Correlation Matrix Method

Perhaps this deserves its own thread. I understand it only in the broadest terms, but it's claimed to be a protocol capable of overcoming replication problems in parapsychology. Caroline Watt has reported "very" significant results using the method.

Here is a short presentation from the Parapsychological Association Convention in Paris this year by Hartmut Grote, entitled "Blind Analysis of a New Correlation Matrix Experiment":

[Laird]

I understand it only in the broadest terms

I'm not sure I can even say that, even after watching the video. Tried to track down details of Caroline Watt's experiment but the best I could find was a one-page abstract: https://www.research.ed.ac.uk/portal/en/...cbd5).html

I'm not sure I can even say that, even after watching the video. Tried to track down details of Caroline Watt's experiment but the best I could find was a one-page abstract: https://www.research.ed.ac.uk/portal/en/...cbd5).html

Yes, that's all I've seen of the Caroline Watt study.

I've now watched the Grote presentation, and I still understand it only in the broadest terms - that they are looking at correlations between a large number of pairs of variables (in this case, for each participant, five variables derived from the way in which they pushed the buttons, against five derived from the output of the random number generator, for each of nine runs per participant, giving 2025 pairs of variables). And then they compare some measure derived from those correlations, to some kind of estimate of what they would expect by chance, to see if there's a significant deviation from chance.

What struck me about this presentation was the "blind analysis" aspect. If I understood correctly, he didn't know what the result of this 18-month experiment was until he ran the analysis program at the end of the talk. And didn't even say what the result was, let alone commenting on it. The synchronisation of the graphics was messed up, which made it confusing, but the result was non-significant, and nowhere close to significance.

[Laird]

The synchronisation of the graphics was messed up, which made it confusing, but the result was non-significant, and nowhere close to significance.

Yep. I assumed that the graphic was supposed to update when he "clicked the final button", which it didn't, but that if it had, we would have seen a significant result. Not sure what the non-significant result was doing there right from the start though...

Yep. I assumed that the graphic was supposed to update when he "clicked the final button", which it didn't, but that if it had, we would have seen a significant result. Not sure what the non-significant result was doing there right from the start though...

I suppose it was some kind of animation, and when the video version was prepared they just used the final static image.

Incidentally, I noticed he had an affiliation to Cardiff University as well as the Max-Planck-Institute. His research interests on the Cardiff website conclude with:
"I also have an interest in the application of precision interferometry techniques to other fundamental physics questions, like the measurement of birefringence of the vacuum, the search for new particles beyond the standard model, and more exotic questions."
https://www.cardiff.ac.uk/people/view/10...te-hartmut

I suppose the last four words are code for parapsychology.

[Laird]

I suppose it was some kind of animation, and when the video version was prepared they just used the final static image.

That seems plausible, yes. But then, as you already asked, why the low significance?

That seems plausible, yes. But then, as you already asked, why the low significance?

From the sceptical point of view, perhaps that could be turned round and we could ask why the high significance in the other studies? I've never had a chance to look at the statistical methods these studies used, but I'm sure that working out the chance distribution of a measure of correlations in a huge matrix of non-independent data isn't an easy thing to do.

Grote says in the talk that he uses a method based on permutations of the data, which sounds a bit like the method Dean Radin often uses for comparing measured correlations between two time series, by applying a lot of random time shifts to one of the series and finding the distribution of the resulting correlations. It would be better (but obviously a lot more laborious) to base the distribution on a lot of different time series, rather than reusing the same data over and over again. Probably the permutation method will give a good approximation in some circumstances, but I can imagine that in others it might systematically over-estimate statistical significance. So possibly there might be some devils lurking in the technical details of these studies.

The Parapsychological Association has made available another video of a presentation at its 2019 Annual Convention, also related to the methods developed by von Lucadou and co-workers. It's entitled "Micro-Psychokinetic Observer Effects on Addiction-Related Stimuli" and is presented by Moritz C. Dechamps of the Department of Psychology at the Ludwig Maximilian University of Munich. In the description on YouTube, Markus A. Maier is also credited:

Here are copies of the papers mentioned as the theoretical basis of this work:
(1) Synchronistic Phenomena as Entanglement Correlations in Generalized Quantum Theory
Walter von Lucadou, Hartmann Römer and Harald Walach
Journal of Consciousness Studies, 14(4): 50-74 (2007)
http://www.deanradin.com/FOC2014/Lucadou2007.pdf
(2) Intentional Observer Effects on Quantum Randomness: A Bayesian Analysis Reveals Evidence Against Micro-Psychokinesis
Markus A. Maier, Moritz C. Dechamps and Markus Pflitsch
Frontiers in Psychology, 9: 379 (2018)
https://www.frontiersin.org/articles/10....00379/full

And this looks like a preprint covering the work presented at the Convention:
How Smokers Change Their World and How the World Responds: Testing the Oscillatory Nature of Micro-psychokinetic Observer Effects on Addiction-related Stimuli
Moritz Dechamps and Markus Maier
https://psyarxiv.com/bwsj5/

And this looks like a preprint covering the work presented at the Convention:
How Smokers Change Their World and How the World Responds: Testing the Oscillatory Nature of Micro-psychokinetic Observer Effects on Addiction-related Stimuli
Moritz Dechamps and Markus Maier
https://psyarxiv.com/bwsj5/

Actually, that was published in the Journal of Scientific Exploration last year:
https://www.scientificexploration.org/jo...sue-3-2019