I agree with Liana. I am a faculty trumpet member of my college wind band, and I exhibit a great amount of creativity in terms of tone quality, dynamics, vibrato, and style even though I don't play any improvised riffs. I heard that a machine was created that detected distinct dynamic levels. When brought to a quality high school band performance, it detected five distinct dynamic levels. When brought to a performance of the Chicago Symphony Orchestra, it detected twenty-eight distinct dynamic levels. So if technology can quantify non-improvisational creativity in one regard, why not others?

It would certainly be interesting to consider performance variability and creativity. One reason why we've not done that so far is that melodic improvisation provides fairly straightforward quantitaive measures of creativity. Also, we've used both professional musicians and amateurs, and amateurs may well be able to produce meaningful melodic improvisation, but cannot be expected to exhibit the exquisit control over expressive performance variables that professional musicians have.

Hi Guy,

Thanks for your note. Your research sounds fascinating. I look forward to reading your final paper and findings. And let me know if you need a professional musician test subject. 

Best wishes for continued success.

-Liana

I'd be curious to know what the quantitative measures are of creativity with melodic improvisation. As a composer, performer, and computer programmer, it's hard for me to imagine it being straightforward. If the measure is of brain activity, I couldn't speak to that - but if it's of musical output, creativity would be extraordinarily difficult to measure quantitatively.

Hello Guy,

I teach classical violin majors and minors, and Violin Pedagogy at NYU String Studies. One of the most profound effects on the musical and technical development of my students is Creative Ability Development by Alice Kanack, a colleague of mine for 30 years. Her method of creative ability via Improvisation for classical musicians (not blues or jazz necessarily) is one of the most enlightening and groundbreaking methods I have seen to date. She is currently involved with brain scientists in the U of Rochester area. She would be a good source for this study.

Hi Ron,

Yes, extraordinarily difficult indeed. We would certainly like to have a quantitative measure of creativity in melodic improvisation, and have discussed it from time to time, but the scope of the problem (particularly with regard to defining the context, i.e. the reference by which something is regarded as creative), has kept us from making any serious attempts. So far, we've only measured melodic complexity, using 0-order melodic entropy (considering the distribution of single notes), 1-order melodic entropy (considering the distribution of bigrams of two consecutive pitches), and the Lempel-Ziv complexity measure.