Control group outperforms mediums in psychic test

I was somewhat surprised to see the SPR retweet this:
https://bigthink.com/surprising-science/...belltitem1

The research team behind this study, led by brain researcher Arnaud Delorme, has previously analyzed brain activity, skin conductance, respiration, heart rate, and peripheral blood flow of self-professed mediums. During one 2013 study, four mediums produced results above chance. Seeking a neural correlate to psychic powers, they write,

"One medium showed a decrease in EEG frontal midline theta waves during accurate responses, suggesting a possible decrease in executive functions associated with successful responses. A limitation of that experiment was its low statistical power due to the relatively small number of participants and number of trials."

...A follow-up study was conducted in 2018. The researchers wanted to know if mediums could identify whether a person was alive or dead based on photographs. After looking over 404 photographs, five of a dozen mediums performed better than chance. As there was no control group, however, the team had to keep testing.

For this study, a dozen professional mediums and a dozen volunteers were recruited. Every photograph was of a deceased individual. This time the task was to identify whether the person died of a heart attack, a car accident, or from being shot. Each image was cropped in such a manner so that you couldn't easily recognize the cause of death. While giving answers, every volunteer was scanned for changes in neural blood flow and heart rate. The results surprised the team:

"Overall, participants were able to detect the cause of death of deceased individuals at statistically robust above-chance levels. Contrary to our expectations that mediums would perform better than controls, the controls performed significantly better than the mediums."

While there were different neural responses between controls and self-professed mediums, they didn't correlate to correct answers.

The full paper can be found here: https://www.sciencedirect.com/science/ar...2620302414

I'm well-aware that this BigThink company have been and still are very biased against anything even remotely religious, spiritual or paranormal (they have done 'interviews' with Dawkins, Nye, Randi among others), but I noticed in the link to the original study that Dean Radin was involved.

Well the SPR has always been more honest than the skeptics, who've been called about by the nobel physicist Josephson and one of CSICOP's own founders. THe SPR found frauds in their heyday, some of them even had the reputation for being like bloodhounds about it.

The variance in medium performance has been know at least since the days of Leorna Piper, though I think the Greeks & Romans recorded "misses" from oracles.

The question has always been why the good hits are so very good - of course one can always say the investigators were idiots or "in on the trick", but when I went through the cases I didn't see a reason to believe that.

That being said, I wouldn't force a belief in mediumship cases onto someone anymore than I'd want someone to force the claims of a fraud like Randi onto me. When there's sketchy history you have to give the doubter some reasonable leeway.

Addendum to my last post ->

Anomalous Information Reception by Research Mediums Under Blinded Conditions II: Replication and Extension

Julie Beischel, PhD et al.

... the reporting of accurate and specific information without prior knowledge, in the absence of sensory feedback,and without using deceptive means. Because the experimental conditions of this study eliminated normal, sensory sources for the information mediums report, a non-local source (however controversial) remains the most likely explanation for the accuracy and specificity of their statements.

I can't deny that it makes it look pretty bad that mediums performed worse than the control. To anyone on the outside that on its own would destroy any credibility that the phenomenon of mediumship might've had, since these were self proclaimed professionals. well clearly their profession means nothing, since average joes outperformed them. That's how I see this being taken and its not without reason. Even if other studies are brought forward that seem to show mediums performing well, this one will likely be the one people remember.

Its things like this that make me wish you could do a longer term training course study where you try teaching people to be meduims or similar and do continuous testing throughout teh process to see if they have a trend of improvement over a year or so. And then compare them to a similar group being trained for something different but getting the same tests as teh medium group. It would really go a long way to prove it one way or another, since if the medium group shows a trend of improvement and accuracy and teh non medium group doest it'd be hard to deny that the phenomenon is real given the amount of data collected over sucha  period of time.


The problem, as always is money, my estimation says a study like this could cost as much as $100 million to do, potentially more. Paying all the researchers, data collection and analysis people, participants, equipment, food, etc for a whole year. In my mind you'd get a thousand people in each group and you'd actually send them on two isolated retreats of sorts, putting them up in some sort of lodging where they more or less live the training as their job to avoid data contamination. At which point I bet it would cost a lot more money. That being said, I bet crowdfunding could solve this problem.

See I wonder why this happened. Because the study is regular guessing, but that isn't what normal mediumship studies are like. Guessing photos and information reception are pretty different things. To someone like me that's the biggest weird standout here.

[Laird]

There are some interesting comments in the Discussion => Limitations section of the paper:

If not due to chance fluctuations, then how might we explain the difference in performance between mediums and controls? Some of the mediums commented that they found it difficult to differentiate between the type of death, as they reported feeling the pain of the deceased individual, but not the cause of that pain. They might have interpreted a given type of pain as a heart attack, but a similar pain could have occurred by being shot in the chest, or by chest trauma associated with a car accident. Given this feedback from the mediums, future experiments involving cause-of-death categories should be designed to avoid similar confusions. Additionally, mediums reported that the time pressure of the experiment did not allow them to really connect with the deceased in-dividuals like they normally would, and so they felt forced to use different strategies to try to respond as fast as possible.

There are some interesting comments in the Discussion => Limitations section of the paper:

Funny how the article doesn't mention that. I'd expect a scientific take to include the limitations or flaws.