You're lying face-up with your head surrounded by a large medical scanner. You've been told that you have to keep your head completely still. You can't look down to see your hands and you can't hear very much beyond the background hum of the machine. A perfect time for some improvisational jazz then....
This was the challenge posed to six professional jazz musicians by Charles Limb and Allen Braun from the National Institute on Deafness and Other Communication Disorders. They were told to play that funky music while undergoing fMRI (functional magnetic resonance imaging), a brain-scanning technique that would enable the two researchers to quite literally see the creative process in action.
Scanning the creative mind
Improvisation is the cornerstone of great jazz performances, and it is a joy to watch musicians deftly adapting pieces of music on the fly to create something familiar yet wholly original. To Limb and Braun, this skill is a fine example of the creative process, itself a "quintessential feature of human behaviour".
The duo approached their study with the assumption that there is nothing mystical or obscure about creativity. Like all other matters of the mind, it is the result of firing neurons, a tangible process that we can understand and even see in action. That's where the fMRI came in, but getting someone to play properly within such restrictive confines is challenging to say the least.
To do it, Limb and Braun ordered a special, custom-built piano keyboard, that lacked any metallic parts that would be affected by the scanner's powerful magnets. Each musician lay down in the scanner with their knees up and the keyboard in their laps. An angled mirror positioned over their eyes allowed them to see the keys, and a small earpiece allowed them to hear what they were playing.
In the 'Scale' task, the musicians were either asked to repeatedly play the simple C major scale within one octave, or to play an improvised melody of their choice using only the same nine notes. This was a fairly basic form of improvisation and Limb and Braun also wanted to capture the much richer nature of true jazz performances.
In the more complicated 'Jazz' task, they were first asked to learn an original jazz composition. Once in the scanner, they either had to play the unaltered tune, or to improvise freely, using the chord structure of the new piece as a guide. In both cases, they listened to a pre-recorded accompanying quartet, which they could use as inspiration during the improvisational phase.
The musicians showed very similar patterns of brain activity during both the Scale and Jazz experiments, despite the different degrees of difficulty. In both cases, certain parts of the brain were consistently activated during improvisation, while others were consistently turned off. In the image below, the red areas represent bits which are more strongly activated when improvising, and the blue areas are bits that are comparatively turned off.

What improvisation looks like
The prefrontal cortex, which controls many of our higher mental abilities from planning to problem-solving, was particularly strongly affected. Large swathes of it were completely shut down, including the LOFC (lateral orbifrontal cortex), which monitors and blocks out inappropriate behaviours, and the DLPFC (dorsolateral prefrontal cortex), which is involved in planning and focused, methodical thinking.
These widespread deactivations could help musicians to avoid over-thinking a task. By turning off parts of the brain involved in self-assessment and focused attention, they could pave the way for actions and ideas that are spontaneous, unplanned and new.
Within the large prefrontal cortex, only one area was strongly activated during improvisation - the medial prefrontal cortex (MPFC). Recent studies have suggested that the MPFC plays a role in our sense of self, and it certainly lights up when people recount personal anecdotes. To Limb and Braun, its heightened activity fits with the idea that musicians use improvisation as an outlet for expressing their own musical voice and experiences.
The fMRI images also revealed that when improvising, the musicians' brains showed more activity in areas involved in hearing, sight and touch, even though they were seeing, touching and hearing much the same things as they were when playing from memory. Their improvised pieces were no faster than the originals nor did they include more notes. Instead, Limb and Braun suggest that during the creative state, a musician's senses are heightened across the board.
The scans imply that there is a very consistent pattern of brain activity linked to creativity, a pattern of heightened senses and self-expression with a lack of conscious control. And the very similar results from the Scale and Jazz scans suggest that this has nothing to do with the complexity of the music and everything to do with the creativity of the performers.
Limb and Braun also tantalisingly suggest that a similar prefrontal shut-down could also be linked to other states of altered consciousness including meditation, hypnosis and most intriguingly, dreaming.
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
For more on the psychology of music, have a look at this post from the classic sight about how music sounds right because it reflects the hidden tones in our speech.
Image of jazz pianist by biskuit
(Mo from Neurophilosophy has also written about this paper but I really couldn't resist doing my own take. It's just such an awesome study)





LOLbachia













Email this entry to a friend
View the Technorati Link Cosmos for this entry

Comments
As a jazz musician, I've been following the commentary on this paper fairly closely. I have a few unrelated comments:
* Range restriction (improvising within an octave or other set interval) is a fairly common exercise used by practicing jazz musicians. It's really helpful in developing skills and coming up with new ideas. Dunno if that's relevant to the use of this technique in the study.
* When I was studying with David Baker, he made the comment that a great solo does not sound predictable, but it does sound inevitable. It's important to remember that improvisation is not random, but that improvisations are built out of commonly understand musical tropes and gestures. Sometimes (cf. Ornette Coleman), those tropes are idiosyncratic, but more often it's built off of culturally shared ideas about what music sounds like. Hence the aphorism, "jazz is a verb." Most jazz is something you do in the context of some existing popular music.
* One of things I stress with younger musicians is the importance of listening to yourself. The way I explain it is that there's a part of your consciousness that tells you what to do, and a part that listens to what you do. The goal is to increase the amount of listening you do, and decrease the amount of telling. The point is, if this act is what corresponds to shutting down the prefrontal cortex, is that this is a learnable skill, and not an emergent property. I suspect that if you could scan someone's brain while playing tennis or basketball, you'd see something similar going on. Cf. the Inner Game series of books.
* Worth mentioning: Most jazz musicians have a repertoire of "crutches," longer bits of memorized phrases and gestures that they can rely on when "pure improvisation" is too taxing. Another practical element is what Dick Sudhalter calls "routining," where a musician develops a particular approach to a particular tune, and may refine that approach over time, but not change it radically from performance to performance.
* I also wonder if this indicates that prior studies have misunderstood the function of the prefrontal cortex. The description of what functions are attenuated doesn't quite jibe with my personal experience of what happens when I improvise. It's close, but not quite right.
Posted by: HP | March 3, 2008 10:54 AM
Great post.
I tutor a systems biology class. At the start of this course, the professor encourages his students to imagine " a kitchen, where the stove is found within the fridge". Which gets everyone thinking. He then explains to the class that we tend to assign a set of rules to our world and that to be creative we have often have to let our minds not work within those rules. He therefore encourages his students to work hard, but to take time off to get into that state, to imagine, to dream.
This work, in a way, confirms his approach.
Posted by: cc | March 4, 2008 5:54 AM
Just wanted to say that I have just discovered this site and thank you! I am not a scientist but a developmental psychologist instead (I think - I don't like labels) and my area is early development. This is all such wonderful stuff! Thanks again
Posted by: Maria Robinson | March 5, 2008 5:12 AM
Thanks all for the really interesting comments.
@HP - it's fascinating to get the perspective of an actual jazz musician. Regarding the role of different brain regions, I expect that our understanding of their functions will greatly improve as technology advances and more studies are done. I tend to see our current knowledge as a very early draft...
@cc - love the stove-in-a-fridge task; might try it myself.
@ Maria - you're welcome, and welcome :-)
Posted by: Ed Yong | March 5, 2008 9:25 AM