Rorschach Audio – art and illusion for sound

The August 2015 issue of Harper's Bazaar magazine carries a two-page feature entitled 'On the discovery of hope' by the magazine's editor-in-chief Justine Picardie – a writer who also found fame and influence as a journalist and the best-selling biographer of Coco Chanel. The article came shortly after last summer's re-issue of Justine's earlier book If the Spirit Moves You, which recalls, in moving detail, her experience of bereavement after the loss of a close family member. 'On the discovery of hope' is a thoughtful and well-written meditation on the experiences that were described in Justine's earlier work, but what's perhaps most striking is the fact that not once does this article mention the huge emphasis that If the Spirit Moves You places on its author's involvement with spiritualist mediums and with so-called electronic voice phenomena (EVP) recording.

The Oxford Companion to the Mind states that during bereavement 'sights and sounds are commonly misperceived as evidence of (the) return' of the deceased, and, in the case of EVP research, an entire belief system has been built around the conviction that it is possible, indeed quite easy, to literally record the voices of ghosts.

Now, an informal consensus does seem to exist within much of the scientific community that it doesn't really do to engage with or to critique beliefs like EVP. An obvious response would be to point out that the core beliefs espoused by EVP practitioners are essentially no more irrational than many other religious beliefs. However, what makes EVP different is the fact that it constitutes an adaptation of essentially primal superstitions to modern technological society. In that context, many EVP enthusiasts can and do proactively promote their findings as being 'objective' and 'scientific'.

In that sense it can be argued that EVP actively demands a response, in order to inform those potentially attracted to EVP exactly why it is that this research is not even remotely scientific. An informed critique of EVP can also be productive for another important reason. Since it can be shown that a whole battery of techniques employed by EVP researchers constitute what amounts to a comprehensive guided tour of certain psychoacoustic phenomena, so debunking EVP provides a compelling narrative structure within which to explain a whole raft of phenomena associated with illusions of sound, with cognitive science and with psychology of perception.

Mechanisms of misperception

I published a critique of EVP research, under the title 'Rorschach audio', as part of the sleeve-notes for a CD of 'classic'' EVP recordings called The Ghost Orchid, which was released by the record company Ash International in 1999. Since then there have been a number of updated 'Rorschach audio' articles, notably a paper that was peer-reviewed by leading academics and published by MIT Press in 2001. Lecture-demonstrations have been held at (among many others) the Science Museum Dana Centre, The British Library, the Oxford University Museum of Natural History and the Freud Museum (as part of the 'Festival of the Unconscious'). In 2007 I received funding to pursue a five-year long Rorschach Audio research project, hosted by Goldsmiths College and the University of Westminster, which led to the publication of the book Rorschach Audio – Art & Illusion for Sound.

The primary metaphor offered by the Rorschach Audio project is a comparison between processes of hearing, and the way in which viewers perceive meaningful imagery in the ambiguous visual figures of the psychoanalyst Hermann Rorschach's famous inkblot tests. He popularised the technique, previously known as klecksography, whereby sheets of paper are folded, opened again, splashed with ink, closed and pressed flat, and then opened, to reveal the iconic symmetrical figures, which are often perceived as resembling meaningful images. Rorschach ink blots were developed for use as a psycho-diagnostic test; and, in examples described by the psychologist Henry Gleitman, symmetrical ink blots resemble 'a head blowing smoke', but also a 'ghost' and an 'angel', those last two being directly equivalent sound imagery typical of EVP.

In the audio equivalent, EVP converts operate in a way that maximises the potential for ambiguous stimuli to be interpreted as meaningful. Badly shielded audio recording devices pick up forms of interference that are well-known to be produced in low-quality audio systems, by stray transmissions from real radio stations, from local taxi cabs, emergency services, intercoms, air-band radio and various sources of electrical interference, and so on. EVP researchers call out to deceased relatives and friends, in the style of a spiritualist séance, then trawl through often very long recordings, in search of the infrequent, abrupt and distorted sounds that manifest in 'response' to questions often asked much earlier. EVP researchers judiciously edit out the sections in which the ghosts didn't seem to have been talkative, then play recordings sometimes speeded up, slowed down, or backwards, but almost always repeated over and over, until they hear those sounds start to resemble personally meaningful messages. After one series of EVP recordings, Justine Picardie says: 'I don't even understand how a telephone works', so presuming she might not know much about radio and audio recording technology either, that suggests a case for improved public understanding of the relevant physics.

A blank canvas

The Rorschach Audio project was inspired by an extraordinary recording, which is frankly a masterpiece of art and science, prepared by the University of San Diego psychologist Diana Deutsch in 1995 (see tinyurl.com/oj4mf2j). Deutsch's recording consists of a narrator repeating two short neutral phrases. These are edited so as to rapidly alternate between stereo loudspeakers, in such a way as they initially come across as completely garbled and without apparent meaning. As described by Shaun Carlson in Scientific American (see tinyurl.com/oa54r9q):within a few seconds of listening to this strange cacophony, my brain started imposing a shifting order over the chaos as I began hearing distinct words... First came 'blank, blank, blank'. Then 'time, time, time'. Then 'no time', 'long pine' and 'any time'. I was then astonished to hear a man's voice spilling out of the right speaker only. In a distinct Australian accent it said 'take me, take me, take me'.

Deutsch's recording provides a convincing demonstration of the process by which ambiguous sounds can provide a 'blank canvas', onto which, with repeated listening, the brain can impose sometimes vivid illusions of audible meaning. This is how EVP listeners can perceive distorted audio snippets as though those sounds were personal messages.

The Rorschach Audio project makes no bones about the fact that the explanatory metaphor it posits is little more than a statement of the obvious. So, while the project acknowledges direct inspiration from Deutsch's recording, it also acknowledges that similar experiments were conducted by the psychologist B.F. Skinner. Similar perceptions have been heard emerging from other repetitive sounds – for instance sounds of steam trains, mill wheels and church bells. The latter gave rise to the famous folk-poem 'Oranges and Lemons', and was also discussed long ago by the artist Leonardo da Vinci. In terms of perceptual creativity, related imagery formed the basis of sound designs employed by the film-maker Jean Cocteau, and similar repetitions were behind a compositional technique employed by the writer Raymond Roussel.

In addition to exploring the (surprisingly rich) cultural history of these kinds of mishearings and misperceptions, the project also explores further aspects of psychoacoustics that relate to EVP. So, the mind's ability to actively project meaning onto ambiguous sense-data, as demonstrated by the Deutsch recording, is also discussed in relation to the 'phantasmal voices' heard (as a result of lip-reading) by the deaf poet David Wright (as quoted by the late neurologist Oliver Sacks). Similar faculties are also demonstrated by the illusion discovered by the psychologist Harry McGurk. In the 'McGurk effect', videos of misleading facial movements are read by viewers in such a way as to actively change the sounds perceived in speech recorded on the video soundtrack. Similarly, a factor within the cocktail party effect, named by the psychologist Colin Cherry, is that a listener's ability to see the facial movements of the person with whom they're talking improves their ability to accurately perceive speech, when speech is obscured by other voices.

EVP researchers also use prompting – prior notification to listeners of exactly what they should expect to hear, to create the illusions that the researcher's interpretations are shared by multiple listeners and are therefore 'objective'. This phenomenon closely matches the experiences of listeners to the sine-wave speech experiments, pioneered by the psychologists Philip Rubin and Robert Remez. Similarly a common aspect of EVP is the practice of adding noise, from sources like radio static and recordings of wind, rain, surf and hissing taps, etc., in order to (paradoxically) improve the success of EVP experiments. This practice closely relates to findings of several psychologists. David Bruce gave test subjects speech recordings that had been rendered barely audible beneath obscuring noise, and listeners reported hearing sentences on subjects they'd previously been told to hear. In an Albert Bregman experiment, listeners were unable to perceive the meaning of sentences in which successive segments had been artificially silenced – the speech was reduced to the equivalent of a series of stutters – but were able to interpolate illusions of continuous speech when the silences were replaced by bursts of noise. And Merckelbach and Van de Ven asked undergraduate students to listen to white noise and instructed them to press a button when they believed they heard a recording of Bing Crosby's 'White Christmas' (without this record actually being presented). A 'non-trivial minority' (32 per cent) consistently reported hearing 'White Christmas'.

These findings show how it is that with perceptually ambiguous source material, repeated listening, the addition of noise, and appropriate prompting, EVP researchers can trick listeners into perceiving ghost voices in stray communications chatter. Given that nearly a third of people can report hearing 'White Christmas' when it's not there at all, it's hardly surprising that EVP researchers can convince believers, when EVP signals are genuinely real (though not supernatural) and when EVP 'test subjects' are often people with the intensely powerful motive of having been bereaved.

Indeed, Justine Picardie quotes Sigmund Freud's discussion of how 'phantasies and unconscious thoughts about life in the womb…afford the deepest unconscious basis for the belief in survival after death, which represents a projection into the future of this uncanny life before birth'. In The Future of an Illusion Freud writes that 'the psychical origin of religious ideas' does not stem from 'experience or (from the) end-results of thinking', but from 'illusions, fulfilments of the oldest, strongest and most urgent wishes of mankind'. Freud further states that 'spiritualists…cannot succeed in refuting the fact that the appearance and utterances of their spirits are merely the products of their own mental activity' and further describes 'the pronouncements and information which they have received' as 'wretchedly meaningless'.

As Freud asserted, 'where questions of religion are concerned, people are guilty of every possible sort of dishonesty and intellectual misdemeanour', and Justine Picardie's mention of a medium who claimed to 'work for the police in murder investigations' provides a chilling illustration of the dangers of charlatanism. As to the ethical dilemma of disillusioning people of belief in the afterlife, Freud asserts that 'those who do not suffer from the neurosis will need no intoxicant to deaden it'. He advocated 'education to reality', saying that 'men cannot remain children forever': people will 'have to admit (to) the full extent of their insignificance in the machinery of the universe'.

Public understanding of science

Many within mainstream science seem to regard engaging with pseudoscience as being beneath their dignity. Yet, for the reasons I have outlined, I believe Rorschach Audio makes a significant contribution to the public understanding of science. Importantly, partly by virtue of the large following that EVP enjoys within sections of the electronic music and contemporary arts communities, this project makes a priority of engaging new audiences and of preaching to the unconverted. Lecture-demonstrations have been held in public museums, art galleries and even working men's clubs, and the book was produced by a publisher whose core readership is drawn by works often concerned with links between the arts and the occult. For a project that never had any professional PR support, it has also been reasonably well publicised – for instance in an Out of the Ordinary documentary on BBC Radio 4 (see www.bbc.co.uk/programmes/b01rg1gh).

In terms of illustrating aspects of the broader context, I remember a friend returning from a public lecture complaining about 'scientists debating how many angels sit on the head of a pin'. So, much of what is promoted to the public as 'cutting-edge' science is so far removed from tangible experiences of day-to-day reality that even the illusions perceived by EVP enthusiasts can (experientially speaking) seem more 'real', and certainly more emotionally appealing, than the information presented in TV documentaries about phenomena that people will never see, hear or feel (superstrings and quantum theory, multi-dimensional universes and black holes, etc.). Therefore, good as they are, some popular science resources, which intend to enthuse the public about the grandeur of nature and the excitement of rigorous science, may alienate rather than engage certain audiences. It is for that reason it's also important to continue engaging beliefs like EVP.

BOX - Illusion and reality

The phenomena that ghost-voice researchers (accidentally or deliberately) exploit arose from the mind's ability to smooth over interruptions and gaps, and to employ intelligent guesswork. These faculties evolved to help listeners recover coherent streams of meaningful speech in situations in which clear perception is compromised. It's the same with vision: consider the Necker cube illusion (as discussed by Richard Dawkins in The Extended Phenotype), kinetic depth effects (of the sort created by an art installation called 'The Analysis of Beauty'), and the blindspot phenomenon. The mental processes that generate what we perceive as illusions are the same processes that generate what we experience as reality. The projections that the mind generates to help us successfully navigate our world are, for most everyday practical purposes, highly reliable representations of reality. It may be tempting to characterise illusions as being perceptual 'mistakes', or as manifestations of what's sometimes termed 'anomalous' psychology, but it is perhaps most fruitful to emphasise the extent to which illusory perceptions emerge as by-products of normal cognitive processes.

Meet the author

'As an installation artist, electronic musician and researcher, much of my work focuses on exploring the creative potential offered by unusual uses and applications of radio technology. Radio intersects with electronic music and contemporary arts cultures in a field of research known as "electronic voice phenomena", whose followers believe they can use radio technology to record actual voices of ghosts. The popularity of such blatant pseudoscience within not only popular culture, but also increasingly within mainstream institutional arts culture, demonstrates a pressing need for public understanding of science projects that actively preach to the unconverted.'

Joe Banks is an artist and researcher.

Some key resources

Banks, J. (2012). Rorschach Audio – Art and illusion for sound. London: Disinformation.
Carlson, S. (1996, December). Dissecting the brain with sound. Scientific American, pp.80–83.
Deutsch, D. (1995). Musical illusions and paradoxes [CD]. La Jolla, CA:  Philomel Records.
Freud, S. (2001). The future of an illusion. London: Vintage. (Original work published 1927)
Picardie, J. (2001). If the spirit moves you. London: Picador.
Raudive, K. (1971). Breakthrough. Gerrards Cross: Colin Smythe.
Sacks, O. (1991). Seeing voices. London: Picador, London.
The Ghost Orchid (1999). An introduction to EVP (CD). Ash International.
Wright, D. (1969). Deafness. New York: Stein and Day.
http://rorschachaudio.com