A foolproof method to validate psi phenomena: an open challenge to the telepathy tapes.

The telepathy tapes presented an intriguing story about non-verbal autistic kids that can read the mind of their parents. In the words of Ky Dickens, the creator of this podcast, telepathy is just the tip of the iceberg when it comes to the psychic abilities on these kids. When you go in-depth into the story, you start running into testimonies of how allegedly telepathic kids can read the mind of people who are far away or even within the distant future.

One of the main criticisms is that these children write with the help of someone else, and that it is the caregivers who, so to speak, put words in the children’s mouths. However, the first season of the podcast presents a very complex framework that cannot be explained by the simple assistance of the caregivers (assuming we are not being blatantly lied).

Body of evidence

I listened to the story, and I honestly loved it. There is good reason to believe in psychic phenomena: scientific evidence. A systematic review and meta-analysis of studies on remote viewing found that individuals tested for their ability to perceive and describe details about a distant or unseen target without relying on sensory input or prior knowledge achieved a hit rate that was 19.3% higher than expected by chance (with a 95% confidence interval: 13.6%–25%). This review included 36 controlled studies conducted up to December 2022.

Of course, skepticism is warranted—and rightly so. Systematic reviews and meta-analyses are only as robust as the quality of the data they synthesize. If the underlying studies are flawed, biased, or methodologically unsound, the resulting conclusions will inevitably reflect those weaknesses. Aggregated results might mask critical flaws in individual experiments, such as inadequate controls, publication bias, or subjective interpretation of results. That was my a priori assumption when I decided to look at individual studies to evaluate their methodology and determine whether their findings hold up under scrutiny. I also decided to read papers that “debunked” remote viewing experiments to save some of my time.

Individual studies

Below, I will present the “average” protocol for remote viewing experiments:

1. A set of images / locations / targets are chosen or generated and put in sealed envelopes.
2. A number is assigned to each of these envelopes.
3. A number is randomly generated to choose one of those envelopes.
4. A person that does not know what the envelop contains guides a remote viewing session and the remote viewer is asked to generate impressions (writing, drawing, or modelling).
5. A judge that does not know the target ranks the impressions based on the description of the remote viewer.
6. A statistical analysis is conducted to check whether the number of hits is significantly different from what you would expect from chance or something similar.

Samples

Before moving to the statistical analysis, I want to provide some examples of targets and impressions from different studies where the blinding between steps 1 and 4 is clearly described. Disclaimer: I am going to cherry pick my favorites:

Blinding techniques	Target	Impression
1. Isolation between the remote viewer and the target selectors (visually, acoustically, and electrically). 2. The remote viewer did not know how the target was selected or who selected it. 3. The target was selected after the viewer was isolated. 4. The experimenters opened a dictionary page arbitrarily and selected the face word that could be drawn. (IEEE) 5. Target sealed in envelope unknown to remote viewer and the session facilitator.
1 to 5
1, 2, 3, 5, 6. Target chosen by a person outside of the research team.
1, 2, 5. 7. Random target at 30 minutes driving distance from research center (list of 100+ targets).	Tennis courts
1, 2, 5, 6 and 7 (Nature)	Stanford’s Hoover Tower	“The area—I have a place—seems like it would be Hoover Tower”.
1, 2, 5, 6 and 7	A boat marina	“What I’m looking at is a little boat jetty or boat dock along the bay. It is in a direction about like that (pointing) from here. Yeah, I see the little boats, some motor launch (sic), some little sailing ships…”
1, 2, 3 and 5 (JSE) PS: The viewers knew the targets were rocks.	“The sample (F) is a beautiful specimen of crystalline halite, which is salt, and in this almost pure form is practically transparent (in fact, looks very much like quartz). It has beautiful cubic cleavage on part of the sample, and I can see through it. This sample was taken in St. Thomas, Nevada. Halite is formed from sedimentary evaporite beds.”	RV1. “I think I’ll settle for a chunk of crystal of some sort, formed by dripping and evaporation. Location by specific state: Northern Nevada?” RV2. “I have the impression I could “look” right through it. My analytical overlay is providing lots of alternatives. Damn, wish it would keep still. Crystal, crystal, crystal ball, glass, crystal clear crystal.”

The case against remote viewing

Most criticisms have focused on the errors that may have occurred AFTER the drawing was made that could bias the statistical estimates. Scott and colleagues, for example, wrote several extensive notes about how the judges might have being cued about the right result. They all have this “gotcha!” spirit. However, whether this is true or not, cueing the judge can not explain away the quality of the above results. I have collected some of this blog post so the public can see by themselves.

Certainly, one could argue against the accuracy of some statistical estimates, but there is no way of denying that at least some viewers are definitely tuning into something. For a brief overview of the statistical debate, I recommend this one-sided video by the former president of the American Statistical Association, Jessica Utts, ironically labeled by google as “parapsychologist” after her findings.

A new scientific paradigm

What is interesting about remote viewing positive findings is that they disprove our assumptions about space-time. On one of the experiments the remote viewer accurately identified two pools, one rectangular and one circular, close to a very prominent water tank, which was not in the target location. However, said tank had been part of the target location years before the experiments were conducted. Was the remote viewer looking into the past?

The alternative scenario, remote viewing into the future, has also been essayed. You can read one of the papers about it, at the very least look at the Figures and captions. The quality of those matches are very similar to the results presented above.

The telepathy tapes challenge

One of the main concerns about the telepathy tapes, particularly from the videos that were shared in the members page, is the presence of auditory, visual, and kinetic cues. However, if this telepathy works somehow like remote viewing does—and a lot of the stories in the podcast certainly suggest traveling across spacetime—we can apply an up-to-date protocol that assumes competence and gets rid of any potential present-day cueing.

Setup

A livestream or otherwise time-stamped camera featuring (1) the kid, accesing whatever they use for communicating and (2) a monitor with a random image generator in a separate room.

Protocol

1. The kid is explained that they must access the information that their parents will be shown in just a few minutes.
2. The kid guesses whatever is shown in step number 3. He or she can use the aid of the carer in whichever form they like. Blinding is not required at this stage, parents can steer the kid as much as they want. Therefore, preserving the bond and environment that had been acknowledged as required by the podcast.
3. We run the random image, number, or word generator and the parent then enters the room to see the image.

A judge could judge whether the images correspond to the respond in step number 2, but words or numbers will certainly give away exact matches without requiring any judge.