by Cindy Heinen
Also see: The article, Gathering Information Using Research Project Gathering Information Using EVPMaker With Allophones
This twelve-month trial was designed to determine whether or not information not known to a participant could be requested and received via Electronic Voice Phenomena (EVP) using EVPmaker with allophones.
A target object was left undisturbed in the same location at the beginning of each month for twelve months. Participants were asked to use only EVPmaker with allophones to produce a sound file containing the information identifying the target. To qualify submissions the project manager screened them for reasonableness. Those possibly containing usable information were submitted to a listening panel of people who were familiar with EVP but who did not know the identity of the target. If a majority of the listening panel heard information as reported by the participant, the submission was considered a valid submission. Submissions were rejected if a majority did not hear what the participant heard. Valid submissions were screened for a positive response by the project manager for inclusion in the study.
This trial did not produce positive responses based on the study’s protocol. However, the abundance of non-protocol EVP captured in the study might suggest EVPmaker is not suited for the type of communication this study was designed to capture. Also, participant knowledge of the target might have to be reexamined as several targets were identified either before the target was placed or after when the participant knew what the target was. Future studies may wish to look at these non-protocol results when designing a follow up study.
Background sound is often used as sound energy during the recording of Electronic Voice Phenomena (EVP). This sound may be ambient noise in the environment or sound purposely added to the recording environment such as the hiss of a radio tuned off station, flowing water or the sound of a fan. These imported sounds are said to supply the sound energy entities need to form voices.
While these sound sources can result in EVP, some practitioners proposed that the sounds or frequencies of the human voice would be optimal for the communicating entities to use to generate a voice. Different methods of experimentation using human vocalizations have been tried including foreign language recordings, some of which were edited and clipped to produce random bits of vocalizations with no discernible words or syllables. Radio-sweep using modified radios popularly known as “ghost boxes” or “spirit boxes,” has also been tried. This involves rapidly changing the tuning dial of a radio to produce pieces of speech.
In the late 1980s, Stefan Bion developed a computer program that he likened to a sound grinder. The software program, called EVPmaker,1 uses a random process to select segments of a buffer in which a raw, audio input file has been loaded. The resulting output file is a stream of randomly arranged short sound segments from the input file. EVP are thought to be formed via manipulation of the random selection process.
In 2008 Stefan Bion offered a file containing seventy-two allophones generated by SpeakJet™ 2 that could be used as the input audio file for EVPmaker. The output of these randomized allophones from EVPmaker, while robotic sounding, were used successfully by some researchers to obtain what they felt were meaningful and relevant communications.
Because of the standardization provided by using EVPmaker with the allophone file provided by Stefan Bion, the program was a good choice to use in a controlled study of EVP. All participants, individuals who attempted to record EVP communication for the study, would be using the SpeakJet™ Allophones2 as the sound source to input into EVPmaker.
The study was designed to ask each communicating entity to perform a task that would indicate understanding and cooperation. This task was to view a specific object that was set out at a specific location. The communicating entity was then instructed to tell the participant what that object was through the participant’s copy of EVPmaker with allophones.
Study question: Can the identity of a target object be recorded in the EVPmaker output file which is of sufficient quality that a listening panel is able to agree on its content?
- A target object unknown to participants was placed in the same location at the beginning of each month for twelve months. Sufficient information about the location to uniquely identify it in the world was announced.
- Participants, who responded to the public announcement, conducted EVPmaker sessions using the allophone file provided by Stefan Bion, but otherwise using any EVPmaker setting. They asked that the target be identified in the output file. There was no limit on the number of attempts.
- Knowing only that the target was an object, participants listened to the output files and determined if a response had been recorded. Whether or not it was a likely response concerning the target object was up to the judgment of the participant.
- Audio files the participant determined could include names of the target were sent to the project manager.
- The project manager compared what the participant heard to the target. If the project manager agreed that the target was indicated in the file, it was sent to a listening panel.
- Not knowing the target object, the listening panel individually reviewed the file and noted what if anything was said in the file. That information was sent to the project manager.
- If a majority of the listening panel heard information as reported by the participant, the submission was considered a valid submission; if not, it was rejected. Valid submissions were screened for a positive response by the project manager for inclusion in the study.
The people who made up the study team were the project manager, the target keeper, three listening-panel members, two alternate listening-panel members and the participants.
- The project manager developed the trial protocol, oversaw the project, communicated with the target keeper, listening-panel members and the participants, analyzed the data and wrote the monthly and final reports.
- The target keeper was in charge of putting a new target into place every month.
- The listening-panel members would review any audio files sent to them by the project manager.
- The alternate listening-panel members would be enlisted to review audio files if one of the original three listening-panel members could not participate.
- Participants were anyone who wanted to take part in this trial and could follow the study’s protocol.
The target for each month was taken from a list that had been prepared prior to the start of the yearlong study. The target objects for the year were only known to the project manager and target keeper. The target keeper would find objects that she felt fit the predetermined list of targets. She would then submit a picture and a brief explanation of the object to the project manager. On the first of the month she would place only this object on a specific shelf on a shelving unit in her home. The object would remain there undisturbed for the entire month. This target object would be taken away and replaced with the next month’s target on the first of the following month.
The following list shows what was requested and what the target keeper put into place for each month. A synopsis of her comments concerning the targets has also been included.
May: An abalone dragonfly pin designed by her husband.
June: One of the target keeper’s recorders.
July: Pink roses in an engraved black vase with “4546 B INDIA” on the underside of the vase.
August: A blue teacup with painted flowers on it. “JAPAN” was stamped on the bottom of the teacup.
September: Zephyr scissors with “CLAUSS NO 78 USA” on one blade. They had blue handles with white paint and black gunk on them. The target keeper noted that she has had them for twenty-five years and thinks of them as “our work scissors.”
October: I Am That, by Sri Nisargadatta Maharaj. It is not a “holy book” in the traditional sense, but the teachings within are certainly considered to be sacred by many.
November: Candle made for the target keeper by her son while in elementary school. It has layers of blue, yellow, red and orange wax poured into a baby food jar.
December: The target keeper’s favorite radio-sweep radio, a Duracell KP028 crank flashlight/radio, also referred to as a “3-in-1.”
January: A child’s pair of BluBlocker sunglasses that belonged to the target keeper’s son when he was little. The word “Taiwan” is printed in white on the inner right arm.
February: A peacock feather given as a cat toy by the pet store.
March: A red, flexible bracelet with the words, “Stronger Now.” The bracelet was bought from two girls who started ARCHFoundation.com.
April: A hammer with very faded print on both sides. One side reads “Diamond Island” and the other reads “Burro Cigars.”
The pictures of the targets (right) and what the project manager was told about them suggested that other target words could be applied to a target. Responses that were considered acceptable for this study had to be words or phrases that contained a word that could be used to identify the object. For example, in March the target object was a bracelet. The bracelet itself was a red, flexible bracelet of the type often used as a charity fundraiser. The best word for the target would be bracelet, but band or wristband would also be acceptable. Words that would be considered adjectives for the bracelet like red or flexible would be noted but not considered a positive response.
Participants for this trial were required to use only EVPmaker with the SpeakJet™ allophone file provided on Stefan Bion’s EVPmaker website. They were allowed to record their sessions using recording equipment of their choice. Files thought to contain information about the target were to be emailed to the project manager, along with text of what the participant believed was said. Inclusion of the practitioner’s voice was considered a valuable plus. Altering the audio file through filtering or noise reduction was not allowed.
Participants did not know what target objects were other than that they could fit on a standard-size wall unit shelf. They also knew the general location of the shelf that held the object. This was at the target keeper’s home in California on the middle shelf on the right-hand side of the wall unit.
After a recording session, each participant was to analyze and interpret their recordings while listening for a reply to the question: “What is the target object on the target keeper’s shelf?” If they heard something they felt might be the target, they would submit an email stating what they heard along with the audio attachment to the project manager.
EVP from EVPmaker is considered an opportunistic form,3 meaning the message is formed from available sound segments in the buffer. But output may also be used as noise for transform EVP 3 which are voices formed from background noise. Participants could submit either type of EVP for this study. If the participant’s interpretation of their submission identified that month’s target object, the project manager would send the files to the listening panel.
A listening panel was organized to review any files that might contain words that indicated the target for that month. It consisted of three primary and two standby individuals who had been enlisted before the study. The listening-panel members did not know what the target objects were and did not talk to anyone about their analyses. After reviewing the audio files, listening-panel members sent their interpretation of the files to the project manager. The project manager then compared the listening panel’s interpretations with that of the participants who sent the files. If two out of three of the listening-panel members heard something phonetically similar to what the participants heard, the submission would be considered a positive response.
In the twelve months of the trial, 648 audio files were submitted. In August, two submissions matched the target object, and in December, three submissions were a match. These submissions were sent to the listening panels but none met the protocol which stated that the target the participant heard must also be heard by at least two out of three of the listening-panel members to be considered a positive response.
Although not what this trial was designed to find, some interesting things did surface during the year of the trial:
- There were seven possibly positive responses that were recorded in September but that appeared to refer to the October target object which was a holy book. The target keeper chose I AM THAT by Sri Nisargadatta Maharaj for October. This book has a black and orange cover and the back of the book is orange. In September, the project manager received seven submissions from three participants that indicated a book. They were: “red book,” “green book,” “bible,” “Book is the key,” “scripture,” “a certified orange book(let)” and “popular book.” The listening panel reviewed all the files and out of the seven, three were identified. They were:
- Book is the key.
- Certified orange book
- Popular book
While these could not be counted as positive responses for the trial, the project manager felt the results should be noted for a number of reasons. First, up to this point in the study, this was the first time that a specific word, “book,” was repeated frequently and was submitted by several participants. Second, some of the references to the target, while not validated by the listening panel, did seem to point towards a holy or spiritual book.
Finally, the project managers and the target keeper both knew what the target was. But in addition to this, both had a connection to the target being prepared for October. The target keeper placed a book she felt connected to because of the book’s wisdom. The project manager, when developing the target list, felt that a holy book would be the one target that would receive the most positive results. She was anticipating the October placement of the book throughout September. While one cannot rule out coincidence for these results, could the knowledge of what the target was and the intentions of the project manager and the target keeper have played a part in the results?
- Group intentions and how they affect the outcome of an event is being studied by scientists. The Institute of Noetic Sciences’, “The Effects of Distant Intentions on Water Crystal Formation: A Triple Blind Replication”4 results suggested that ice crystals formed from water exposed to distant intentions were more aesthetically pleasing than ice crystals formed from water from a control group. And the majority of Lynne McTaggart’s intention experiments5 have suggested that intention can affect plants and human project outcomes. EVP and participants’ intentions might also lend themselves to some interesting studies.
- As documented in the chart, three submissions were accepted by the project manager for the December target, which was a radio favored by the target keeper for radio-sweep. These submissions were “Her radio,” “The answer … radio,” and “Transistor.” The project manager accepted “transistor” because small, handheld radios were often in the past referred to as “transistors.” Knowing what to listen for, the project manager could identify these responses but the listening panel, who did not know what to listen for, did not hear what the participants reported hearing. The question arises that, if the listening panel knew what the participants heard, could they have picked it out of the audio file?
How much does confirmation bias affect the analysis of skilled listeners? There were instances after the month’s target was revealed that participants reviewed audio they recorded during the month of that target and subsequently reported to the the project manager that they found audio that matched the target. The effect of knowledge about the target during the recording process needs to be explored and possibly reconsidered.
- Another common result was the recordings of what participants felt were communicating-entity comments. Most of these comments were normal EVPmaker random selection EVP. Most were quite clear and many commented on the communicating entity’s ability or lack of ability to be able to help with the trial. Also on a monthly basis, there were submissions that were comments directed, by name, to the participant or to the project manager. Some submissions were negative in nature and a few did not make any sense. While several participants submitted these “comment EVP,” the majority were submitted by one participant.
Listening Panel Challenges
Throughout the trial, the listening panel was sent audio files to listen to and analyze. None of the listening-panel members knew what the targets were. Often, what the listening panel reported hearing was not phonetically close to what the participant heard. They also seldom agreed with one another as to what the same audio file was saying, if they heard anything at all. This may indicate the challenges one faces when trying to understand and analyze very short audio clips from EVPmaker using allophones.
The clipped, robotic sound of the short audio files, along with having no word cues, may make it difficult for listeners to find any recognizable words. It might prove useful to submit longer audio files to the listening panel so they could hear the participant’s voice asking the question, then have the communicating entity’s reply. However, this has its own sets of challenges as the communicating entity did not always come in after the question was asked. The entity responses and comments could come in anytime during the recording sessions.
Words created through EVPmaker, while having proper vowel sounds, may have dropped consonants, for example, the word “hammer” might be heard as “ammer.”6 Since EVPmaker voices seem to create modified words it might be useful to train the listening panel to be able to identify these “new words.”
Number of Participant Challenges
Several things need to be noted that may have had an effect on the results. While 648 submissions were sent to this study, the number of participants was low. There were never more than four participants per month and often there were only two or three. Also, from August 2010 until the end of the study, the bulk of the submissions were from only one individual. So while the sample size was respectable, the number of different participants was very small. The study was designed as a group project. Having a greater number of participants could likely have changed the results. The reason few people participated may have been because trying to find an unknown target in a sea of vocalization fragments is a difficult task.
Another point to note is the rigidity of the trial’s protocol itself. It is often apparent in the recording of EVP that what works for one person does not necessarily work for others. Also, successful techniques have been known to sometimes show a decline in productivity for no apparent reason.7 All these points should be addressed in the development of follow-up experiments.
Based on the files submitted by a small number of participants, the results indicate that gathering specific information (a target) that was unknown to the participants was not accomplished using EVPmaker with allophones and following the protocol for this study.
Although not a part of this study, it does appear that “comment EVP” that identified the target was frequently recorded. These results suggest that following this trial’s protocol, EVP could be recorded using EVPmaker with allophones but specific informational EVP was infrequent. The reason for this remains unknown.
The “holy book” results in September might offer a nudge towards what else might be needed to obtain information-gathering communication. An experiment could be designed to compare the number of targets identified between participants who know what the target is and those who do not. A similar study could be done to compare the results of participants who spend time having focused intentions towards a target and those who don’t. A participant’s knowledge of the target might also play into this idea about intention.
The development of such intention experiments would need to take more into account than the technical aspects of EVP communication. It also would have to explore if consciousness; intent and attitude play a part in EVP communications.
Finally, knowing what to listen for and how to interpret the speech coming through EVPmaker might need to be addressed. A tutorial for participants and listening-panel members on the communicating entity’s unique formation of speech in EVPmaker might change the outcome of any future EVPmaker experiments.
The author wishes to express her appreciation and thanks to Tom Butler for his guidance during the course of this trial and assistance in the preparation of this paper. Also the author wishes to express her gratitude to the individuals who assisted in this trial including the persons who took on the positions of target keeper, listening-panel members and all those who were participants in the study.
- Bion, Stephan. EVPmaker. tonbandstimmen.de/evpmaker/index_e.htm.
- SpeakJet,™ speakjet.com, Magnevation LLC.
- Butler, Tom. “Locating EVP Formation and Detecting False Positives,” atransc.org/locating-false-positives/, 2010.
- Radin, Dean, Nancy Lund, Masaru Emoto, and Takashige Kizu. “The Effects of Distant Intentions on Water Crystal Formation: A Triple Blind Replication” Petaluma, Institute of Noetic Sciences, Journal of Scientific Exploration, 22, No. 4, pp. 481–493, 2008.deanradin.com/papers/emotoIIproof.pdf
- McTaggart, Lynne. The Intention Experiment: Using Your Thoughts to Change Your Life and the World, Free Press, 2007. ISBN-10: 0743276957, ISBN-13: 978-0743276955. theintentionexperiment.com
- Butler, Tom. “EVPmaker with Allophones: Where are We Now?” atransc.org/evpmaker-study-where-are-we-now/, 2011
- Butler, Tom and Lisa Butler. There is No Death and There Are No Dead. Reno: AA-EVP Publishing, 2004, atransc.org/books-atransc/