Wednesday, February 27, 2008

Science Messes with Jazz (Again?)

Having staked my doctoral degree on a thesis about computer music, I am always interested in what happens when the scientific community decides to make similar ventures into the world of music. One of those ventures was reported this morning on Net News Publisher:

Scientists funded by the National Institute on Deafness and Other Communication Disorders (NIDCD) have found that, when jazz musicians are engaged in the highly creative and spontaneous activity known as improvisation, a large region of the brain involved in monitoring one’s performance is shut down, while a small region involved in organizing self-initiated thoughts and behaviors is highly activated. The researchers propose that this and several related patterns are likely to be key indicators of a brain that is engaged in highly creative thought. NIDCD is one of the National Institutes of Health. The study is published in the Feb. 27 issue of the journal Public Library of Science (PLoS) One.

Unfortunately, this study reveals more conflicts than insights. This is best illustrated by two additional excerpts from the Net New Publisher account. One is a summary of the experimental design:

During the study, six highly trained jazz musicians played the keyboard under two scenarios while in the functional MRI scanner. Functional MRI (fMRI) is an imaging tool that measures the amount of blood traveling to various regions of the brain as a means of assessing the amount of neural activity in those areas.

The other involves how the subject was introduced in the opening paragraph:

When John Coltrane was expanding the boundaries of the well-known song “My Favorite Things” at the Village Vanguard in May 1966, no one could have known what inspired him to take the musical turns he took. But imaging researchers may now have a better picture of how the brain was helping to carry him there.

That introduction is important because it exposes the pitfalls of trying to study something as subtle as the practice of music in a laboratory environment. Scientific investigation can rarely assess any phenomenon that is not normative, and Coltrane's practices were anything but normative. The November 17, 1962 Paris performance of "My Favorite Things" clocked in at almost 24 minutes; and the extended durations of Coltrane's solos even earned him sardonic criticism from Miles Davis! This is such a "statistical outlier" that I doubt that any examination of "six highly trained jazz musicians" will reveal very much. In other words the most interesting hypotheses that would try to link brain activity to improvisational behavior are likely to be the ones that account for the behavior that is least normative.

More important, however, is that improvisation is a social practice that goes far beyond the relationship between soloist and instrument, which the experimental design tried to capture. Improvisation is driven by the relationships that are taking place across the entire ensemble. One cannot understand Coltrane's behavior without taking into account the other members of his quartet. Any experiment that does not take this social dimension into account can only provide an impoverished body of data. In fairness, however, the idea of designing an experiment that would yield more valid data is so challenging as to be virtually impossible, at least with current equipment.

This is an excellent example of what happens when we run up against the limitations of what Thomas Kuhn called "normal science." We need to reassess both the questions we ask and the methods we design to answer them. Kuhn called this a "paradigm shift." Unfortunately, when funding for scientific research is limited, "normal science" tends to prevail; and the result is a collection of well-funded scientists running in circles like a dog chasing its tail!


dk said...

Stephen, from my experience as an improvisor and music therapist, I agree that to include all the social elements of the improvised experience is challenging for "normal science." I am encouraged though by the increased interest of the established scientific community in this area.

Based on your expertise in this area, do you think that their conclusions, that the dorsolateral prefrontal cortex shows decreased activity during the improvising condition are accurate?

thanks, dk

Stephen Smoliar said...

Actually, it seems that quite a lot has happened in "wet brain" theory since I posted this. Indeed, at a "meta" level, I might argue that each advance in imaging technology seems to broaden the scope of what we would call "normal science." My most recent post on the brain-music connection was entitled "
Your Brain on Music is a Highly Distributed Processor
" and concerns results just presented at the annual meeting of the American Association for the Advancement of Science in San Diego. The quote that stuck with me came from Dr. Aniruddh Patel: "Music engages huge swathes of the brain - it's not just lighting up a spot in the auditory cortex." From this I would conclude that dorsolateral prefrontal cortical activity needs to be part of the model, but it is only one node in a causal network. We probably do not yet know the full extent of that network. For that matter we may need something better than causal chains for reasoning about such highly distributed systems.