Protocol For Medium Demonstration With SGU and Skeptiko
Listen to the Podcast: Dr. Steve Novella on Psychic Medium Research Protocols
Protocol For Medium Demonstration (Draft): A cooperative protocol between the SGU and Skeptiko
1) Alex Tsakiris will select 3 mediums to be tested. He can use whatever criteria he wishes.
a. The mediums must know they are being tested and be willing to subject themselves to the protocol as outlines here.
b. The mediums will agree ahead of time in writing that the protocol is fair and that they are able to conduct a successful sitting under the conditions of this protocol.
2) The SGU will recruit at least three sitters, each sitter will receive a reading from 1 medium. The sitters will be:
a. Unknown prior to this study to either the SGU, Alex Tsakiris, or any of the mediums to be tested.
b. The sitters will be told that they are to be volunteer subjects for a study involving a psychic reading, but will not be given further details.
c. The sitters will be given a questionnaire (detailed separately) prior to the sitting for the purpose of documenting as much personal information as possible for later scoring. The sitters will not be told the specific purpose of the questionnaire.
d. The SGU will coach each sitter prior to the reading so that they understand they are not to give any information to the medium during the reading.
e. Alex Tsakiris will have the opportunity to meet and question each sitter just prior to the reading, but the sitters will be instructed not to divulge any personal or demographic information about themselves.
3) The Readings
a. For each reading the medium will have no contact with or information about the sitter prior to the reading.
b. Each reading will last 30 minutes and will be recorded (audio and video if possible) to aid in later scoring and for objective documentation.
c. The medium is allowed to ask the sitter only yes/no questions.
d. The sitter is allowed to answer questions either “yes”, “no” or “I don’t know.” The sitter is not to divulge any specific information about themselves to the sitter.
e. If the sitter inadvertently reveals inappropriate information to the medium that will not disqualify the sitter or the reading, but that information will be considered when scoring the results. Specifically the medium will not get credit for any “hits” that result directly from information they received from the sitter.
f. After the formal reading has concluded the medium and sitter will not interact.
g. The SGU and Skeptiko will monitor the reading to their satisfaction, but neither can interfere with the reading – they can monitor only.
4) The Scoring
a. The scoring will be done cooperatively between SGU and Skeptiko.
b. Each factual statement of the medium will be scored as either to vague to score, general, or specific
c. Each statement will also be compared against the biographical information obtained from the questionnaire and scored as either a miss, a partial hit, a direct hit, or insufficient information to score
d. A list of the factual statement in written form will also be produced and agreed upon by SGU and Skeptiko and the sitter will be allowed to score each statement as either a miss, partial hit, direct hit, or not sure.
e. Hits only apply to the sitter or those known to the sitter.
f. Future predictions will also not be scored as hits.
May 21st, 2008 at 4:45 pm
I didn’t listen to the podcast (yet?) but I have some suggestions…
The protocol will leave the medium open to charges of cold reading.
Having the medium give the reading with proxy sitters and with no feedback from the sitter can eliminate charges of cold reading.
The protocol also leaves the results open to charges of experimenter bias.
For example, what if the name of a spirit given by the medium is Rickey but the sitter gives the name of the spirit as Richard? A biased skeptical experimenter would probably say it’s a miss a biased believing experimenter would probably say it’s a hit.
If sitter is only allowed to say yes or no, some sitters may say yes and some may say no to a variation on the name or a vague bit of evidence. This is not just a problem in scoring that can be resolved afterwards, it can effect how the medium will proceed with the reading. In a large experiment with many sitters these differences in sitters may average out but in an experiment with three mediums and as few as three sitters it can have an effect. This is another reason why feedback from the sitter should not be allowed.
Who decides what is vauge? Are there written guidelines that distinguish vague from specific? Sometimes what is vague or general can be less than clear cut. If I ask a spirit to show me how he spent time with the sitter, and the spirit shows me a field of wheat, is that vague or specific? If the sitter lived in the city and only went out into country once in his life to a farm where he met one other person who is the spirit, it is specific. If the sitter was a farmer and had family reunions every year of his life living on the family farm it might be vague since there could be many spirits who might be identified by the evidence of the wheat field.
If you use proxy sitters, you can eliminate charges of experimenter bias by judging the reading by asking the sitters identify their own reading from a group of two or more readings where one of the readings in the group is the reading the medium gave for their proxy sitter. This eliminates ambiguity and bias in judging since you only get a hit or a miss (the sitter identifies his own reading or not).
If the readings are very general or very detailed it doesn’t matter. For the sitter to correctly identify his own reading requires specificity. This way the results are independent of subjective distinctions like vague, specific, etc. However you need more than three readings per experiment to get a good statistical sample and the number required depends on the frequency of hits.
The protocol does not specify how the results will be assessed as to the probability they occured by chance. Before a medium agrees that the protocol is fair he will want to know what meaning will be attributed to possible results. There must be agreement on what will be considered a demonstration of paranormal phenomena for the protocol to be deemed fair. (What’s the point of the experiment unless you are trying to demonstrate paranormal phenomena?)
What steps will be taken to verify the information provided by the sitters?
It is a common occurence for sitters to remember information a day or more after the reading. How will sitters be able to correct information or scoring if they realize mistakes occurred afterwards? How can a medium agree it is fair if there is no way to correct such mistakes?
A questionare is not sufficient to extract all relevant information from the sitter since a spirit that is not related to the sitter or who was only slightly known to the sitter might come through during the reading. In such a case the significance of evidence given by the medium is may be remembered only after the experiment is over. There should be a way for the sitter to correct mistakes made in scoring at a later date. Possibly a follow up interview one week later would be acceptable.
The protocol does not say what process will be used to select sitters except that they will be selected by SGU. What steps will be taken to ensure that bias will not effect how sitters are selected? How can the medium agree the process is fair without this information? Skeptical sitters might have higher standards for recognizing evidence (ie answering “yes”) and this could effect the medium’s ability to do the reading.
All contact between the sitters and SGU should be video taped and audio taped just as all contact between the sitters and the medium is taped.
How can the medium know that they will be able to conduct a successful sitting under the protocol until they try it? What if the medium gets to the laboratory and finds there is distracting background noise or distracting activity in the general area? Who determines if complaints by the medium are reasonable?
(I assume everyone is familiar with the work of Gary Schwartz, it is a useful example.
http://veritas.arizona.edu/papers/Beischel%20EXPLORE%202007%20vol%203.pdf )
May 25th, 2008 at 5:33 am
Firstly, may I congratulate you on a fascinating podcast. I recently stumbled across the site and downloaded a few key episodes. Initially I downloaded episodes with Blackmore and Wiseman but then that led me onto all the others. Congratulations on creating a podcast and interviews that are truly skeptical - i.e. questioning, and even questioning the so-called skeptics.
The mediumship protocol that was recently posted in draft form is a step in the right direction but not complete yet. I will look around the site and more recent podcasts that I haven’t listened to yet in case there’s more information about what you’re attempting here. There appears to be an attempt to replicate (and perhaps improve upon) Schwartz’ work. Perhaps I’m wrong. If I’m right then can I suggest that you look back at the wealth of published work in this area by psychical researchers and early parapsychologists who pretty much ‘nailed’ the protocol (for want of a better expression). The sorts of key figures I’m thinking of are Saltmarsh, Soal, Pratt & Birge. Recent replications of these early protocols and scoring methods are 100% robust against criticisms of sensory leakage, lack of control, temporal cues etc. If, once you’ve read these early studies, and recent replications, you decide to veer off on a separate protocol than that is fine, though justifying it in Parapsychology circles would be difficult.
Will be following your progress intently and, once again, congratulations on the creation of a truly listenable series. Excellent!