Afterlife Experiments Now Available at

Afterlife Experiment by Gary E Schwartz, PhD - March 2002


Evidence of Information Retrieval Between Two Mediums
(Campbell "White Crow" Readings)

Journal of the Society for Psychical Research, 2001

Evidence of anomalous information retrieval between two research mediums:
Telepathy, network memory resonance, and continuance of consciousness
Gary E. R. Schwartz, Ph.D. and Linda G. S. Russek, Ph.D.
Human Energy Systems Laboratory
University of Arizona 

Can mediums receive highly accurate and specific information under laboratory controlled blinded conditions: (1) prior to the scheduled time of the reading, and (2) during the reading in the absence of visual, auditory, or other potential sensory cues? A research medium investigated for three years in the Human Energy Systems Laboratory at the University of Arizona participated in a novel experiment with a unique sitter she did not know. The medium was in Tucson, Arizona; the sitter (who is also a research medium) was in Los Angeles, California. The experimental design involved three phases: Phase 1, a pre-reading contemplation period where the medium attempted to receive information about the sitter's deceased loved ones before the reading began; Phase 2, a silent-read period where the telephone was on mute so that the sitter could not hear the information received by the medium; and Phase 3, an actual telephone reading involving dialogue between the medium and the sitter. Specific information regarding names and relationships during Phases 1 and 2 were more than 90% accurate; the conditional probability of guessing the primary information by chance was less than one in 2.6 trillion. Information obtained during Phase 3 extended the findings, increasing the p value to 1 in 17,000 trillion, including four pieces of specific information unknown to the sitter and later confirmed. The design rules out conventional explanations of fraud, cold reading, vague information, statistical coincidence, selective sitter memory, and sitter rater bias. Three anomalous mechanisms may be involved: telepathy with the living, network memory resonance with the living (a superpsi hypothesis), and the existence of intentional, organizing consciousness. Paraphrasing William James, the present findings represent a genuine "white crow" research reading in mediumship science.

Contemporary research on anomalous information retrieval has been advanced through the cooperation of research-oriented mediums with exceptional talent and integrity (Schwartz et al, 1999; Schwartz and Russek, 1999; Schwartz, Russek, Nelson, and Barentsen, 2001; Schwartz, Russek, and Barentsen, 2001) and scientists open to exploring their abilities.

In order to establish anomalous information retrieval, it is essential to rule out (1) conventional sources of information (e.g. visual and auditory) obtained intentionally (e.g. through fraud or cold reading) or unintentionally (e.g. via subtle cueing associated with the clients / sitters / subjects breathing), (2) experimenter bias or error, and (3) rater bias on the part of the clients / sitters / subjects (e.g. selective memory effects or inflated ratings of information) (reviewed in Gauld, 1983). 

Research in the Human Energy Systems Laboratory (HESL) has employed single-blind experiments where in the first part of the session, research mediums (i.e., mediums who engage in controlled research studies) are completely blind to the sitters (e.g. Schwartz, Russek, Nelson, and Barentsen, 2001; Schwartz, Russek, and Barentsen, 2001). The mediums do not see or hear the sitter. The mediums must report information out loud in the first ten minutes with no knowledge of the sitters' age, sex, appearance, tone of voice, and personal history and beliefs. Double-blind experiments where the sitters (1) do not hear the readings when they are performed, and (2) later score blinded transcripts (containing their personal readings and control readings of other sitters), are on-going in HESL.

The present report was written because of its research design and the evidential nature of the data. In Schwartz, Russek, Nelson, and Barentsen (2001), five mediums were investigated. We suggested that "the particular mediums participating in this research may be examples of five ‘white crows' of anomalous information retrieval."

Extending James' metaphor further, we suggest that the particular research design and reading reported here may be exemplary of a "white crow" research reading in mediumship science.

As the late Carl Sagan (and others - Sagan originally learned the phrase from Marcello Truizzi) have said, "extraordinary claims require extraordinary evidence." The present findings are consistent with this recommendation.

The single blind sitter-silent paradigm (termed the Russek procedure in HESL), was creatively extended by Laurie Campbell, LC, chair of our mediumship research committee. LC began doing private readings in the summer of 2000, and wished to extend the Russek procedure to her private practice. It occurred to LC that during her pre-reading contemplation sessions (the term "contemplation" was proposed by Stanley Krippner, LC calls them "meditations") which regularly occur approximately one-half hour before a scheduled reading, that she could request information purportedly from the sitter's departed loved ones before the reading began.

LC noticed that she typically received specific information - names, relationships, causes of death, personal description - that she could record prior to each reading and then confirm or disconfirm, item by item, with the sitter in the first part of the actual reading. At the time this report was written, LC had recorded data on over 100 readings using this pre-reading contemplation procedure (termed the Campbell procedure in HESL); her accuracy ratings range from 50% to 95%.

We learned of her procedure in December, 2000, and decided to conduct a controlled blinded laboratory experiment to investigate her claims. The design of the experiment is described below, followed by the data from one reading with a unique sitter whose evidence is so extraordinary as to deserve special scrutiny.

A traditional experimental report summarizing group statistics is in preparation. Statistically significant evidence for anomalous information retrieval was found for each of the sitters investigated in the experiment. However, it is the uniqueness and extraordinarily evidential nature of the particular reading highlighted in this detailed report that justifies focusing on this "white crow" research reading.

[Note: Given our commitment to integrity in research - both the experimenters and the mediums - we report all potential sources of error and / or fraud in the conduct of this experiment. Rather than hide these details, we express them openly, and let the reader reach his or her own conclusion about possible sources of contamination in the design.]

Experimental Design
Selection of Sitters

Sitters were chosen by the experimenters (the authors, GERS and LGSR) who met the following criteria: (1) the sitters had experienced significant personal losses of beloved individuals (2) the sitters were sincerely interested in the possibility of a spiritual reality and the survival of consciousness hypothesis (3) the sitters were mature individuals of exceptionally high integrity committed to the rigors of experimental research, and (4) the sitters were willing to score information collected in both the single blind and double blind conditions after the readings were conducted and transcripts were prepared.

The medium (LC) was blind to the selection and identities of the specific sitters.

The particular sitter whose data are reported here (George Dalzell, GD) is important because (1) he is professionally trained as a psychiatric social worker in Los Angeles County in California, (2) he comes from a highly educated academic family (he was educated at Northwestern University in Chicago, and his father, grandfather, and great grandfather were educated at Yale University), (3) he is deeply interested in research, and has validated data from four deceased individuals (described below), and (4) he has secretly been a medium for the past few years since the death of a dear friend (Michael, M) who was one of the four persons GD invited to serve as "departed hypothesized co-investigators" (Schwartz et al, 1999; Russek et al, 1999) in the present experiment.

The last fact is especially important, because GD has extensive data documented in a forthcoming book (Dalzell, 2001) that provides evidential support for the continued existence of M.

At the time of testing, LC and GD had never met (in person, by telephone, letter, or email). [Note: the mediums were willing to have their phone records examined; the skeptic will point out the possibility that if LC and GD were cheating, they could have used phone booths or phones of friends].

LC had been told of GD's existence (the authors of this report mentioned to her GD's interest in HESL's research a few months earlier). However, LC was not informed that GD would be one of the sitters selected to participate in this experiment. Moreover, LC was never informed that GD would be considered to serve as a sitter in any laboratory investigations. LC had not been shown GD's manuscript (Dalzell, 2001). The manuscript was kept out of sight of LC. [Note: a number of details obtained by LC in the readings were not only absent from the book, but they were unknown to GD prior to the readings. Hence, the thesis that LC somehow read the manuscript is improbable.]

GD knew of LC's existence (GD had read reports from the laboratory describing research with LC; Schwartz et al, 1999; Schwartz and Russek, 1999; Schwartz, Russek, Nelson, and Barentsen, 2001). GD was told that LC might be the medium in this experiment (he knows HESL is working with a number of research mediums). Given GD's academic background and professional standing, as well as his personal interest in this research area, the opportunity to participate as a research sitter in this experiment was of professional and personal significance to him. Engaging in fraud in the laboratory, if detected, would greatly jeopardize his career as a social worker as well as a medium. He was made aware (as are all mediums and sitters) that if fraud was detected in the laboratory, it would be exposed.

The experiment was conducted in Tucson, Arizona. LC was flown in from Irvine, California to participate in research. The sitters were located in Arizona and California. All readings were conducted over the telephone. Distances between the medium and the sitters ranged from approximately ten miles to greater than one thousand miles.

The sitters were telephoned by GERS and LGSR and invited to participate. The procedures were described in detail, including the pre-reading contemplation periods and the sitter-silent periods. Sitters were encouraged to attempt to personally connect with their deceased loved ones and invite them to communicate with LC prior to their scheduled appointment times with LC. Sitters were told that as in our previous experiments (Schwartz et al, 1999; Schwartz and Russek, 1999; Schwartz, Russek, Nelson, and Barentsen, 2001; Schwartz, Russek, and Barentsen, 2001), LC would be completely blind to their identity.

The sitters were reminded of the need for absolute integrity, and that they would be required later to carefully score transcripts from the readings. They understood that some of the transcripts they would recognize, and others they would not (Phases I and II described below). We explained that they would not be told ahead of time which of the blinded transcripts were their transcripts until they had completed the scoring and we had analyzed the data.

[Note: LC stayed at the home of GERS and LGSR for the weekend of the experiment. When one of the experimenters was speaking with a potential sitter on the telephone in a given room (to schedule an appointment), the other experimenter was with LC in another room (thus ensuring that LC was not "overhearing" the phone conversation. From the time that sitters were selected to the completion of the experiment, LC never left the home except when she was with the experimenters. Since the exceptional nature of the data reported here was not anticipated ahead of time, the experiment did not include additional desirable controls such as independent observers of all the procedures to establish that the experimenters were not engaged in conscious or unconscious deception. However, the experimenters are exceptionally sensitive to the possibility of unconscious "leakage" if not conscious collusion. It was deemed more effective to watch LC in order to rule out any possibility of the LC communicating with the sitters - recall the sitters were contacted only one day before data collection to participate in the experiment - then to have LC staying in a hotel away from the experimenters.]

Readings were scheduled for specific times (e.g. 2:30, 4:30, 6:00 p.m.). The procedure for each reading was as follows:

Phase 1: Pre-Reading (Campbell) Procedure. One-half hour before a scheduled time, LC would conduct her pre-contemplation period, in seclusion and silence. She would write down the information she received during the pre-reading period.

[Note that the blinded pre-reading contemplation phase eliminates all possible visible and auditory cues (as well as olfactory cues), and therefore eliminates conventional explanations of cold reading, subtle cueing, and medium fraud, as possible explanations of the findings.]

Phase 2: Sitter-Silent (Russek) Procedure. At the appointed time, depending upon the specific sitter, either the sitter telephoned the experimenters (GESR answered the phone) or the experimenter telephoned the sitter. A Sony digital video tape recorder was used to record the initial reception of the sitter and the conduct of the Russek Procedure.

The sitter was reminded that the telephone would be placed on mute (hence the sitters would not be able to hear LC speaking), and that they were to hold the telephone to their ear for the duration of the ten minute sitter-silent period. We confirmed empirically that the mute button worked effectively, and that the sitters could not detect words spoken by GERS before he handed the muted telephone to LC. Moreover, LC sat a few feet from the answering-machine that contained the mute button. She held the phone with her left hand, and wrote notes with her right hand. There was no possibility of her secretly attempting to manipulate the mute button.

When the telephone was placed on mute, the telephone was handed to LC. Since the sitter's telephone was not placed on mute (not all sitters had a mute button, and we wanted the sitters to focus their attention as if the telephone was being used in this phase of the reading), sporadic noises generated by the sitter and his or her environment could sometimes be heard by LC if she held the telephone in a normal listening and speaking position. The sporadic noises were distracting to LC. Hence, LC held the telephone with the instrument turned away from her ear, thus minimizing potential distractions.

[Note: One reviewer suggested that LC could have used cues such as breathing coming from the phone as feedback for cold reading. However, since the sitters could not hear LC, the "feedback" would have to be from sitters unconsciously receiving information from LC via telepathy, and then communicating agreement through subtle changes in breathing. This speculation would be a novel super-psi cold reading-type hypothesis.]

The Sony video camera recorded the Russek procedure during which time LC shared out loud what impressions she was receiving. Note that at no time did the experimenters refer to the sitters by name, and LC had not yet heard the sitter's voice. Phase II lasted approximately 15 minutes.

[Note that the blind sitter-silent phase eliminates possible visible and olfactory cues (and hence cold reading, subtle cueing, and fraud) as well as useful auditory cues (the sitter could not hear LC; and LC was not using sporadic distracting auditory cues to shape her responses).]

Phase 3: The Actual Reading Procedure.  When the sitter-silent period was completed, the experimenter took the telephone from LC, turned off the mute button, and started a Radio Shack telephone tape recorder that recorded both LC's voice and the sitter's voice. With the telephone on mute, LC's voice could not be recorded on the telephone tape recorder. However, LC's voice was recorded via the auditory channel of the video tape recorder.

The experimenter explained to the sitter that she or he would now hear LC's voice, and that LC would explain how she conducted a normal reading. LC then introduced herself and explained how she conducted a normal medium-sitter dialogue reading. LC then read, item by item, the content received during the pre-reading contemplation, and asked the sitter to confirm, question, or disconfirm the information. Hence, the sitter was not blind to the pre-reading contemplation content (the Campbell procedure) in this experiment.

However, we requested that LC not read the sitter-silent period content (the Russek procedure) in this experiment. This material was saved for later blind scoring by the sitters.

As will be discussed below, the pre-reading (Campbell) procedure can be used for blind scoring by sitters as well. However, this was not the purpose of the present experiment. The purpose was to document whether the pre-reading procedure generated discrete and specific accurate information under single-blind laboratory conditions.

The material obtained during Phase 3 (the actual reading) are obviously complicated by the dialogue established between the medium and the sitter. However, as will be clear, the content received during the actual reading complemented and extended the content received during Phases 1 and 2, thus providing compelling supportive evidence for anomalous information retrieval by LC.

Results for GD
Names and Relationships for Phases 1 and 2.

The experimenters knew that GD planned to invite M as a departed hypothesized co-investigator. However, the experimenters were blind to the other individuals GD invited to purportedly participate in the research.

Subsequently, according to GD, he invited four specific individuals:
(1) M,  (2) a deceased aunt named Alice, A), (3) his father, Bob, B, and (4) another close friend, Jerry, J.

During Phases 1 (Pre-Reading) and 2 (Sitter-Silent) combined, LC reported:
(1) that she was being told that this reading was for someone named George (GD),
(2) that there was a deceased friend named Michael (M),
(3) that there was a deceased person named Bob (B) (relationship not specified),
(4) that there was a deceased friend named Jerry (J),
(5) that there was someone with a strange name that sounded like "Talya," "Tiya," and "Tilya" (T) (LC's phonetic spellings of the name she heard),
(6) that there was a deceased person named Alice (A) (relationship specified in Phase 3)
(7) there was deceased dog whose name began with an S. 

However, LC ended Phase 2 with a statement indicating that the primary deceased person for this particular sitter was a male named Michael (M).

[Note - except for the unusual name "Talya" (GD calls her "Tallia," names will typically be listed in this paper by initials; LC provided full names, not initials, for these individuals in Phases 1 and 2.]

Table 1 lists all names and initials recorded in writing by LC during Phases 1 - 3 (save for three scientists names, mentioned for completeness in the Discussion section). The sitter used a 0-3 ratings scale: (0) unknown to sitter (possibly a mistake, e.g. Joyce), (1) known but not close - e.g. GD knows of a "Shermer" but doubts that "Sherm" (LC's spelling) is important to the reading, (2) known and moderately close - e.g. sitter knows a "Fred" and LC might be referring to him, and (3) known and close (e.g. the names GD, M, B, A, J, "Talya", S, and K).

The names of the four people specifically invited by GD - M, A, B, and J - were received by LC during Phases 1 and 2. In addition, it turned out to be factually correct that GD had a friend with the unusual name that sounded like "Talya" (T) (GD called her "Tallia"), and that GD had a beloved dog with an S name (LC's guess about the actual name was similar in sound but not precisely correct).

Table 1 - Summary of Names, Relationships, and Ratings 

George (G) **  Self  Living  3
Jerry (J)**  Friend  Deceased  3
John  Great Grandfather  Deceased  2
B name (Becky, Barbara, Betty)  Friend  Living  1
Maureen  Friend  Living  1
Robert / Bob (B) **  Father  Deceased  3
Talya / Tily / Tilya (T)**  Friend  Living  3
S (Suzane) (S)**  Dog Deceased  2+
Sherm? (LC spelled)  Known Living  1


Michael (M)** Friend Deceased  3
Jerry (J)** Friend Deceased  3
Joyce ----- -----  0
Fred Friend Living  2
Francis ** Friend Living  3
Albert or Alfred ** Friend of friend Deceased  3
Alice (A) ** ----- Deceased  3
Elaine ----- -----  0


Michael (M)** Friend Deceased  3
Marcus ** Friend Living  3
Jerry (J)** Friend Deceased  3
Albert ** Joel's friend Deceased  3
Alice (A)** Aunt Deceased  3
George (G)** Self Living  3
Arthur Friend of friend Deceased  2+
Katherine (K)** Grand daughter Living  3
Joe / Joseph ----- -----  0


** Refers to people or pets of special significance to GD. ----- means that no information was provided by JC. Letters in parentheses refer to individuals used to calculate the conservative conditional probability, first for Phases 1 and 2, then including Phase 3. Adding the remaining names to the calculation would serve to inflate the overall conditional probability.

[Note: It is unfortunate that we did not think to have GD write down the names of the people invited 24 hours before the reading, and have this document notarized. One reviewer suggested that maybe GD was deceiving us and / himself about the people he invited to the reading so as to help the sales of his book when it was published. This speculation has no basis in fact, and is entirely inconsistent with GD's professional and personal history. If this was the case, it could not explain other facts in the data, such as the 4 pieces of information obtained by LC, unknown to GD, that he subsequently confirmed after the readings.]

LC reported receiving other names; three were not recognized by GD and therefore could be scored as errors (listed in Table 1 as Joyce, Elaine, and Joe / Joseph).

What is curious is that GD did not know any living people with these three very common names. However, both experimenters (had we been the sitters) do have close relationships with a Joyce, an Elaine, and a Joseph.

For the purpose of conservative scoring in Phases 1 and 2, we limited our analysis to the specific names, relationships, and details of the four people invited by GD (M, B, J, A) plus two other clearly important individuals (a person and a pet) close to GD (T, S).

As will be clear from the actual reading (Phase 3) described below, not only were each of the four primary people described accurately by LC, but four additional facts not known by GD and later confirmed by sources close to GD indicated that exceptionally accurate information was obtained for GD's deceased family and close friends.

The combined probability of names and relationships reported by LC during the pre-reading and sitter-silent periods is highly statistically significant. It is estimated conservatively as p < one in 2.6 trillion (1 in 2,624,400,000,000 ).
The procedure used for calculating this conditional probability is described below.

Conditional Probability Calculation
If we estimate that there are at least 15 common American male names (e.g. common male names of people well known by the experimenters include Al, Bill, Bob, Edward, Gary, George, Harry, Howard, John, Larry, Michael, Mark, Sam, Steve, Tom), and 15 common female names (names of people well known by the experimenters include Alice, Beverly, Cathy, Jane, Joyce, Joan, Kate, Karen, Linda, Lynn, Mary, Margaret, Martha, Rita, Susan), we can conservatively estimate that the probability of LC getting a specific name correct for a given sex is 1 in 15.

[Note - the number 15 was selected by the authors to be a conservative yet fair number. Obviously, other common names could be added to the list (e.g. additional common names of males known by the experimenters include Daniel, James, Paul, Ralph, Terry, and additional common names of females known by the experimenters include Ann, Judy, Lisa, Sara, Shirley). Increasing the estimated number of possible common names would only make the conditional probabilities all the more improbable by chance. For example, the conservative 1 in 2.6 trillion estimate would be multiplied by at least 5 * 5 * 5 * 5 * 5 * 5 (an increase of 15,625 for the six names given), to less than 1 in 41,000 trillion ). The specific examples of male and female names listed above were selected after the experiment was completed and the experimenters decided to attempt to calculate conditional probabilities for the findings. Subsequent analyses of first name frequencies at the University of Arizona and the US Census bureau, reported in Appendix A, document that 1 in 15 is a conservative estimate.]

[Note: One reviewer suggested that once the sitter gets a single name correct, the p value should be reduced accordingly. For example, 1 in 30 - male plus female - should be reduced to 1 in 29, with each name picked. However, this assumes that readings involve single names. In a recent experiment, a medium reported hearing "Michael times 2 - Michael times 2" - not realizing that the name of the son who died, followed by the name of the father died, were Michael Junior, and senior, respectively. The medium not only reported indicated that the son and father were both deceased, and the father died after the son, but he correctly implied that both had the name Michael (Schwartz, Russek, and Barentsen, 2001). Given that 1 in 30 is an underestimate to begin with, the assumption that the p value should be reduced will not alter substantially the overall value of the probabilities.]

Relationship can be conservatively estimated as 1 in 12 (e.g. mother, father, daughter, son, grandmother, grandfather, aunt, uncle, wife, husband, female friend, male friend).
[Note - ideally one might wish to give more weight to older persons such as grandparents and parents, and less weight to friends and children. For the present data, doing so would serve to make the conditional probabilities even more significant; we purposely adopted a relatively conservative procedure for this reading.]

The conditional probabilities were calculated as follows:
 LC said that this particular sitter was male (1 in 2) and his name was G (1 in 15). For this information, the probability would be 2 * 15 or 1 in 30 (p<.033). [Note - when a name is selected without a relationship given, the pool of possible names (male and female combined), as calculated here, would be 2 * 15, or 30 possible names].

LC said that the primary deceased person was a male friend and his name was M A male friend (1 in 12) named M (1 in 15) would be 12 * 15 or 1 in 180 (p<.006). The combined probability of G and M, so identified, is 30 * 180 or 1 in 5400.

LC said that there was a deceased friend name J. Friend named J would be 12 * 15 or 1 in 180 (p<.006). The combined probability of G, M and J, so identified, is 30 * 180 * 180 or 1 in 972,000.

 LC said that there was a deceased person named B. LC did not specify the precise relationship. Person named B would simply be 2 * 15 (p<.03). The combined probability of G, M, J, and B, so identified, is 30 * 180 * 180 * 30 or 1 in 29,160,000.
LC said that there was a deceased person named A. LC did not specify the precise relationship. Person named A would simply be 2 * 15 (p<.03). The combined probability of G, M, J, B, and A, so identified, is 30 * 180 * 180 * 30 * 30 or 1 in 874,800,000.

The combined probability value of 1 in 874,800,000 is an underestimate of the actual combined probability.

For example, consider what happens when we add the highly unusual T name "Tallia" (which LC spelled "Tilya") to the conditional probability. LC did not comment what the relationship was. However, the question arises, what is the probability of spontaneously guessing the unusual name Tallia / Tilya by chance? A conservative estimate is 1 in 100 (p<.01). When we add T to the combined probability of G, M, J, B, and A this brings the conditional probability to 30 * 180 * 180 * 30 * 30 * 100, or 1 in 87,480,000,000.

Also important is the dog with an S initial. LC did not provide information about the sex of the dog, though she later described the dog accurately (see below). If we estimate that of 26 letters in the alphabet, picking S by chance is conservatively 1 in 15 (clearly names starting with Q's, X's, and Z's are highly improbable, whereas names like Alice, Bob, Charles, Debbie, Edward, Frank, Gary, Harry, Jerry, Kathy, Larry, Mary, Peter, Susan, Tom), and the selection of dog is maybe 1 in 2 (dogs and cats are the primary pets that mediums seems to mention), reporting a deceased dog with an S name is at least 1 in 30 (p <.03). Add the S dog to G, M, J, B, A, and T brings the conditional probability to 30 * 180 * 180 * 30 * 30 * 100 * 30, or 1 in 2,624,400,000,000.

In sum, for these six names, one initial, and relationships per se, the conditional probability is at least p < one in 2.6 trillion.

[Note that by comparison, neither author has (1) a deceased person B, or (2) a deceased close friend M, or (3) a deceased close friend J, or (4) a deceased person A, or (5) any known person named Tallia, or (6) a deceased dog S. Clearly, some people will have one or more of these name-relationship pairs, but virtually no-one (except GD) will have the complete pattern of all the name-relationships received by LC in Phases 1 and 2. This hypothesis was empirically investigated with a control group and is reported in Appendix A.]

Additional Content during Phases 1 (Pre-Reading) and 2 (Sitter-Silent)
LC reported meaningful content during Phases 1 and 2 that was clearly relevant to GD and his deceased loved ones.

[Note: One reviewer correctly noted that many of these items were, by themselves "vague" and could apply to many people. What make this information important is the constellation of descriptions, specially when the more vague information is interspersed with highly specific content. Just because a medium reports vague information at times does not necessarily imply that the mechanism is "cold reading." The information, though general, may still apply to the individual in question.]

During Phase 1 (pre-reading) LC wrote that the sitter (who had yet to be telephoned) was concerned with "truth that is held within the soul's journey - journey of the soul's path - truth from someone with an M name" and that the sitter was preparing to "stand up and be counted." GD is preparing to stand up and be counted with the publication of his book (Dalzell, 2001). The names G, M, B, T, and a small dog were mentioned.

LC saw "candles burning." GD informed us that he had lit a candle just prior to the beginning of the experiment, an act seldom practiced by GD.

During Phase 2 (sitter-silent), LC reported East Coast and California (both correct - GD comes from the East Coast, and he currently lives in California), seeing science and books associated with the sitter and his deceased father (both true), new discoveries with radio and television in the future (all true), recent advancement with new directions (true), father deceased (true), the name "Michael" key to the reading (true), and Jerry.

Other material received by LC was more general and could apply to many people (e.g. that the sitter was loving and caring). It should be noted that LC said that GD's mother was deceased; GD informed us that this was an error; his mother was living and in good health at the time of the experiment.

As in the previous research (Schwartz et al, 1999; Schwartz and Russek, 1999, Schwartz, Russek, Nelson, and Barentsen, 2001; Schwartz, Russek, and Barentsen, 2001), it was during the actual reading (Phase 3), that the most extensive and detailed information was retrieved that replicated and extended the information obtained during Phases 1 and 2.

Phase 3 (Actual Reading) - Content outlined as it emerged, including four examples of information not previously known to GD. After reviewing Phase 1 with GD, LC began the reading by focusing on M. She also mentioned that J had passed recently (in the past 6 months, which was true), and that A was also strongly present. LC described M as a partner (which was true) and that M was GD's "muse" (an interesting phrase - remember that LC was blind to the identity of the sitter, though she now knew that the sitter was male and that the names she had received previously were accurate and important to the sitter).

She described M as seeing the "world" through "lots of glass." She returned to this fact at various points in the reading. LC could not interpret what "M was showing her." It turned out that M was an international purser and flight attendant for Lufthansa Airlines. He flew all over the globe, literally seeing the world through the glass windows of airplanes.

She described M's personality accurately - not only as loving and caring, but obsessively neat and "pristine" (true).

LC then moved to J. She saw him from the East Coast, Brooklyn area (true), and that he was "drinking and smoking" (true).

It is important that GD did not know at the time of the reading that J had lived in Brooklyn; this was confirmed afterward by a friend of J and GD. She saw him as often "intoxicated" and sitting on "bar stools" (he was an alcoholic for more than half of his life). She also said that he "stopped" drinking before he died (true).

After describing an "A" named male (details that were true), she returned to M. She saw M in a white kitchen that was "cozy" with "stone" (true).

LC said that M showed her where he lived: somewhere in Europe, and his parents have a "heavy accent" (M was German). LC reported that M was showing her a big city, and then M was traveling through the countryside to his home. Along the road to his house LC was shown a river and "centuries old stonework" (true).

LC claimed that M showed her an old, stone "monastery" on the edge of the river on the way to his parent's home. This information was not known to GD prior to the reading. After the reading, GD telephoned M's parents in Germany and learned that there is an old abbey / church along the river's edge on the way to their house, and that they had held a service for M in this monastery-like stone building a few weeks prior to the experiment.

LC then described the older Aunt A, her great sense of humor (true), but related that A was experiencing "compassion and sorrow" for her grand daughter Katherine (true) who was having difficulties and was "uncontrolled." LC indicated that K was currently receiving "healing." This detail was not known to GD prior to reading. After the reading, GD telephoned K's mother who informed him that K was indeed having serious difficulties and had sought psychological counseling in the week prior to the experiment because she was in crisis.

Also interesting was that GD originally thought that LC had misspelled Katherine's name; GD believed that Katherine's name was spelled "Catherine". GD learned that LC was correct.

LC described more details about A's cottage, and then moved back to GD himself. LC said she was being shown by M that GD's life was about to become "noisy" and be "turned upside down." This is true. With the publication of GD's book, his secret life as a medium will become a matter of public record, and he may have to face professional complications in his role as a psychiatric social worker in good standing with Los Angeles county Department of Mental Health, as well as his role as Psychiatric Admissions Coordinator at a Hospital in Glendale, California.

The experimenters found it interesting and evidential that LC reported M showing her GD and "white coat" / clinicians in a hospital prior to the reading. It turned out that GD had performed psychological evaluations in the emergency room of a Los Angeles hospital, and he had been there just prior to the reading.

The experimenters also found it interesting and evidential that LC saw the small dog (colors and personality description reported by GD to be accurate) being near a favorite tree and water. GD later informed us that he and his dog spent many hours at a special tree near the water, where his deceased dog was buried after living to be eighteen years old.

The complete reading lasted more than one hour. Substantial information was received. Space precludes reporting the complete reading.

[Note - the full transcript with detailed commentary from the sitter will be placed on a website for future scholars to examine.]

Some of the information received by LC can be labeled as general and is applicable widely (e.g. that Aunt A was loving, or M was compassionate). GD said that this information was at least 90% accurate. Though this information accurately portrayed the deceased and living family and friends of GD, given that this kind of information is general and widely applicable, it is not considered evidential in the context of the present report.

However, much of the information received by LC was highly specific and clearly organized for specific individuals (e.g. the pattern of "partner," "M," "world through glass," "small house in Europe," "old stone monastery," "on the river's edge," "lived life to the fullest"). GD said this kind of specific information was also at least 90% accurate. Because this information is so highly specific and precisely organized, it is decidedly evidentiary.

The design of this experiment is unusual because of the pre-reading contemplation period (Phase 1 - the Campbell procedure), the sitter-silent period (Phase 2 - the Russek procedure), (3) the demonstrated integrity of both the research medium (e.g. Schwartz, Russek, Nelson, and Barentsen, 2001; Schwartz, Russek, and Barentsen, 2001) and the sitter (Dalzell, 2001), and (4) the special qualities of the sitter (who is a psychiatric social worker, senior administrator, and recently a research medium himself).

The pre-reading contemplation condition in this experiment is especially important because it supports LC's hypothesis that she is able to receive information before readings actually begin. Since her private clients schedule appointments months in advance, it is plausible that their hypothesized deceased loved ones will be waiting for the scheduled moment in time when they can communicate through LC to their loved ones (LC's hypothesis; superpsi hypotheses are discussed below).

Future experiments can be designed to explicitly investigate the "scheduling variable" and determine if LC is more accurate during pre-reading contemplation periods if (1) appointments are scheduled, and (2) sitters invite their deceased loved to communicate prior to the reading.

Integrity of the Sitters, Procedures, and Experimenters
Having conducted research with LC for three years under increasingly controlled laboratory conditions, it is highly improbable that LC has engaged in fraud with her private clients. Skeptics who might wish to speculate that LC was using the internet or an investigator to get information about her private clients before the readings, and then using the pre-reading information to impress her clients, have no basis or justification to offer such an unfounded criticism. 

As mentioned previously, extreme care was taken to insure that only the experimenters' knew of the selection of the three sitters (their names, sexes, ages, and locations). The sitters were selected after LC arrived in Tucson. She stayed at the experimenters' home (and therefore was under close scrutiny). LC did not have a cell phone. Phone records can document that one brief telephone call was placed to GD on December 16th, 2000 (he was not home then); one call was made by GD that evening returning the experimenter's call to discuss the upcoming reading (he spoke only with GERS; LGSR was with LC in a separate room while GERS spoke with GD); and one call was made by the experimenters at 6:00 p.m. on December 17th, 2000 for the session.

In our three years of laboratory research using carefully controlled conditions where LC is blind to the sitter's identity (e.g. Schwartz, Russek, Nelson, and Barentsen, 2001; Schwartz, Russek, and Barentsen, 2001), the possibility of fraud as well as cold reading have been essentially ruled out. The present report indicates that the systematic pre-reading observations collected in LC's private practice can be replicated and extended under controlled and blinded laboratory conditions.

Paraphrasing William James and his opinions of Mrs. Piper's integrity (quoted above), we are "willing now to stake as much money on Laurie Campbell's honesty as on that of anyone we know, and we are quite satisfied to leave our reputations for wisdom or folly, so far as human nature is concerned, to stand or fall by this declaration."

However, in science, the data ultimately stand or fall on the integrity of the experimenters. We mentioned in the methods section that we did not anticipate observing such an extraordinary research reading. Had we done so, we would have included additional procedures to establish the integrity of the experimental procedures (e.g. we would have had independent observers witness all of the procedures). The experimenters are well informed about the potential of deception, and we will report evidence of it when it is observed in the laboratory (Schwartz, Russek, and Nelson, 2001).

As described in Appendix B, the HESL is concerned about integrity in science in general, and mediumship science in particular. We carefully explain four types of integrity in the statement. Everyone involved with HESL is requested to read and sign our "integrity action pledge"; mediums are required to sign our "exceptional abilities participants" form.

[Note: One reviewer of a draft of this paper, Michael Shermer, questioned the scientific value of including reference to the use of an Integrity Action Pledge in our laboratory. He recommended that this fact be excluded from this report. One anonymous reviewer proposed that the information be deleted because "its presence only raises the reader's suspicion that the authors find adhering to these minimal standards a somehow noteworthy achievement." Another reviewer considered the information "unnecessary and pointless." However, we believe that in light of the unexpected and unusual nature of these findings, that the HELS's emphasis on integrity should be included in this particular paper. If the experimenters did not have integrity, for example, we would have excluded the fact that LC stayed in the experimenter's home to avoid misinterpretation of the experimenter's motivation. Since two anonymous reviewers raised questions about the integrity of both LC and GD, the fact that HELS employs explicit information about integrity seems worth noting. Of course, signing a piece of paper will not protect against "fraudsters," as one anonymous reviewer put it; however, it does put the fraudsters openly on notice.]

Possible Rater Bias and the Conservative Calculation of Conditional Probabilities
The specific information received during both the pre-reading contemplation condition (Phase 1) and sitter-silent condition (Phase 2) cannot be explained as due to rater bias in GD's scoring of the information.

Details regarding names and relationships are discrete, precise, and can be independently verified by living family and friends. The fact that GD has conducted independent research attempting to discover if his friend's consciousness continues (reported in Dalzell, 2001), and that GD's family and friends provide independent confirmation of the facts reported by LC, serves as essential cross-validation of the information received by LC, and appears to preclude telepathy and superpsi in some cases (see below).

In Schwartz et al (2001), the average accuracy of information scored by the sitters from the transcripts was 83% for the actual readings and 77% for the silent periods. These studies were conducted proximally (i.e. locally) - the medium and sitter were in the same room.

However, the present experiment was conducted "long distance" (i.e. nonlocally) - from Tucson, Arizona to Los Angeles, California, a distance of over 1000 miles. The accuracy of the specific information obtained during the actual reading (Phase 3) was above 90%. Moreover, the combined accuracy of the names and relationships obtained during the pre-reading contemplation (Phase 1) and sitter-silent (Phase 2) conditions was also above 90%.

It is important to recognize that the calculated conditional probability of the pattern of names and relationships correctly retrieved by LC (p less than one in 2.6 trillion by chance) underestimates the actual conditional probability. Since this information was received before LC knew anything about the sitter (LC was both "blind" and "deaf" to the sitter), the information retrieval cannot be attributed to cold reading or subtle cueing.

As mentioned in the results section, none of the six primary names received by LC during the pre-reading and sitter-silent periods in the context of their specific relationship to the sitter applied to either experimenter. Curiously, the three names received by LC without relationship information that GD rated as "0" (unknown and therefore possible errors) - Joyce, Elaine, and Joseph - were names of people close to the experimenters (and would have been scored highly had we been sitters).

It should be noted that if important information obtained during the actual reading (Phase 3) is added to the conditional probability from Phases 1 and 2, the p value increases in significance. For example, it is possible to estimate the probability of having a deceased "aunt with the name Alice" who in turn has a living "grand daughter with the name Katherine."

Following the procedure described in the Results section, LC indicated in Phase 3 that A was an aunt. Including the specific relationship adds 1 in 6 (rather than 1 in 12, because sex was already included in the Phase 1 and 2 estimate). The combined probability of G, M, J, B, T, S and A, so identified, is 30 * 180 * 180 * 30 * 30 * 100 * 30 * 6, or 1 in 15,746,400,000,000.

Granddaughter named K is 12 * 15 or 1 in 180 (p<.006). The combined probability of G, M, J, B, T, S, A, and K, so identified, is 30 * 180 * 180 * 30 * 30 * 100 * 30 * 6 * 180 or 1 in 2,834,352,000,000,000.

In addition, it is important to recognize that LC did not say that G, M, J, B, or T had a granddaughter named K. LC only said that A had a granddaughter K. Hence, the specification that granddaughter K belonged to A and not G, M, J, B or T can be estimated conservatively as 1 in 6. The combined probability of G, M, J, B, T, S, A, and K belonging to A is 30 * 180 * 180 * 30 * 30 * 100 * 30 * 6 * 180 * 6 or 1 in 17,006,112,000,000,000.

In other words, with the addition of aunt A with a grand daughter K, the conditional probability increases to 1 in 17,000 trillion.

[Note - most of the names not attached to specific people, be they common (e.g. John) or less common (e.g. Marcus or Maureen) were excluded from the calculation of the conditional probabilities (the B and A names were included because they were specifically invited by GD prior to the reading). We did not include additional names despite the fact that they could be seen as directly related to the specific sitter (e.g. Marcus is the name of one of M's dearest friends). Excluding names like John, Marcus and Maureen, for example, in the calculation of the conditional probability more than balances excluding the three probable errors (Joyce, Elaine, and Joseph) in the calculations.]

As mentioned in the Results section, the estimate of 1 in 15 for common names is quite conservative. Raising the estimate to 1 in 20 increases the conditional probability 5 * 5 * 5 * 5 * 5 * 5 for the six common names calculated, to the highly improbable value of 1 in 265,720,500,000,000,000,000, or 1 in 265 million trillion. If actual data are used to calculate the probabilities, the p values are even more significant. Appendix A presents data based on a sample of subjects from the University of Arizona (n=88) as well as a sample of subjects from the US Census Bureau (total n more than 6 million).

One curiosity is noted for the sake of completeness and integrity. In the actual reading (Phase 3), LC brought up the names of three deceased well-known scientists in context of GD: Albert Einstein, David Bohm, and Carl Jung. In LC's notes taken during the actual reading, their names were mentioned in the context of words such as "group," "inspire," science exploration," "work with Einstein," and "agreement."
In the process of reading a draft of this paper, GD informed us that in addition to the four primary individuals, he also invited "spirit scientists to help facilitate the experiment." The fact that LC mentioned Einstein, Bohm, and Jung in this particular reading (and she has never mentioned three senior scientists in the context of any previous research reading conducted over the past three years) is interesting in the context of GD's declaration.

GD recognizes that whereas most of the data can be readily confirmed by independent sources (e.g. by Michael's family or Jerry's friends), other information can not (e.g. that in his role as a research medium, he invited "spirit scientists to help facilitate the experiment"). To address the concerns of the most committed skeptics, GD indicated that he would be willing to undergo a professional lie detection test to help establish his personal integrity. LC and authors would be willing to do the same if it would help address the unfounded concerns of the most cynical critics of this research.

An Integrative Triune Approach to Anomalous Information Retrieval
Since fraud, selective memory on the part of the sitter, and sitter rating bias, have been essentially ruled out in this research, it becomes meaningful to consider possible alternative / anomalous / paranormal hypotheses.

There are three primary alternative (anomalous / paranormal) hypotheses that can account for the present findings (Gauld, 1983; Schwartz et al, 1999; Schwartz, Russek, Nelson, and Barentsen, 2001). We suggest that all three may be involved in anomalous information retrieval, and that they share a common dependence on info-energy systemic resonance relationships (described in Schwartz and Russek, 1999). Other hypotheses may also be possible, including novel hypotheses that have yet to be conceived (a position recommended by Michael Shermer).

The first hypothesis is telepathy with the living. The premise is that the medium is reading the conscious mind of the sitter (locally and / or nonlocally). Given the level of awareness and experience of the sitter in the present experiment, telepathy with the sitter needs to be seriously considered.

However, the inclusion of Phases 1 and 2 in this experiment, coupled with the observation that four pieces of specific information were not known to the sitter prior to the reading:
(1)the stone church along the river (for M),
(2)Aunt A's living grand daughter's crisis (K),
(3)the correct spelling of K's name (GD thought it was spelled with a C), and
(4)J's living on the east coast in Brooklyn, make the simple telepathy hypothesis insufficient to account for all the information retrieved by LC.

The second hypothesis is often termed "superpsi" (e.g. Braude 1992). One version of the super-psi hypothesis can be thought of as an extended unconscious telepathy / systemic resonance mechanism with everyone presently living (Schwartz and Russek, 1999). Simply stated, (1) LC resonates with the experimenters, (2) the experimenters resonate with the sitter (e.g. the senior author has spoken with GD numerous times and met him three times in person), (3) the sitter resonates with his living family members and friends (including family members and friends of M, A, B, and J), and (4) the information is retrieved unconsciously through systemic memory resonance (Schwartz and Russek, 1999). The network of dynamic info-energy relationships is accessed unconsciously by LC. The majority of the present findings are consistent with such a network memory resonance hypothesis.

Given that the telepathy hypothesis (1) and the network memory resonance hypothesis (2) (one example of a superpsi hypothesis, Braude 1992) are both plausible in principle, the question arises, do hypotheses 1 and 2 together account for all the data in this research reading? We suggest no.

Close examination of the languaging used by LC indicates that she is not simply reporting memories and images. LC is also reporting intentions and interpretations reflecting the information processing of "entities" (her words), or dynamically changing info-energy systems (our words, Schwartz and Russek, 1999).

When LC describes how M is interpreting future changes in GD's life, for example, the languaging not only implies that M is living, but the precise way the information is being organized is recognized by GD as reflecting M's mind and personality.

In other words, it is the specific intentional and organizing nature of the way the information is received by LC and reported to GD that suggests that LC is not simply reporting the conscious or unconscious memories of GD and his extended network of family and friends. We term hypothesis 3 the organizing consciousness (or soul) hypothesis (reflecting the fact that it is the precise details of the organizing nature of the information that implies the continued existence of an intentional, living consciousness (soul).

[Note: One reviewer suggested that LC might be "role-playing and impersonating discarnate entities. This speculation, in the abstract, is correct. However, in the concrete, given the highly blinded nature of Phases 1 and II, the role-playing hypothesis seems somewhat unlikely in LC's case.]

In view of the decades of substantial and replicated experimental research in parapsychology (e.g. meta-analyses reviewed in Radin, 1997), it is prudent to consider the possibility not only that all three hypotheses may be true, but that hypotheses 1 and 2 (telepathy and superpsi) may be prerequisite mechanisms for discovering hypothesis 3 (e.g. the organizing consciousness). Hypotheses 1 and 2 may be intimately involved in the discovery of hypothesis 3. In other words, we propose that hypotheses 1 and 2 may provide the mechanistic foundation that allows hypothesis 3 to be documented (e.g. mediums claim that they engage in "telepathy" with the deceased). Together, they reflect an integrative triune approach to dynamic anomalous information retrieval.

Implications for Future Research
Future research can exploit the power of the pre-reading contemplation (Campbell) and sitter-silent (Russek) procedures in both single blind and double-blind experiments. Stimulated by the present findings, we are currently using the Campbell procedure (Phase 1), combined with the Russek procedure (Phase 2), with sitters who do not receive an actual medium-sitter dialogue condition (i.e. no Phase 3). Hence, the mediums never hear the sitter's voices and the sitter's never hear the medium's voices. Information obtained from both the Campbell and Russek procedures are then mailed to the sitter's under blinded conditions. Transcripts from five sitters (the sitter's own transcript and another four sitter's transcripts, which serve as controls) are scored, item by item, by each sitter. If the sitters can accurately determine which transcripts belong to them, this establishes anomalous information retrieval under strict double-blind conditions.

In one ongoing double-blind experiment, five mediums are serving not only in the role of mediums, but they are also serving in the role of sitters. Double-blinding the Campbell and Russek procedures makes this kind of experimental design possible.

[Note: After this report was written, GD, in the role of a medium, attempted the LC pre-reading contemplation paradigm with a sitter. He reported receiving accurate information of name, cause of death, age of death, personal descriptions, that matched the primary person invited by the sitter. The approximate accuracy was 85%. GD is currently serving as one of the mediums and sitters in the above double-blind experiment]

Future research can determine if characteristics of the sitter matter. For example, when mediums and spiritually open people serve as sitters in a double-blind experiment, is more specific and detailed information retrieved than when disbelievers and skeptics serve as sitters?

Publishing White Crow Research Readings
It is likely that future findings will not be substantially greater (in terms of accuracy) than are the present findings. Clearly, future experiments can replicate the basic observations reported here and document that they be generalized to other mediums, with other sitters.
However, in the same way that Michael Jordan (arguably the greatest basketball player who ever lived) achieved only a few 60+ point games, it seems likely that "Michael Jordans of research mediumship" will only have a few 90+% percent accuracy research readings under controlled laboratory conditions.

It is hoped that this paper, focused on the findings of a William James "white crow" research reading, will encourage other investigators to explore the potential of these experimental procedures and replicate the findings. Future papers using these procedures, replicating evidence of white crow research readings, can help firmly establish the existence of anomalous information retrieval, and potentially the continuance of consciousness.

Braude, S. E. (1992). Survival or super-psi? Journal of Scientific Exploration. 6,2:127-144.

Dalzell, G.E. (2001 in press). Messages: Evidence of Life After Death. Charlottesville, VA: Hampton Roads Publishing.

Radin, D. (1997). The Conscious Universe: The Scientific Truth of Psychic Phenomena. San Francisco, CA: HarperCollins.

Russek, L.G.S., Schwartz, G.E.R., Russek, E. and Russek, H.I. (hyp) (1999). A possible approach for researching purported spirit communication: An empirical-anecdotal investigation. Advances in Mind-Body Medicine. 15,4: 295-301.

Schwartz, G.E.R. and Russek, L.G.S. (1999). The Living Energy Universe: A Fundamental Discovery that Transforms Science and Medicine.  Charlottesville, VA: Hampton Roads Publishing.

Schwartz, G.E.R., Russek, L.G.S., et al. (1999). Potential Medium to Departed to Medium Communication of Pictorial Information: Exploratory Evidence Consistent with Psi and Survival of Consciousness. The Noetics Journal. 2,3: 283-294.

Schwartz, G.E.R., Russek, L.G.S., Nelson, L.A. and Barentsen, C. (2001 in press). Accuracy and Replicability of Anomalous After-Death Communication Across Highly Skilled Mediums. The Journal of the Society for Psychical Research.

Schwartz, G.E.R., Russek, L.G.S., and Nelson, L.A. (2001, submitted for publication). Purported anomalous perception in a highly skilled individual: Observations, interpretations, compassion.

Schwartz, G. E. R., Russek, L.G.S., and Barentsen, C. (2001, submitted for publication).


Gary E. Schwartz, Ph.D.
All Rights Reserved.

 Site Design By: Online In Olalla