Using error correction codes in ESP (PSI) experiments (Blog)

[jacob]

Discuss here the following article from mind-energy.net blog.
A member of our parapsychology forum , advised using FEC (Forward Error Correction) codes in psi experiments. This should help, in his opinion, create a more remarkable result out of psi test. Here’s an edited excerpt of his post (a little technical):

Most ESP tests show only a slight (but consistent) deviation from the expected results (e.g. in a test where the nominal chance of a hit is 50%, someone consistently scores 52%). In the long run, such an effect is statistically significant, but it is by no means “impressive” by human measures. BUT, consider these:

3. ...So, instead of trying to apply ESP on “raw” tests (e.g. testing telepathy with Zener cards), and just comparing the results with the expected hits, why not make the objects (e.g. the Zener cards) code some digital data (the “message”), protected with a heavily redundant error-correction code (e.g. one that increases data size to 1500%, but only needs 10% of the transmitted, protected data to recover the original). As the “receiving” subject senses the cards, assuming that we are witnessing actual psychic ability, the received results will likely contain enough signal (that is, receptions produced by ESP, not by random chance, i.e. noise) to recover the original message.

If you’re not familiar with FEC, here’re a couple of real examples which I gave later in the forum. I studied some of it in university but it was quite a few years ago so I might not be 100% accurate, so beware.

A basic example would be a Compact Disc. You know that even if it has scratches in a moderate amount, the player will still play it will. This is since it can reconstruct the correct information using data that is still available (if there’s enough of it). A more basic example of actual ECC would be transmitting everything 3 times. Then, if one of 3 is received with error it’ll still be able to select the right result (2 of 3). Of course, there are much better ways.

Since I liked his idea I decided to email Dr. Dean Radin, one of the leading parapsychologists today. His first reply on the topic was as follows:


Continue reading "Using error correction codes in ESP (PSI) experiments"

Click here to view the article.

[anonymous]

Why haven't these experiments, which prove information can be transmitted psychically, led to the acceptance of parapsychology by mainstream science?

[jacob]

I think these papers require examination, and I'm not a very qualified researcher myself. I agree that the method of using FEC can help to find the evidence for a PSI type transmission. But I haven't specifically looked into the studies that Radin provided.

There's also another thread started on this topic by Miguel at Error Correcting Codes. He posted a quote from one of the studies there. For some reason he thinks the quote explains that it didn't work although I didn't understand it the same way myself. Perhaps it's my English ;-)

[anonymous]

Originally Posted by jacob

There's also another thread started on this topic by Miguel at Error Correcting Codes. He posted a quote from one of the studies there. For some reason he thinks the quote explains that it didn't work although I didn't understand it the same way myself. Perhaps it's my English ;-)

I agree with you. I think your english is okay.

I couldn't find many of these papers on-line. What I could find didn't seem to match the original proposal which was to use telepathy to transmit information. These papers seem to be on pk, precognition, and remote viewing. They do say that democratic votings improves accuracy.

Telepathy may give an even stronger result because you have two living people both making some effort so you might get a better result especially if they are attuned to each other.

Here's what I found:

Radin, D. I. (1990-1991). Statistically enhancing psi effects with sequential analysis: A replication and extension. European Journal of Parapsychology, 8, 98 - 111.


http://66.102.1.104/scholar?hl=en&lr...Parapsychology

The experiment, consisting of five blocks of 1,000 sequential analysis trials each, produced by a single subject, showed evidence for psi and replicated the finding that statistical averaging techniques can be used to enhance the “raw” hit rate.

This next paper was not on Dean Radin's list:

REDUNDANCY IN PSI INFORMATION
Implications for the Goal-Oriented
Hypothesis and for the Application of Psi
By J. E. KENNEDY

http://66.102.1.104/scholar?hl=en&lr.../psi/jp79.pdf+

But it references a study (RYZL, M. A model of parapsychological communication. Journal of Parapsychology, 1966, 30, 18-30.) that was on the list:

Ryzl (1966) reported that 19,350 binary trials (P = 1/2) combined by MV and a coding procedure predicted with 100% accuracy 15 decimal digits. A simple sequential sampling method was used until a strong majority was acquired for each decimal call. The published results cannot be evaluated in terms of efficiency of psi although the 61.90% binary scoring rate (CR = 33.10, but about one-third of the 19,350 trials were actually index trials) was increased to 100% for the decimal targets (equivalent CR = 6.5) which indicates that scoring enhancement clearly took place. The experiment was designed and carried out specifically to "furnish experimental proof that . . . application of ESP is possible in principle" (p. 22), so the experimenter's interest presumably focused on the outcome of the decimal trials.

This paper also discusses the papers Miguel quoted. The point of the paper is that the design of the experiment and the goal for the experiment will effect whether or not redundancy will improve accuracy. I think what he is saying is that psi is efficient so that if the goal of the experiment can be accomplished without redundancy then redundancy will not improve the results. To my understanding, that must mean that in some cases redundancy will decrease the accuracy per trial by providing more trials to generate the same overall accuracy. Therfore you have to be careful how to design the experiment if you want to get better results by using redundancy.



I couldn't find the Puthoff papers but I found something by Puthoff that referenced them and gives a clue to their contents:


http://www.biomindsuperpowers.com/Pa...itiatedRV.html


CIA-Initiated Remote Viewing
At Stanford Research Institute

by H. E. Puthoff, Ph.D.
Institute for Advanced Studies at Austin

Examples include the apparent lack of attenuation of remote viewing due to seawater shielding (submersible experiments) [8], the amplification of RV performance by use of error-correcting coding techniques [19,20], and the utility of a technique we call associational remote viewing (ARV) to generate useful predictive information [21].8"

references 19 & 20 are the Puthoff papers on Dean Radin's list:

Puthoff, H. (1985). Calculator-assisted psi amplification. In R. White and J. Solfvin (Eds.), Research in parapsychology 1984. Metuchen, NJ: Scarecrow Press, pp. 48-51.

Puthoff, H. May, E.C. & Thomson, M. J. (1986). Calculator-assisted psi amplification II: Use of the sequential-sampling technique as a variable-length majority-vote code. In D. Weiner & D. Radin (Eds.), Research in parapsychology 1985. Metuchen, NJ: Scarecrow Press, pp. 73-77

[jacob]

Ryzl (1966) reported that 19,350 binary trials (P = 1/2) combined by MV and a coding procedure predicted with 100% accuracy 15 decimal digits. A simple sequential sampling method was used until a strong majority was acquired for each decimal call. The published results cannot be evaluated in terms of efficiency of psi although the 61.90% binary scoring rate (CR = 33.10, but about one-third of the 19,350 trials were actually index trials) was increased to 100% for the decimal targets (equivalent CR = 6.5)

I wish I knew all the statistical terms used here, like CR.
But if I understood correctly, for the binary trials he got 61.9% correct "guesses" and for 15 decimal digits that were transmitted using the 19350 binary trials that result was 100% accuracy, right?

So this study shows that there is some ESP going on. By the way, what were the trials consisting of? What's the experiment?

Interesting idea about goal orientation. I'm finishing reading a book that specifically discusses the importance of research of "need-oriented" psi occurrences. I'll report on it later in more detail.