Psi and brain activity?

[derem]

Been going through old threads for awhile, so I’m not sure whose still active here haha

So I’ve gone through a lot of psi phases and I’d say I’m more skeptical than not, but there’s one thing that does kind of seem somewhat convincing

In terms of studies, I have 3 that concern me here

This study didn’t neccessarily prove psi at all, since the results were at chance. But how is it that these researchers seem to consistently get results that kind of defy logic/causation?

Essentially the idea here is that they can detect a certain pattern in a brain state that occurs when a subject is shown a correct target vs a wrong one, and these researchers have done this experiment 4 times now.

I can’t post articles, but if you look up the bolder it’ll show up as the first or second option on the browser, I have the abstract tho (for the first one it’s on semantic scholar, the second one is on ncbi)

Differential event-related potentials to targets and decoys in a guessing task

Event-related brain potentials (ERPs) were recorded from 20 sub- jects performing a computerized, forced-choice guessing task. On each of 40 trials, ERPs were elicited by digitized images of 4 playing cards, sequentially presented on a video monitor for 150 ms. After the last card was presented, subjects guessed which of the 4 cards would be the target for that trial. Fol- lowing the subject’s guess, the computer randomly selected one of the 4 cards to be the target and presented this as feedback; the remaining 3 cards served as nontarget decoys for the trial. We found that a negative Slow Wave measured at 150–500 ms post-stimulus had greater amplitude when elicited by targets than when elicited by nontarget decoys (p .05). This result indicates an ap- parent communications anomaly because no viable conventional explanation of the ERP differential could be identified. It is the fourth study in our labora- tory employing essentially the same design to yield this or a similar ERP ef- fect.
Keywords:

I realize it’s not neccessarily proof of psi at all, but how is this possible? I can think of 2 ways.

1. Some sort of measurement error,
2. Fraud or file drawer error

All this is basically saying is that when we see something that’s a correct option vs an incorrect option regardless of whether the subject knows whether or not the option is correct or incorrect, and controlled for differences in the target.

The first test was done with Malcolm Bessent, whose a known “psychic.” The second and third test were done with non psi selected participants

Persinger study on ingo swann

In the present study, the artist Ingo Swann, who helped develop the process of remote viewing (awareness of distant objects or places without employing normal senses), was exposed during a single setting of 30 min. to specific patterns of circumcerebral magnetic fields that significantly altered his subjective experiences. Several times during subsequent days, he was asked to sit in a quiet chamber and to sketch and to describe verbally distant stimuli (pictures or places) beyond his normal senses. The proportions of unusual 7-Hz spike and slow wave activity over the occipital lobes per trial were moderately correlated (rho=.50) with the ratings of accuracy between these distal, hidden stimuli and his responses. A neuropsychological assessment and Magnetic Resonance Imaging indicated a different structural and functional organization within the parieto-occipital region of the subject's right hemisphere from organizations typically noted. The results suggest that this type of paranormal phenomenon, often dismissed as methodological artifact or accepted as proofs of spiritual existence, is correlated with neurophysiological processes and physical events. Remote viewing may be enhanced by complex experimentally generated magnetic fields designed to interact with the neuromagnetic "binding factor" of consciousness.

Persinger on remote viewing. He studied a guy called ingo swann, a known remote viewer, and found a correlation in terms of his accuracy vs high alpha wave spikes and the same pattern in the experiment above. I’m not sure how good his overall accuracy was. I’m not sure how consistent this result is overall however.

So I guess I have 2 questions from this

1. Are there any problems with the first paper in terms of data analysis, or selective reporting of the brain data, because that’s the only real problem I can think of there. But how is the idea that there is a predictable brain activity when we see a correct target vs an incorrect target even possible?

2. What do we make of the idea that remote viewing (and psi related things in general) accuracy seem to have a correlation with brain activity? Regardless of the total hit rate or not, the fact that most experiments seem to find a correlation of high alpha activity and psi accuracy is really weird.

Now I’m not sure how constant this correlation is, and I find that finding studies that don’t replicate or show no effect are harder to find and probably don’t get published as much, so that could explain the second effect. Furthermore, if it was real you would think there would be a study right now where they attempt to discern the accuracy just by using the brain activity to test it but I don’t think this has ever been done (like basically if there’s a trend in data, researchers are generally not gonna publish a paper showing an opposite trend, specifically parapsychology researchers)

I’m stumped for the first one though. Not neccessarily that it proves or disproves psi, but just the idea that you can tell if someone sees a correct or incorrect target by trends in brain activity, regardless of prior knowledge. I haven’t seen anything responding to or against this paper either so I’m pretty confused here.

Again, I don’t think there’s any evidence of psi ability, but more so that it just doesn’t make sense how something like this is possible outside of fraud

Edit: found something a bit interesting

Experiment one found slow waves of activity from the 400-500ms mark after exposer

Experiment Two, I couldn’t find a lot of info on, but while they found slow wave differences it was in one hemisphere rather than both like in experiment one

Experiment three found it from the 150-400ms mark after exposure, the 400-500 mark approached significance but didn’t reach it. They said this was their hypothesis that they’d find trends in both spots, and they added the first one after seeing a trend here and revisiting their earlier data, but somehow I doubt that lol

Experiment four found it from the 150-500ms mark after exposure.

So while slow waves were constant, from one to two it was only found in one hemisphere, and from 2 to three they changed the timing, and from 3 to 4 they decided to make the time the 2 timing they contested from 3 to 4 combined

In the abstract they said the hypothesized this, (for the three one) but from a book excerpt from the author
I wonder how much of this was post hoc, and if there were possible failed experiments before this that they didn’t produce. 1 and 2 occurred in the same year (1992), although it was the same subject. 3 and 4 on the other hand, occurred in 1998 and 2002.

Persinger also lowkey doesn’t strike me as someone super reliable, considering the god helmet replication thing where he got salty at a replication attempt by another team that he apparently approved of because it failed

Also, there was a 2008 study on a similar task by an fmri. I don’t think an fmri measures the same things, but they found nothing, and the study was probably better

 

[Elaedith]

Haven't had time to read those articles in detail, but you might want to look at this meta-analysis of the anticipation effect which considers some possible explanations in the discussion.

 

[derem]

Haven't had time to read those articles in detail, but you might want to look at this, the anticipation effect which considers some possible explanations in the discussion.

Wait, I’m confused

It basically says that there’s a larger effect in higher quality studies, and that there’s a clear effect?

In the discussion it seems to list a bunch of possible explanations and then say why they are unlikely, and the only ones that seemed somewhat plausible were fraud and statistical dishonesty (the second of which tbf i heard from Louie savva is rampant), but then it says even with that it’s not fully likely

And in the conclusion, it basically says they don’t know how this anticipatory effect is possible then right? Like it says it should be explainable by normal means but they can’t think of a legitimate way other than fraud or statistical dishonesty (like the way they report it kinda)

It doesn’t necessarily forward the psi idea but wouldn’t this push more towards the idea that there is an effect?

 

[Elaedith]

Wait, I’m confused

It basically says that there’s a larger effect in higher quality studies, and that there’s a clear effect?

In the discussion it seems to list a bunch of possible explanations and then say why they are unlikely, and the only ones that seemed somewhat plausible were fraud and statistical dishonesty (the second of which tbf i heard from Louie savva is rampant), but then it says even with that it’s not fully likely

And in the conclusion, it basically says they don’t know how this anticipatory effect is possible then right? Like it says it should be explainable by normal means but they can’t think of a legitimate way other than fraud or statistical dishonesty (like the way they report it kinda)

It doesn’t necessarily forward the psi idea but wouldn’t this push more towards the idea that there is an effect?

They seem to be leaning towards the conclusion of a genuine, non-paranormal effect. The claim that it is a natural effect has to be partly based on prior probability along with the fact that the effect is not generally of the kind that is actually predicted by psi researchers, even though it does seem anomalous. I think the main point is that it needs independent replication by multiple researchers who are not interested in psi, and with consistent methods and analyses.

Just in the first paper you linked to, I noted a few things. First, the effect reported was only for one hemisphere, hence the significant result is for a stimulus category x hemisphere interaction. The p value is rather oddly reported as p 'less than or equal to .05' rather than giving an exact p value. In fact, the exact p value with F (1, 19) = 4.21 = .0542, so just on the margin of significance. The sample size is small. With only one of several main and interaction effects significant, a study with low power and marginal p value, there isn't a lot of evidence against the null hypothesis. The p value gives the probability that the effect could occur if the null is true, not the probability that the null is true.

 

[derem]

They seem to be leaning towards the conclusion of a genuine, non-paranormal effect. The claim that it is a natural effect has to be partly based on prior probability along with the fact that the effect is not generally of the kind that is actually predicted by psi researchers, even though it does seem anomalous. I think the main point is that it needs independent replication by multiple researchers who are not interested in psi, and with consistent methods and analyses.

Just in the first paper you linked to, I noted a few things. First, the effect reported was only for one hemisphere, hence the significant result is for a stimulus category x hemisphere interaction. The p value is rather oddly reported as p 'less than or equal to .05' rather than giving an exact p value. In fact, the exact p value with F (1, 19) = 4.21 = .0542, so just on the margin of significance. The sample size is small. With only one of several main and interaction effects significant, a study with low power and marginal p value, there isn't a lot of evidence against the null hypothesis. The p value gives the probability that the effect could occur if the null is true, not the probability that the null is true.

Reading the followup papers of the meta analysis by mossberg made it seem like he was an advocate of some sort of psi effect, mossenburg often writes with radin and tressoldi definately seems like a psi advocate in his later experiments of thgis

However, to be honest, the small size of the effect makes me think it probably is some sort of combination of the file drawer effect, expectation bias and faulty statistical analysis by the original studies. What i find even more interesting is that their attempt to explain away the possible faulty statistical analysis of the original researchers seemed really lacking to be honest.

I thought the significant result was significant enough in the right hemisphere that it was significant overall, though i might have misread it. What do u mean by F(1,19)?

That being said, i was reading a phd by louie savva and he seemed like he has read most presentiment studies, and has said that he isnt impressed at all bu them and alot of results are a combination of filedrawer, biases and straight up fraud. I found it interesting that in a follow up meta analysis, they showed a box chart with the significance and effect size of each experiment with a box score chart

What was interesting was one of savva's experiment was not significant and one was just as significant as the rest, but he himself has said his study was unconvincing and in retrospect the significant results seemed to be from faulty analysis and wishful thinking or something like that, and it was pretty much in line with all the other studies in terms of effect soze. He also did 3 more informal studies he did that werent published were all non significant.

Outside of tressoldi, whose results look a good deal more positive and consistent than the rest, 9/22 of the studies look non significant which I feel makes it pretty questionable considering the effect size is so small even in succesful studies. Their attempt to explain away publication bias, i coildnt really understand it but it seemed farfetched to me, and when i researched it it didnt seem lole a super convincing way to do so either (they used some formula, but apparently that makes sweeping generalizations of possible unpublished studies being similar to the current ones which seems pretty odd)

I feel the biases could explain alot of the results, and im also curious as to if the raw data is as impressive in all these experiments as is claimed.

What do you mean when you say only one of several effects by the way? Theyve essentially tested the same effect multiple times so wouldnt that rule out them looking for one rather than having a hypothesis one would be different and testing it? or am i misunderstanding?

In terms of p valie, i understand what it means but something i dont fully get. So 5% is usually the cutoff, but if the p value is 5.5% then, assuming its not by a bias, theres a 5.5% chance this result would be gotten by chance?

 

[Elaedith]

Reading the followup papers of the meta analysis by mossberg made it seem like he was an advocate of some sort of psi effect, mossenburg often writes with radin and tressoldi definately seems like a psi advocate in his later experiments of thgis

However, to be honest, the small size of the effect makes me think it probably is some sort of combination of the file drawer effect, expectation bias and faulty statistical analysis by the original studies. What i find even more interesting is that their attempt to explain away the possible faulty statistical analysis of the original researchers seemed really lacking to be honest.

Not sure who Mossberg is; do you mean Mossbridge? Looking at the paper again, I think you are right. Although it seems to state rejection of a paranormal effect, it seems to simultaneously imply real effect that violates the laws of physics (which is normally what's meant by paranormal). I wouldn't trust the conclusion of the authors, although it does suggest some possible explanations that could be followed up. I do think that p hacking, publication bias and faulty statistics are plausible explanations.

I thought the significant result was significant enough in the right hemisphere that it was significant overall, though i might have misread it. What do u mean by F(1,19)?
<snip>

What do you mean when you say only one of several effects by the way? Theyve essentially tested the same effect multiple times so wouldnt that rule out them looking for one rather than having a hypothesis one would be different and testing it? or am i misunderstanding?

In terms of p valie, i understand what it means but something i dont fully get. So 5% is usually the cutoff, but if the p value is 5.5% then, assuming its not by a bias, theres a 5.5% chance this result would be gotten by chance?

I was looking at the results for the slow wave amplitude measurements which was a repeated measures ANOVA with three topographical factors and one stimulus category factor. The only 'significant' effect reported was the hemisphere X stimulus category interaction. This was because they apparently only found a difference in SW activity for the right and not the left hemisphere. This seems to be fishing because there is no apparent reason to predict this interaction and in fact no clear predictions. If the effect in the right hemisphere was large enough to produce an overall effect of target versus non-target they would have reported this. The F(1,19) are the degrees of freedom that can be used with the F value to calculate the exact p value which they have not reported. P value of .05 means there is a 5% probability of obtaining this effect when the null is true. This is often misinterpreted as the probability that the null is true.