by Nora Schultz
06 July 2009
WHAT does the human brain sound like? Now you can find out thanks to a technique for turning its flickering activity into music. Listening to scans may also give new insights into the differences and similarities between normal and dysfunctional brains.
Brain scans created using functional MRI consist of a series of images in which different areas light up with varying intensity at different times. These can be used to determine which parts of the brain are active during a particular task.
To turn such scans into music, philosopher Dan Lloyd at Trinity College in Hartford, Connecticut, identified regions that become active together and assigned each of these groups a different pitch. He then created software that analyses a series of scans and generates the notes at these pitches as the corresponding brain areas light up. Each note is played at a volume that corresponds to the intensity of activity.
When Lloyd fed the software a set of scans of his own brain taken as he switched between driving a virtual-reality car and resting, he found that he could detect the switch-over in the sounds.
Lloyd then gave the software scans taken from volunteers with dementia and schizophrenia, and from healthy volunteers. The brains of people with schizophrenia switched between low and high activity more erratically than healthy brains, allowing the two types of brain to be distinguished by sound alone.
While this difference is also clear from looking at the images, Lloyd's collaborator Vince Calhoun at the University of New Mexico in Albuquerque, says there are variations in the music from people with schizophrenia that are not visually obvious. "It almost sounds like there is more background warbling," he says.
He suggests that these "unsteady rhythms and cadences" may be indicative of dysfunction in the brain. Lloyd also identified sounds and rhythms in the brains of people with dementia that distinguished them from healthy volunteers.
Could identifying such aural differences ever be useful? Daniel Levitin, a neuroscientist at McGill University in Montreal, Canada, thinks they might. He says brain music's killer application might be in allowing researchers to home in on patterns that suggest a particular region is interesting and that wouldn't be detectable using the eye alone. They could analyse these regions more closely using conventional imaging.
His colleague Didier Grandjean at the University of Geneva in Switzerland says that brain music might help identify temporal patterns in particular. "Melodies are a much better way to build complex mental representations over time than anything the eye can do," he says.
Lloyd is also keen to explore the aesthetic aspects of brain music. "It's not quite like composed sound but it's not random either, it's 'almost music'. My students are putting it on their playlists."