Audio technique makes people feel that their doppelganger is in the room with them
Hallucinatory experiences are pretty common, despite widespread beliefs to the contrary.
12 November 2021
By Emma Barratt
Researchers estimate that around 40% of the population regularly experience hallucinations to some degree, whether that be feeling your phone vibrate only to find no notifications, or a full-on out of body experience.
Though hallucinations are common, some are harder to study than others. One of the trickiest has been autoscopic hallucinations, during which people experience a doppelganger of all or part of their body in the space around them.
Though this hallucination has been reported to clinicians, its rarity has made it difficult to observe in an experimental setting — so researchers have instead attempted to induce the phenomenon in the lab. Now, Marte Roel Lesur and team at the University of Zurich have developed a way to experimentally induce it in people who have no history of hallucinations using only auditory cues.
Previous research has placed significant emphasis on the role of vision in producing these hallucinations, and some visual methods have even been developed to induce these kinds of experiences experimentally. For example, in 2007 researchers at University College London were able to shift participants' perceptions of where their bodies were in space using a virtual reality video feed of their backs, positioned a few metres behind where they were sitting. But people sometimes also report auditory hallucinations during this phenomenon, leading the team in Zurich to develop a new method that relies on audio cues alone.
In this study, published in Consciousness and Cognition, participants (N=25) produced a series of tightly controlled binaural recordings of themselves walking into a room and reading excerpts from The Little Prince. One set of microphones were placed in the ears of a dummy sitting across the room from the participants, in order to create recordings with true-to-life location cues. Another set of microphones recorded from the ears of the participants themselves.
Participants then sat with their head in the same location and position as the dummy head used for the recordings. While watching a virtual empty room through a VR headset, the participants were played three binaural audio recordings as separate conditions: the recording that had been taken from their own ears, in which they heard themselves from their own location (the self-egocentric condition); the recording that had been taken from the dummy, in which they heard themselves from across the room (the self-allocentric condition); and a recording that had been taken of a gender-matched person speaking from across the room (the other-allocentric condition). After each of these conditions, they answered a 14-point questionnaire measuring how closely participants felt that there was another presence in the room, and how closely they identified it as themselves.
In the second block, participants were played audio of either themselves or matched others navigating around the room, reading from an experimental script — "One, two, three, four, five, point at my feet." Participants were then asked to point a virtual laser pointer to the floor of the location they heard the audio originating from, in order to give a measure of perceived distance from the speaker. After all the trials were done, they answered two questions to assess if they'd had any previous hallucinatory experiences.
Data collected from the 14-item questionnaire indicated that participants strongly felt a presence walking around them while the self-allocentric audio played (they didn't report a very strong sense of a doppelganger during the self-egocentric audio). Participants also self-identified with that presence, feeling a sense of body ownership and agency when it was linked to their own voice, as opposed to that of the gender-matched others.
These results suggest that some aspects of bodily self-consciousness can be affected by audio in those with no history of psychiatric or neurological issues. These participants experience what the authors describe as an "auditory doppelganger", with whom they strongly self-identify. One participant even went so far as to report that they felt "as if I was going around myself." This effect may offer a new avenue through which to investigate bodily self-consciousness in experimental settings.
Perhaps most interesting, however, is that when asked to point to the locations voices originated from, those with no history of hallucinations perceived their own voice in the allocentric condition as originating closer to them than the voice of another person. Conversely, those with a history of hallucinations tended to locate their own allocentric voices as further away from themselves than the voices of others.
Though the authors of this paper emphasise that their findings are preliminary, and based on a very small sample in the hallucinator group (N=4), they offer some speculation as to why this difference may exist, citing previous research on patients with psychosis showing reduced capacity to recognise their own voice. However, even if those in the hallucinator group experienced difficulties in distinguishing their own voice, it's unclear why they would estimate their voice to be further away than the voice of another person. As this finding was contrary to the team's original expectations, this issue is likely to be more fully explored by future research.
The study was quite small, which may raise questions as to validity and generalisability of these results. Similarly, sorting of participants into those with and without a history of hallucinations was done using two questions, which is perhaps lacking in the detail required to conduct this kind of group separation. Temperature and electro-cardiac measurements taken from participants during the blocks did not differ significantly between conditions, which may also indicate that the experience of a doppelganger was not particularly strong. This research is preliminary in nature, though, and these points can be improved in further studies.
Further reading
About the author
Emma L. Barratt (@E_Barratt) is a staff writer at BPS Research Digest
