Bone conduction helps you distinguish between your own voice and the voice of others, according to new research published in Royal Society Open Science. Bone conduction refers to the process by which sound waves travel through bones in your skull and inner ear, instead of through the air in your ear canal.
Our own voice often sounds strange and unfamiliar when played back to us in a recording. This is partly the result of the lack of bone conduction, which alters the acoustic properties of our voice. This has made it difficult to study self-voice perception, making it one of the least studied aspects of self-awareness. The researchers behind the new study sought to use relatively new technology — bone-conduction headphones — to overcome this issue.
“We found it fascinating that hearing our own voice feels so unnatural to us,” explained study author Pavo Orepić, a postdoctoral research fellow at the Laboratory of Cognitive Neuroscience at the University of Geneva. “Moreover, hearing own voice is often accompanied by feelings of discomfort. This is strange because our voice is the sound we associate most with ourselves, it plays an essential role in our everyday lives, and it is our main means of communication.”
The researchers conducted three studies to investigate voice discrimination in human subjects. The participants in all three studies (84 in total) were right-handed, reported no hearing deficits, and had no history of psychiatric or neurological disorders.
In each study, participants were asked to vocalize the phoneme /a/ for 1-2 seconds. Audio recordings of their vocalizations were then normalized for intensity and duration, and background noises were removed.
The resulting 500 ms clips were then combined with preprocessed voice recordings to generate a number of voice morphs. These voice morphs spanned a continuum between the two voices. In other words, the voice morphs contained 85% of person A’s voice and 15% of person B’s voice, 70% of person A’s voice and 30% of person B’s voice, and so on.
The voice morphs were presented to participants through either air-conduction (using laptop speakers or headphones) or bone-conduction headphones.
Bone-conduction headphones work by transmitting sound vibrations through the bones of the skull. The headphones typically rest on the cheekbones or the temple, and use tiny transducers to convert electrical signals into vibrations that are sent directly to the inner ear.
In Study 1 and 2, participants were instructed to indicate whether a voice morph they heard more closely resembled their own or someone else’s voice. The voice morphs were between an unfamiliar person’s voice and their own voice (Study 1) or between a familiar other’s voice and an unfamiliar person’s voice (Study 2).
In Study 3, participants were accompanied by a friend of the same gender and similar age. Each pair of participants performed the self-other and familiar-other voice discrimination tasks (air or bone) as in Study 1 and 2, respectively. The researchers also examined whether participants could recognize their unmorphed recorded voice.
Orepić and his colleagues found that bone conduction improved voice discrimination if the task involved self-voice morphs. Results from Studies 1 and 3 indicated that participants were better at telling their own voice apart from someone else’s voice when they heard the sounds through their bones instead of through the air.
But bone conduction only helped people distinguish their own voice from the voice of others. Study 2 demonstrated that it didn’t help the participants distinguish between the familiar and unfamiliar voices of others.
“When we speak, our skull vibrates and our brain processes not just the sound of our voice, but also these vibrotactile sensations. Our study showed that participants better recognize their voice when they hear it through bone-conduction headphones, compared to regular headphones,” Orepić told PsyPost.
“By adding an additional vibrotactile signal to the sound of our own voice, we made the experience of hearing our own voice more natural. Our study suggests that our voice is not just a sound, but fundamentally a multisensory construct.”
Interestingly, participants often mistakenly identified familiar voices as their own. This happened even when the familiar voice sounded different from their own. This suggests that we are more likely to confuse a familiar voice as our own because we are used to hearing it.
“It was surprising that participants more often misattributed familiar voices as their own, compared to unfamiliar voices, regardless of which voice (familiar or unfamiliar) was acoustically more similar to their own voice,” Orepić said. “This shows that familiarity mechanisms affect own-voice perception.”
The researchers also noted that the way we naturally perceive our own voice also includes motor signals related to speech production, which can influence the perception of our voice. “When hearing our own voice in natural conditions, while speaking, there are inevitably motor signals present that were not tested in our study,” Orepić explained. “The presence of such motor signals and the associated intraoral and pharyngeal sensory speech-related cues may also exert additional effects on own-voice perception.”
The findings have clinical relevance for conditions such as schizophrenia.
“This work could serve as a scaffold for clinical investigations of a very common and highly distressing psychiatric symptom—auditory-verbal hallucinations, i.e. ‘hearing voices,'” Orepić told PsyPost. “Although we still do not know why psychiatric patients hear voices in their heads, there is ample evidence suggesting that hallucinating other voices is closely related to the inability to recognize own voice.”
The study, “Bone conduction facilitates self-other voice discrimination“, was authored by Pavo Orepic, Oliver Alan Kannape, Nathan Faivre, and Olaf Blanke.
