by Tom Butler, 2009
Also see listening panel trials on radio-sweep examples
A Research Study into the Interpretation of EVP
Radio-sweep technology, popularly known as “ghost boxes” or “spirit boxes,” is examined as a technology used for recording Electronic Voice Phenomena (EVP). The results of a session reported in the ATransC Idea Exchange were used for a blind, online listening test similar to previous tests reported in the online ATransC Journal as EVP online listening trials. The generally negative results are reviewed and reasons why the technology may not be suited for trans-etheric communication are discussed.
“Radio-sweep” is a technology that involves rapidly changing the tuning of a radio receiver to produce a sound track composed of bits of sound from whatever radio programming is on the air and from whatever radio station is detected by the radio at the time. In theory, the communicating entity somehow arranges for the radio programming of local stations to be producing the required sounds at the moment they are required and that the sweep will detect those sounds at the right moment to produce the desired message.
Radio-sweep technology, popularly known as “ghost boxes” or “spirit boxes,” has become a popular technology represented by its advocates as a way to record EVP. It can be accomplished by manually tuning a radio, but a number of modified radio receiver devices are now being sold as EVP recording devices ranging from a few hundred dollars to over $1,200. A survey of the literature produced by manufacturers indicates that there have been no controlled studies of this technique to establish that it actually produces EVP.
As part of the Association TransCommunication mission to provide guidance to members about trans-etheric phenomena, this technology was examined to evaluate its capability of producing EVP, how it might do this and whether or not it can improve understanding of trans-etheric communication. There have also been frequent complaints that examples of radio-sweep results did not seem to actually contain intelligent information. At the same time, many members have reported great success with the technology, and this dichotomy required that such an evaluation included an examination of our current assumptions about EVP formation.
A companion article, EVP Formation, describes how EVP are thought to be formed and addresses current understanding of how EVP is heard and reported.
Online Listening Test
A study of radio-sweep was conducted using an example considered typical of the technology. This example was posted in the AA-EVP Idea Exchange with the comment:
“I used a Mini-Box and heard”:
Reported EVP: “Big Circle.”
“I asked: ‘Is the Big Circle there?’”
Reported EVP: “Circle, Big.”
Reported EVP: “Is it —-?”
Reported EVP: “Is it?”
Reported EVP: “Might be!”
“Let me know what you hear. I only cut out bits of silence and my first comment to make it fit.”
This example was obtained using one of the Mini-Box radio-sweep devices sold by the apparently defunct Paranormal Systems for $300 (as of early 2009). The manufacturer describes it as “…a useful tool and a new way to establish spirit communications.” The example for analysis was selected because eight of eight members commenting in the thread stated that they heard the example as it was reported.
With the exception of “is it,” which is a clearly enunciated phrase, I was unable to hear the examples as reported. To assure that it was not just my inability to make out the reported message, I broke the example into the same segments reported by the practitioner and posted them on ATransC.org as a new listening test. They were labeled as “Example 1” (through 5) and an unlabeled text field was provided for the website visitor to indicate what was heard. This same procedure has been used for previous listening tests resulting in average correct word recognition of 25.2%. See: EVP Online Listening Trials
The test was stopped after forty-one entries were received because a decisive outcome had been obtained. The results were:
Example 1: “Big Circle” — Zero recognized words (%Rw = 0.0%). Common response were “This is Butler,” “puffin” and “buckle.”
Example 2: “Circle, Big” — Zero recognized words (%Rw = 0.0%).
Example 3: “Is it —-?” —Ten of a possible 123 words were reported for %Rw = 8.13%. “It” was reported, but in many different contexts other than what was expected.
Example 4: “Is it?” —Forty-one of a possible eight-two words were reported for %Rw = 50.0%.
Example 5: “Might be!” — Zero recognized words, %Rw = 0.0%. Commonly reported words were “Hi,” “I’m” and “Spring.”
Here is the original sound track.
- Examples 1-3 and 5 are mostly sound fragments that would most likely be reported as artifact noise if found in a digital recorder.
- Example 4, “Is it,” is composed of two clearly spoken words, and its high %Rw indicates that the listening test works. If such a clearly spoken example did not have a high %Rw, then it would be necessary to question the validity of the test.
- The “Is it” segment is a case of a randomly, but naturally occurring sound segment. Story telling is then used to make it seem part of a meaningful response.
- The use of short examples has been questioned; however, in the other trials a one-word example scored the lowest while two-word examples did overall as well or better than the three or more word examples. The previous trials indicate that, if there are recognizable words present, then there should be at least a few correctly reported words for each example. See: EVP online listening trials
Radio-Sweep Audio Output
Potential voice and voice-like sounds in radio-sweep includes:
- Chaotic sounds that are inappropriately given meaning (sometimes known as Pareidolia).
- Clearly spoken words that a practitioner incorporates into a story about the message that is meaningful.
- Sounds that invoke meaningful impressions in the practitioner, which are then explained as messages.
- Transform EVP formed from the noise produced by the sweep.
Altered Perception and Story Telling
In EVP Formation, a companion article intended to explore how EVP are formed, it is noted that there are a number of ways mundane sounds are mistaken as EVP. The most common way follows the process:
- The practitioner asks for information during the recording.
- Sounds are heard, either live or on the resulting recording.
- The practitioner “hears” what is expected in the sounds.
- The practitioner reports what was “heard” and listeners hear what is suggested.
This is not malicious intent, but a natural response to trying very hard to find a particular kind of information in a chaotic signal. This appears to be especially common in if the chaotic sound has a staccato pace, as we have seen the effect in both radio-sweep and EVPmaker output.
An interesting explanation as to how practitioners and listeners might find EVP where there are none is found in the Gestalt Laws of Perceptual Organization, which include:
The Law of Proximity: Stimulus elements that are close together tend to be perceived as a group.
The Law of Similarity: Similar stimuli tend to be grouped; this tendency can even dominate grouping due to proximity.
The Law of Closure: Stimuli tend to be grouped into complete figures.
The Law of Good Continuation: Stimuli tend to be grouped as to minimize change or discontinuity.
The Law of Symmetry: Regions bound by symmetrical borders tend to be perceived as coherent figures.
The Law of Simplicity: Ambiguous stimuli tend to be resolved in favor of the simplest.
A reasonable conclusion is that the practitioner heard what was expected. “Big Circle” is an important part of ATransC culture, and hearing this term after asking for someone in the Big Circle to comment is natural, especially considering the low quality of the sound file. The next step would be to imagine a story that would allow what was thought by the practitioner to have been said to make sense. Next, the listeners simply conform by hearing what they are told is present in the recording.
Either the actual sound file had the reported utterances (except for Example 4, it did not), or if not, the practitioner may have intuitively sensed the response. By this, I mean that the radio-sweep output could be used as a technology for divination much as other intuitive aids such as Tarot cards or tea leaves. When Tarot cards are laid out for a reading, the practitioner has an array of visual/intellectual cues that can be used to develop a story; but the meaningfulness of the story is largely the result of the practitioner’s intuitive ability. In the same way, a radio-sweep sound file contains audible cues from which a story may be developed, but the meaningfulness of the story would be largely the result of the practitioner’s intuitive ability. In effect, the practitioner becomes an oracle intuitively reading the radio-sweep output.
It should be noted that this observation is not intended to detract from the practitioner’s ability. Other research has clearly shown that various forms of mediumship and/or intuitive sensing are valid techniques for trans-etheric information access. It is not my intention to say that information reported by radio-sweep practitioners is not meaningful or accurate. Methods of evaluating the information content, other than those used in this study, must be used for such a determination.
Radio-Sweep as a Source of Noise for Transform EVP
As discussed in the article, EVP formation, the traditional method for EVP is the recording of the phenomenal utterances by transforming available audio-frequency noise into voice. In fact, it has been shown that virtually any noise is apt to be transformed into voice. The primary output from radio-sweep is noise, and as can be expected, it is common to find examples of transform EVP in the output sound file.
The presence of transform EVP in radio-sweep output is a confounding problem for the evaluation of the technology. Radio-sweep can produce EVP which results in meaningful information; however, the evidence indicates that, when transform EVP is produced using radio-sweep, that technology is being used as a novel way to produce noise for ordinary EVP formation. The radio-sweep output does not appear to be phenomenal in itself.
Transform EVP Formation and Physical Mediumship
To compound the problem of evaluating the veracity of radio-sweep for EVP, it has been noted that some practitioners do produce EVP using the swept dial of a radio as a sound source. The rarity of such practitioners suggests that other processes are involved.
Recent observations indicate that the ideal audio-frequency energy for transform EVP formation is both chaotic favoring human voice frequencies (200 to 4000 Hz) and with many short transients. For instance, a recorder with a lot of noise but without a lot of amplitude changes is not as effective as a recorder with noise that has many perturbations in the noise heard as clicks, pops and very short (stuttering-like interruptions in the noise. As it turns out a very rapidly scanned radio spectrum often produces such noise. For example, the radio-sweep results we have heard reported by some practitioners as EVP, and that do appear to be EVP, have been produced using a manual sweep on a radio with a round tuning dial. Rapidly turning a small tuning dial from stop to stop (probably half a second) results in a sufficiently short “dwell time” on individual stations that only bits of voice are heard, much as if a phoneme file was being used instead of radio-sweep.
EVP produced by radio-sweep should be formed of many voices and music components, yet in the meaningful examples produced by some practitioners, the voice is typically all one person speaking for the entire sweep. This is what would be expected for transform EVP using the radio-sweep noise as a sound source.
Direct Radio Voice (DRV) such as that produced by Marcello Bacci and Anabela Cardoso, meaningful messages are produced from radio broadcasts that are thought to have an etheric origin. This is thought to be a form of physical mediumship produced by Bacci and Cardoso using a radio as a sort of high-tech séance trumpet. This is a very rare form of phenomenon that may also be produced by some EVP practitioners.
In other words, some practitioners appear to produce meaningful and reliable EVP using radio-sweep technology. However, once again, the radio-sweep output does not appear to be phenomenal in itself.
Violation of Self-determination
While the idea that we have self-determination or free will is faith-based, it does raise an important question. I am not aware of any instances in which we have been forced to do something by our etheric communicators. In fact, there are many examples in which they seek to protect us. For radio-sweep to be a viable technique for EVP, it seems necessary that programming is exactly as required for the intended message. That implies that radio announcers are forced to speak words that are required for the message. If this is the case, then it is a clear violation of our self-determination. In effect, the radio announcer is forced to say “Hello Tom” at the exact moment a practitioner sweeps the dial past that station if the intended utterance is “Hello Tom.”
Why did eight of eight listeners on the discussion board report hearing what the examples were reported to have said while online listeners did not? Perhaps the suggestion of what will be heard in not so clear sound is all that is needed to entrain the mind of the listener to hear exactly that. This tendency to hear what is suggested is most evident with examples that are of very poor quality. EVPmaker using live voice and radio-sweep examples have such a confusing, staccato pace that they tend to confound the mind, making it difficult to “lock onto” the actual sound stream. The result is that the listener may be forced to depend on instructions for what is to be heard.
The three techniques that have been decisively shown to produce EVP are audio recorder using noise (transform EVP), EVPmaker using allophones and speech synthesis. All three depend on available physical energy and processes for voice formation. This is discussed in the article, EVP Formation. Radio-sweep depends on the availability of the right sound being present at the exact moment the sweep selects that station. In fact, the entities appear to use most efficient methods for communication and do not routinely make people do things for the sake of communication. We are aware of no precedent indicating that EVP have been formed by first creating physical energy and/or causing physical processes. The only trans-etheric influence we have seen evidence for appears to manifest as the subtle energy usually described as “psi energy.” The processes most commonly influenced by psi energy are random, and in EVP, this is seen as the influence of random noise. There is no empirically demonstrated evidence we are aware indicating the entities are able to cause someone to do something in order to communicate via EVP.
It is important to note that when evaluating radio-sweep, it has been demonstrated that the noise produced by the sweep process is sometimes used for transform EVP. As such, it is possible to find a few words formed from the noise, but in this mode, radio-sweep is just an expensive way of producing noise for voice formation.
We have been examining radio-sweep since an ATransC member began working with it years ago. While we have not been able to find a reason to think the technology produces EVP, we have found substantial reason to think it does not. Certainly one cannot permanently close the door on any technology, but until properly designed research produces empirical evidence that radio-sweep produces EVP, our policy must be that radio-sweep does not produce EVP as advertised.