Image by Lars Stenman/Getty Images

To study how auditory and verbal hallucinations work, Swiss scientists developed a robotically-assisted technique to make people who have no history of mental illness hallucinate voices that are not there.

In an interview with PsyPost, Pavo Orepic, a University of Geneva postdoctoral researcher who co-authored a recent paper in the journal Psychological Medicine about the phenomenon, explained his experimental methodology, which included a poking robot, "pink noise," and recordings of both the participants' own voices and the voices of other people.

Though there has been some research suggesting that auditory verbal hallucinations — "hearing voices" — are the result of the brain failing to recognize its own thoughts as self-generated and processing them as outside stimuli, the psychiatric and neurological communities still don't have a satisfying explanation.

While these auditory hallucinations are most often associated with the psychiatric disorder schizophrenia, which is characterized by such experiences, people with other mental health issues — and sometimes even folks who haven't been diagnosed with any mental health issues — sometimes hear voices that aren't there, too, and scientists still don't know why.

To try to start untangling it all, Orepic and his colleagues at the Swiss Federal Institute of Technology Lausanne's Laboratory of Cognitive Neuroscience recruited 48 healthy subjects who were fluent in French and who were unaware of the study's purpose — which, provocatively, was to make them hear voices.

First, the Swiss researchers recorded the patients' own voices saying nine one-syllable words that have negative connotations — because, as PsyPost points out, many auditory verbal hallucinations involve negative or mean words — as well as the voices of people unknown to the participants saying the same words.

Recordings in hand, the blindfolded participants were sat in between two parts of a robotic unit: one in front of the subjects, which they were asked to poke, and one behind them, which poked their backs in return. The experimenters timed these pokes to be both simultaneous — that is, the robot poked the participants at the same time they poked it — and on a brief delay, which saw the robots poking them a few seconds after they first poked it.

After introducing subjects to the poke-bot, the researchers tested out participants' hearing thresholds by playing their own voice recordings, played randomly within a larger interval of "pink noise," which is lower in frequency than white noise and considered more calming. This was then repeated with the recordings of other people's voices occurring within the pink noise intervals.

Following the hearing threshold tests, researchers then got to the meat of the experiment by playing, at various times, the same pink noise from the previous portion without any voice recording additions, which sometimes was played synchronously with the pokes and sometimes just after.

Diagram of the researchers' "task block design" featuring the poking robot. Image via Psychological Medicine/Cambridge.

Participants were asked to push a button when they heard voices in the pink noise and later to describe what they heard, which the researchers used to determine whether they were hallucinating their own voices or the recordings of other people. These "false alarms," as the researchers described them, most often occurred during the synchronous pokes — and fascinatingly, subjects indicated that they more often heard the recordings of their own voices in the noise than those belonging to strangers.

Orepic told PsyPost that along with the auditory false alarms, he and his fellow researchers also "observed that people who are more prone to delusional thinking were more likely" to hear voices.

"This supports the theory that experiences like hallucinations and delusions are present across the phenomenological spectrum even in the general population," he added.

Let's be honest: it's a baroque experiment that blurs the line between Pavlov and pareidolia, and it's not entirely clear what the takeaway is — except, perhaps, that our brains remain deeply mysterious biological mechanisms that are capable of alternately discerning incredibly obscure patterns or of imagining them entirely.

More on voices: Scientists Say New AI Can Translate What Chickens Are Saying


Share This Article