Science Watch

In her 30 years as a music therapist, Connie Tomaino has seen hundreds of people incapacitated by stroke, Parkinson's disease or accidents learn to walk again—to the cadences of carefully chosen music.

Tomaino is the senior vice president for music therapy at Beth Abraham Family of Health Services, a residential and community-based health center in New York. She was a college trumpet player and pre-med student three decades ago when she realized that she could combine her two passions in the relatively new field of music therapy.

"Even before I came here in 1980, there was an understanding that, for patients with Parkinson's disease or brain injury, music could help them spontaneously move in a way that they couldn't move on their own," she says. "Clinically we knew that this worked, and we've been able to perfect how it's used therapeutically over the years."

What they didn't know was precisely why it worked, Tomaino says. Now, using such brain imaging techniques as fMRI and MEG, researchers are beginning to answer that question. They're finding evidence that music and motion are neurologically intertwined, and that motor areas of the brain may be key to our ability to perceive music's beat.

"What science is telling us now is what the mechanisms behind [music therapy] are," Tomaino says.

Where in the brain?

One of the scientists behind that finding is Jessica Grahn, PhD, a postdoctoral researcher at the Medical Research Council's Cognition and Brain Sciences Unit in Cambridge, England. Grahn is also a musician—a pianist—and she's particularly interested in how people extract a steady "beat" from many different rhythms.

A beat is what makes you want to tap your foot to a song at a steady pace, but it's not at all clear how the brain finds that regular beat among a song's complicated rhythms. Now, Grahn's research suggests that motor areas may be involved.

In a study published last year in the Journal of Cognitive Neuroscience (Vol. 19, No. 5), Grahn used fMRI to scan people's brain activity as they listened to different types of rhythms. Those rhythms were simply strings of different-length computer-generated beeps: "Doooo Doo Doo, Doooo Doo Dooo Doo."

Only some of the rhythms produced a perception of a beat—the ones where the beep lengths were all integer multiples, such as one, two, three or four. To produce a beat, the researchers also had to arrange the beeps in groups where beep lengths added up to four (2-1-1 | 3-1 | 4).

That's because most Western music is in 4/4 time, Grahn says, and she wanted to use rhythms that she knew would produce a beat experience for her participants. People from other cultures, whose music is structured differently, might experience beat and rhythm differently.

"If you took a tribe from South America, would they feel these particular rhythms as having a beat? I don't know," she says. "But we wanted to test people with rhythms they were used to moving to."

Grahn found that when people listened to the beat-generating rhythms, their basal ganglia and supplementary motor areas lit up. When they listened to the non-beat-generating rhythms, those areas stayed dark. That suggests that we may use the areas—both of which are involved in motor control and coordinating movement—in detecting beats as well.

When in the brain?

John Iversen, PhD, a neurobiologist at the Neurosciences Institute in San Diego, is using a different imaging technique to study a similar topic: magnetoencephelography (MEG). This technique measures magnetic fields produced by neurons' electrical activity to examine the brain's activity as it listens to rhythms and beats.

MEG doesn't pinpoint the location of brain activity as precisely as fMRI, but it works on a much finer time scale. With MEG, Iversen can examine the brain's response to each note in a rhythm, second by second.

"When you play a sound, the brain responds with oscillations over a wide range of frequencies, like a bell ringing," he says.

In one study, recently accepted in the Annals of the New York Academy of Sciences, Iversen and his colleagues asked participants to listen to a very simple rhythm—two 45-millisecond tones followed by a 200-millisecond rest.

This is called an ambiguous rhythm because listeners can choose to hear the beat on either the first tone (DA-dum, DA-dum, DA-dum) or the second (da-DUM, da-DUM, da-DUM). In different trials, Iversen and his colleagues asked participants to hear the beat on either the first or second tone (all of the participants were trained musicians to ensure that they understood the concept of beat).

The researchers found that when people listened to the tone they imagined to be the beat, their brain activity at the upper end of the beta range—between 20 to 30 Hz—increased by an average of 60 percent. That end of the beta range is active in motor processing, and also active when distant areas of the cortex communicate—areas such as the motor and auditory systems, Iversen notes.

Edward Large, PhD, a neuroscientist at Florida Atlantic University, seconds the idea that the auditory and motor systems communicate via the beta band—and that this communication is the key to beat perception.

"I think there's a distributed system responsible for beat perception," he says. "It's happening in the basal ganglia and SMA [supplementary motor areas] but also in the auditory pathway."

In a study published in 2005 in Cognitive Brain Research (Vol. 24, No. 1), he used electroencephalography to image participants' brain activity as they listened to different rhythms. All of the rhythms had a regular beat, but occasionally the "beat-note" would be missing, with silence in its place.

Large found bursts of beta activity that spiked around the beat. The activity even spiked when the beat-note was missing, suggesting that just the anticipation of a beat could trigger it.

That beta band activity, he explains, could reflect the link between the auditory and motor processing areas.

The research neuroscientists and psychologists are conducting will inform music therapy, and eventually make it even more effective, says Tomaino, who co-founded the Institute for Music and Neurologic Function at Beth Abraham in 1995 to bring researchers together with practicing music therapists. For example, recent research that suggests that simply imagining a beat and hearing it in one's head can activate motor areas in a similar way as actually hearing it. That suggests it might be helpful to teach patients to sing or imagine rhythms on their own, without an external source, Tomaino says. "There was a gap between neuropsychologists and clinical music therapists, who usually don't sit in the same room," she says. "I wanted to encourage them to come together, so we can get to better treatment."


Lea Winerman is a writer in Washington, D.C.