“When jazz musicians improvise, they often play with eyes closed in a distinctive, personal style that transcends traditional rules of melody and rhythm,” says Dr. Charles Limb, a former research fellow with the National Institute on Deafness and Other Communication Disorders (NIDCD) and a gifted jazz saxophonist himself. “It’s a remarkable frame of mind.”
If you’ve ever been in “the zone,” making it up as you go along, or even seen someone hitting that sweet spot, you know it’s more than remarkable. It’s spiritual, it’s transcendent and it’s addictive.
Now, we have a clearer picture of how the brain helps us do that, a cognitive context for creative improvisation.
Limb and his fellow researcher at NIDCD’s (which is part of The National Institutes of Health) Division of Intramural Research, Dr. Allen Braun, chief of the division’s Language Section, both assumed that, as mystical as a musician might look following their muse, creativity is a matter of firing neurons. It’s tangible. We can understand it, and even see in action. That’s what Limb and Braun wanted to do: view, in real time, the brain functions of musicians during improvisation. But how do you see what musical improv (and beyond that, improvisation of any sort, from problem solving to having a conversation) looks like from the inside out? How do you view a brain on jazz?
The World’s Smallest Jazz Club
Laying on your back in a functional magnetic resonance imaging (fMRI) machine, unable to move your head, see your hands or hear much of anything over the drone of the machine is not the ideal situation in which to show of your musical skills. Until we invent X-ray goggles, though, functional MRI (which shows the amount of blood traveling to various parts of the brain so we can measure the amount of neural activity in those areas) is our best bet, so Limb and Braun had to turn the scanner into a tiny concert hall.
Six trained jazz pianists, three from the Peabody Institute and three who heard about the study through gossip in the local jazz community, lay down in the machine with their knees bent and were given some special equipment for their performance. A keyboard specially designed for the experiment (it was shortened to fit inside the machine tube and had its metal parts removed so the machine’s powerful magnets wouldn’t attract them) was rested on the pianists’ knees and a mirror was placed over their eyes so they could see the keys. The pianists also wore fMRI-compatible headphones so their music wouldn’t be drowned out by the din inside the tube.
And then, they played.
“We all do ‘do, re, mi,’ but you have got to find the other notes yourself.
“Because musical improvisation incorporates a broad range of melodic, harmonic, and rhythmic invention that is intrinsically difficult to control,” says Limb in his paper, Neural Substrates of Spontaneous Musical Performance: An fMRI Study of Jazz Improvisation, “we designed two paradigms, one that was relatively low and one that was high in musical complexity.” In the “low” paradigm, named Scale, the musicians were asked to play an ascending or descending scale, and were allowed to improvise in the “high” paradigm, named Jazz. This allowed the researchers to compare brain activity during the performance of a simple task to that during the performance of a more complex, creatively demanding one.
The Scale paradigm was based on the C major scale. The musicians first played the scale up and down in quarter notes along with a metronome, something most any accomplished musician has done countless times while practicing. They were then asked to improvise, but were limited to playing those same quarter notes within the C major scale. “Although the musicians were indeed improvising, it was a relatively low-level form of improvisation, musically speaking,” Limb said in a NIH press release.
In second paradigm, Jazz, the researchers aimed to “reproduce the high degree of musical richness of a jazz performance.” First, the musicians played a blues melody, written by Limb, that they memorized in the days before the experiment. They were accompanied, via the headphones, by a pre-recorded backing band. They then improvised again, using the chord structure of Limb’s composition as a guide and the backing band as inspiration.
That’s What Jazz Looks like to Me
Once things quieted down, Limb and Braun analyzed the brain scans. All six musicians showed similar brain activity patterns, and the researchers found that, during improvisation, certain parts of the brain were consistently activated while others were consistently turned deactivated.
The prefrontal cortex, the region of the brain’s frontal lobe that controls many of our higher mental abilities, is where the majority of changes happened. The dorsolateral prefrontal cortex, which is involved in intellectual function and action, and lateral orbifrontal cortex, which monitors and blocks out inappropriate behavior, acting as our self-censor, displayed a pattern of deactivation, almost to the point of shutdown.
On the other hand, the medial prefrontal cortex, which hasn’t been fully explored but is suggested to be involved in self-initiated thoughts and behaviors, became highly activated. Other brain scan studies have shown that this same region is very active when people tell anecdotes or make up stories.
The researchers say that the suppression of the musicians’ self-monitoring mechanisms and firing up of the “story telling” part of our brain makes sense given the notion that improvisation is an outlet for a musician to express their individual musical voice.
The brain scans also show that, during improvisation, there was increased neural activity in the sensory areas responsible for touch, hearing and vision, despite the fact that there was no significant change in what the musicians were touching, hearing or seeing when they switched from the Scale paradigm to the Jazz paradigm. Limb thinks that brain might “ramp up its sensorimotor processing in order to be in a creative state.”
The most interesting finding is that the brain scans from the two improv sessions were nearly identical, the same pattern of activation and deactivation described above occurred whether the musicians were improvising within the one-octave scale or had free reign to do whatever they wanted over Limb’s tune. This lends some support to the idea that, basically, this is what creativity looks like. If the difference in neural activity between the memorized and improvisational paradigms was the result of increasing complexity, then there should have been a greater difference between the two improvisation sessions, also, since the Jazz paradigm improvisation was more complex than its Scale counterpart. Braun concludes that “there is no single creative area of the brain—no focal activation of a single area. Rather, when you move from either of the control tasks to improvisation, you see a strong and consistent pattern of activity throughout the brain that enables creativity.”
Limb says that this pattern of brain activity may also be present during other types of improvisational behavior. He and Braun plan to use similar experiments to see if the brain activity they have found also occurs when other artists, like writers or painters, and non-artists are asked to improvise.
On that note,further experiments with musicians wouldn’t be a bad idea, either. This experiment overlooks an important element of improvised music, especially in jazz: the social factor. Jazz improvisation is about more than just the soloist and their instrument, the musicians play off of each other and the musical relationships unfolding during each measure affect the output of every person in the group. Designing an experiment that accounts for the social dimension of musical improvisation would be nearly impossible with the equipment we have now, though. Even if we could line up four or five fMRI machines and make sure the musicians could hear each other clearly, it would be no small feat to play the saxophone, much less the drums, inside the scanner.
Limb and Braun’s study was published in the February 27th issue of Public Library of Science (PLoS) One, an open-access journal. The full paper is available to read in its entirety on the web.
Reference: Limb, C.J., Braun, A.R., Greene, E. (2008). Neural Substrates of Spontaneous Musical Performance: An fMRI Study of Jazz Improvisation. PLoS ONE, 3(2), e1679. DOI: 10.1371/journal.pone.0001679
 Dr. Limb is now an otolaryngologist at the Johns Hopkins University School of Medicine and faculty member at the university’s Peabody Conservatory of Music.