Hallucinations tend to be associated with psychosis, but the reality is more complicated than that. Some people who hear voices don’t suffer from other mental health problems, and the voices they hear aren’t distressing. These “non-clinical voice-hearers” provide an important opportunity to understand hallucinations without the complications of mental illness or medication.
A preliminary study published this week in the journal Brain reports that non-clinical voice-hearers were more likely to detect language in a recording of distorted speech. Voice-hearers also showed some different patterns in brain activation as they listened. The results could help to explain why some people are more likely to hear voices, as well as help to direct future research on the topic.
Hearing meaning in noise
Ben Alderson-Day, the lead author on the paper, is a psychologist at Durham University whose research focuses on auditory hallucinations. To investigate differences of perception in voice-hearers, Alderson-Day and his colleagues used sine-wave speech, which strips out some of the most vital acoustic properties of speech and leaves something that sounds kind of like a series of clicks and whistles. It’s possible to understand it—once you already know what it says, or once you’ve listened to quite a bit of sine-wave speech. (Listen to some examples here.)
Participants weren’t told that their task involved voices or speech. They were just told that they would be listening to a series of sounds while in the fMRI machine and that they should press a button when they heard a “target sound” that was “noisier” than the other sounds. The mix of sounds they heard consisted of sine-wave speech, as well as other samples that sounded pretty much the same as sine-wave speech, but scrambled to the point where they didn’t encode any intelligible language at all. The “target sounds” were unintelligible sine-wave speech that had been edited further to sound different.
Participants were told how far along they were in the task by being given markers saying “Block 1,” “Block 2,” etc. When they were finished, they were asked whether they had noticed any speech in the sounds they had heard and, if so, when they had first noticed it. Only half of the control participants reported noticing the speech, but three-quarters of the voice-hearers noticed it. They also picked up on it earlier than the control participants.
Data from the fMRI showed an interesting pattern: both groups shared a similar pattern of brain activation when listening to unintelligible sine-wave speech. When the speech was intelligible, however, voice-hearers’ brains had activation in some additional areas. That points to the voice-hearers’ brains being better at discriminating between the intelligible and unintelligible noise. The results suggest that people who hear voices aren’t just imagining speech in any distorted sound—they’re better at picking it up when it’s actually there.
Primed to hear speech
Participants were then told that there had been language-like sounds in the task and were played some examples of sine-wave speech until they could decode them fairly accurately. After this training, they repeated the task; again, the participants were just listening for target sounds, but now they knew that there was speech elsewhere in the recording. This time, the two groups showed no difference in perception of speech.
The training didn’t seem to affect either group’s brain responses to the unintelligible sounds, suggesting that the training didn’t cause anyone to imagine language where there wasn’t any. And the two groups had similar responses when it came to intelligible speech, suggesting that the voice-hearers didn’t get an extra boost from the training—once all the participants had the expectation that the sounds were language-like, they were equally good at hearing them. That suggests that the major difference between the two groups is their default expectation, or priming.
This was a small, preliminary study, so its results will need to be confirmed with further research. Recruiting voice-hearers for research is a tricky process, because many people aren’t inclined to share the fact that they hear voices. Studies using expensive fMRI equipment are often small, and the authors of the paper point out that the sample size of 12 voice-hearers and 17 control participants is smaller than generally recommended. But it’s the biggest neuro-imaging study conducted with non-clinical voice-hearers, which means that the results are particularly useful for helping to illuminate the most important questions for researchers to ask next.
This study points to voice-hearers’ brains as having a different response to ambiguous stimuli: when they have the option to perceive meaning in noise, they do so at a higher rate than people who don’t experience auditory hallucinations. It suggests that hallucination may develop from “ordinary perceptual processes,” the researchers write—and it also illustrates “the continuity of mundane and unusual experience.”
Powered by WPeMatico