they has been taken into account. When asked this question, they refer to a justification offered by Jessica Utts that it theoretically should not materially alter the results. The simple explanation is that though you may prefer to choose the right-facing target, and may do so on every trial, you should still only be right about half the time, as a random selection of targets will ensure that a right-facing target will only be selected about half the time. Yes, the variation expected from random sampling means that this 50-50 proportion is only a general tendency and that you will see usually see something a bit different from 50-50 and occasionally a lot different from 50-50. But this variation, and the probability of obtaining random targets which are a lot different from 50-50 and coincide with the choice of target by the subject, can presumably be described by the same distribution obtained by random sampling. Parapsychologists seem to consider this sufficient to mostly fail to take your concerns into account, and to occasionally attempt to see if they can discover this effect on post hoc analysis (as Bem attempts in this paper).

Now, I'm not sure that even theoretically this would work. When I've tried to play with this idea, it looks like this alters the variation so that the variance would no longer be accurately described by, for example, the binomial distribution, meaning that statements about expected probabilities based on a binomial distribution would be inaccurate. I haven't fully explored this though, so I may be wrong. And it is of little importance compared to the much bigger problem that it turns out in practice that the distribution of 'randomly selected' targets seems to be markedly different from the expected distribution. The few times we have been given the data on the distribution of targets, that distribution has been markedly different from chance. So we already suspect that statistical tests based on random sampling won't give us valid answers, regardless of whether they could in theory.

Linda

Bem has also used parametric tests, so binominal distribution is not an issue afaik. Even if the data-set was not binominal, the resulting skewness can be easily tested and corrected for. It should not alter the results much if at all, since Bem used a pretty large number of observations. (at least the studies Ive read up till now)

 

[Orphia Nay]

I just found this thread by doing a tag search for "Daryl Bem". I'd just come across this article:

http://www.psychologytoday.com/blog...finally-discovered-evidence-psychic-phenomena

In one of the studies, college students were given a list of words and after reading the list, were given a surprise recall test to see how many words they remembered. Next, a computer randomly selected some of the words on the list as practice words and the participants were asked to retype them several times. The results of the study showed that the students were better at recalling the words on the surprise recall test that they were later given, at random, to practice. According to Bem, practicing the words after the test somehow allowed the participants to "reach back in time to facilitate recall."

Is it just me, or is this totally underwhelming? This in no way seems paranormal to me. It's just aiding recall by giving reminders. Where's the precognition?

In another study, Bem examined whether the well-known priming effect could also be reversed. In a typical priming study, people are shown a photo and they have to quickly indicate if the photo represents a negative or positive image. If the photo is of a cuddly kitten, you press the "positive" button and if the photo is of maggots on rotting meat, you press the "negative" button. A wealth of research has examined how subliminal priming can speed up your ability to categorize these photos. Subliminal priming occurs when a word is flashed on the computer screen so quickly that your conscious brain doesn't recognize what you saw, but your nonconscious brain does. So you just see a flash, and if I asked you to tell me what you saw, you wouldn't be able to. But deep down, your nonconscious brain saw the word and processed it. In priming studies, we consistently find that people who are primed with a word consistent with the valence of the photo will categorize it quicker. So if I quickly flash the word "happy" before the kitten picture, you will click the "positive" button even quicker, but if I instead flash the word "ugly" before it, you will take longer to respond. This is because priming you with the word "happy" gets your mind ready to see happy things.

In Bem's retroactive priming study, he simply reversed the time sequence on this effect by flashing the primed word after the person categorized the photo. So I show you the kitten picture, you pick whether it is positive or negative, and then I randomly choose to prime you with a good or bad word. The results showed that people were quicker at categorizing photos when it was followed by a consistent prime. So not only will you categorize the kitten quicker when it is preceded by a good word, you will also categorize it quicker when it is followed by a good word. It was as if, while participants were categorizing the photo, their brain knew what word was coming next and this facilitated their decision.

Again, why is this supposed to be proof of psychic or psi powers?? It's just suggestion, or prompting. There's no prediction of future events.

Is there any better "evidence" than this in the actual paper?

 

[blobru]

In the first study, test subjects seem better at recalling words they see twice, once before and once after the test, than at recalling words they only see once, before the test; in the next, subjects seem to categorize images as emotionally "positive" or "negative" quicker if they are 'primed', after categorizing the image, with a word that has the same positive or negative affect than with a word that has the opposite affect. Bem's hypothesis is that the differences in performance are being caused by events after measurement of performance! That would be a startling finding. If the test results hold up to scrutiny and can be replicated, it would be the best evidence for something like psi [precog] I'm aware of (a large and tight enough study where deviations, though weak, are statistically significant). However, barring precogs, how it fares under peer review remains to be seen.

 

[Orphia Nay]

Nope, I'm still totally underwhelmed. Giving someone a reminder or a hint about something they've seen makes it easier for them to recall or categorise something they've seen for a moment. What on earth makes this paranormal??? The incredulity people are displaying is astounding.

 

[blobru]

Except the reminder or hint is given after they recall or categorise (...cue spooky music...) Iow, in one experiment, you're given a list to study that includes "dog" and "cat". Then you're tested on how many words you can recall from the list. After this test, you're asked to write out certain words from the list. Bem's data seem to show that those (actually a subset of subjects, self-identified 'extraverts') asked to write out "dog" after the test were slightly better at recalling "dog" on the test; those asked to write out "cat" were slightly better at recalling "cat"; etc.

 

[Orphia Nay]

Except the reminder or hint is given after they recall or categorise (...cue spooky music...) Iow, in one experiment, you're given a list to study that includes "dog" and "cat". Then you're tested on how many words you can recall from the list. After this test, you're asked to write out certain words from the list. Bem's data seem to show that those (actually a subset of subjects, self-identified 'extraverts') asked to write out "dog" after the test were slightly better at recalling "dog" on the test; those asked to write out "cat" were slightly better at recalling "cat"; etc.

Ah, thanks, that helped explain the test better.

I just had a look at the paper, and there were only 100 participants in the test. I don't think that's a big enough sample size for the results to prove anything paranormal.

 

[Ersby]

I've asked Bem for the data.

~~ Paul

Have you heard anything? I ask because I looked at experiment 1, and for the first forty trials there are three potential types of targets (erotic, negative and neutral - each one appears 12 times in the 36 sessions) so wouldn't that make the expected hit rate on each type 33.3%?

 

[squig]

Ah, thanks, that helped explain the test better.

I just had a look at the paper, and there were only 100 participants in the test. I don't think that's a big enough sample size for the results to prove anything paranormal.

sample size is taken into account by the test. With a higher sample size, the same effect would be "more significant"

edit: actually only 40 ppn are used for the control samples, strange...

Have you heard anything? I ask because I looked at experiment 1, and for the first forty trials there are three potential types of targets (erotic, negative and neutral - each one appears 12 times in the 36 sessions) so wouldn't that make the expected hit rate on each type 33.3%?

They choose between 2 doors, one empty one wiith the image, so 50-50 right?

 

[Aepervius]

Thanks, the wiki seems a bit too complicated for me to follow through but if I understand correctly, the gist is that every form of observation in the qm-sense is an interaction with the particle observed unlike human "observation", am I correct?

There is more to QM (like quantization, uncertainty principle and de-coherence to pass a few more keyword), but yes, this is the gist. We human when we "observe" are actually mostly passively receiving info (sound wave/tympanum, light wave/retina, substance/touch etc...) while not sending any in a meaningful way (*).

(*) we are always sending light in the infrared spectrum and reflecting light in the visible spectrum, but unless what you "observe" (non-physical meaning) can be affected by those thru direct social interaction (read . they see you and react to your presence), there is no influence.

 

[fls]

Bem has also used parametric tests, so binominal distribution is not an issue afaik. Even if the data-set was not binominal, the resulting skewness can be easily tested and corrected for. It should not alter the results much if at all, since Bem used a pretty large number of observations. (at least the studies Ive read up till now)

The parametric tests were performed on data which still depended upon generating targets randomly, so it doesn't really help (the clue is that he still attempted some analyses to 'fix' this problem on those tests). The number of observations (over which hit rates are generated) ranges from 1 to dozens or more. These are the conditions under which it has the most potential to cause problems.

Linda

 

[fls]

Ah, thanks, that helped explain the test better.

I just had a look at the paper, and there were only 100 participants in the test. I don't think that's a big enough sample size for the results to prove anything paranormal.

Even smallish sample sizes can be sufficient to establish an effect with an appropriate research design. The much bigger problem is that the experimental design is not set up to test whether or not there is an effect from psi, but rather whether or not heterogeneous data can be grouped and regrouped in ways that lead to between-group differences. Then if any between-group differences are demonstrated, a failure to be sufficiently clever to account for those differences post hoc is meant to represent 'psi'.

Linda

 

[squig]

The parametric tests were performed on data which still depended upon generating targets randomly, so it doesn't really help (the clue is that he still attempted some analyses to 'fix' this problem on those tests). The number of observations (over which hit rates are generated) ranges from 1 to dozens or more. These are the conditions under which it has the most potential to cause problems.

Linda

I wasn't talking about a lack of randomness though
Assuming the randomness used was indeed random, a preference for the right or left within participants doesn't change that.

Im not sure what you mean with the second part. (maybe we are talking about a different experiment? I replied to a remark about experiment 1)

 

[Beth]

Even smallish sample sizes can be sufficient to establish an effect with an appropriate research design. The much bigger problem is that the experimental design is not set up to test whether or not there is an effect from psi, but rather whether or not heterogeneous data can be grouped and regrouped in ways that lead to between-group differences. Then if any between-group differences are demonstrated, a failure to be sufficiently clever to account for those differences post hoc is meant to represent 'psi'.

Linda

I'm not sure what you mean by this? In what ways was the data grouped and regrouped?

 

[Ersby]

They choose between 2 doors, one empty one wiith the image, so 50-50 right?

Ah, yes. I was thinking they were guessing the kind of picture, but you're right. It was the position they were guessing. Stupid mistake. Thanks for putting me right.

 

[Paul C. Anagnostopoulos]

I've not heard back from Bem yet.

~~ Paul

 

[fls]

I wasn't talking about a lack of randomness though
Assuming the randomness used was indeed random, a preference for the right or left within participants doesn't change that.

I thought Ersby did a fairly good job of illustrating this in post 23 where the random choice of target happened to select mostly right-facing pictures.

Bem also discusses this in regards to the first experiment starting on page 11 (under The Role Of Random Number Generators), and the second experiment on page 17 (under Results and Discussion).

Im not sure what you mean with the second part. (maybe we are talking about a different experiment? I replied to a remark about experiment 1)

You stated that it shouldn't alter the results much because the number of observations was large. I was pointing out that the number of observations was not large (in the paper under discussion) and was sometimes as low as one (considering parapsychology experiments in general). Ersby was talking about the second experiment in the post I was commenting on (although this issue isn't confined to that experiment).

Linda

 

[squig]

Ah, yes. I was thinking they were guessing the kind of picture, but you're right. It was the position they were guessing. Stupid mistake. Thanks for putting me right.

I think the entire paper has horrible to read method sections and Bem makes the descriptions a lot more complicated than they need to be. Im sure that I misinterpreted at least 5 other things myself

I thought Ersby did a fairly good job of illustrating this in post 23 where the random choice of target happened to select mostly right-facing pictures.

Bem also discusses this in regards to the first experiment starting on page 11 (under The Role Of Random Number Generators), and the second experiment on page 17 (under Results and Discussion).



You stated that it shouldn't alter the results much because the number of observations was large. I was pointing out that the number of observations was not large (in the paper under discussion) and was sometimes as low as one (considering parapsychology experiments in general). Ersby was talking about the second experiment in the post I was commenting on (although this issue isn't confined to that experiment).

Linda

But the entire point of random is that it does not coincide. If the random generator is truly random, there is no need to worry about that. Yes maybe for one participant 60% coincides and for the other 40% but they average each other out because it is random.

About the sample sizes, with cases over n=30 (per cell), normal distributions is an assumption that is not strict (the tests are robust against violation of this assumtion). In all 9 experiments, the number of total observations is much larger than 30. And even so, he has used non-parametric tests that are specifically designed for these cases.

--

A small thing I noted b.t.w. is that in experiment 1 it is claimed that positive images are not enough to lead to "psi" but later experiments (the backwards priming ones) show an effect of positive primes

 

[Paul C. Anagnostopoulos]

A small thing I noted b.t.w. is that in experiment 1 it is claimed that positive images are not enough to lead to "psi" but later experiments (the backwards priming ones) show an effect of positive primes.

Also, why do the erotic images work in Experiment 1, but not the negative ones? How were the subjects primed for this distinction?

~~ Paul

 

[Elaedith]

Haven't really followed Bem's research, but these studies don't sound hard to replicate. Surely the obvious question is how replicable the effects are, and are all failures to replicate published?

I feel tempted sometimes to try to replicate this type of study since I conduct computer-based behavioural experiments and have access to everything needed. The problem is that my colleagues would think I've gone bonkers.

 

[Paul C. Anagnostopoulos]

I feel tempted sometimes to try to replicate this type of study since I conduct computer-based behavioural experiments and have access to everything needed. The problem is that my colleagues would think I've gone bonkers.

As them what the hell they think is going on. Tell them you're so fascinated to learn that you're willing to try some replication.

~~ Paul