Confirmation Bias and the Art of Music

In blog posts lately I’ve been noting confirmation bias and its role in determining how teachers and players determine the best pedagogy or best way to practice.  My personal example of fooling myself into thinking I could accurately predict a player’s embouchure type by looking at their anatomy alone is one example of what I mean by this.  Crunching the numbers showed that very few of the physical characteristics I thought would be helpful predictors turned out to be statistically significant.  There are plenty of other examples of how our biases can even change how we perceive the exact same performance.  Science itself is a process which strives to distance ourselves from confirmation bias and control for it in such a way that we don’t fool ourselves.

But there is another side to this discussion that I haven’t really written about too much before, how confirmation bias affects the musical perceptions and enjoyment of the art music making.  Writing for the Scientific American blog, Samuel McNerney explores this topic.

Let’s observe music, a popular topic in the psychology world. One of the common themes to emerge from the literature is the importance of patterns, expectations, and resolutions. Many authors argue that enjoyable music establishes a known pattern, creates expectations, and resolves the expectations in a predictable way. As neuroscientist Daniel Levitin, author of  This is Your Brain on Music explains, “as music unfolds, the brain constantly updates its estimates of when new beats will occur, and takes satisfaction in matching a mental beat with a real-in-the-world one.” This is one reason we repeatedly listen to the same songs and bands, we know exactly what we are going to get, and love it when they fulfill our preconceived expectations.

Which leads me to think a little bit about the value placed on surprise in some musical styles.  In a recent lecture about bebop in one of my classes I discussed how improvisers would often unexpectedly change directions and this was a quality that was prized among fans and musicians performing in this style.  While many listeners found bebop improvisations jarring and too random for their tastes, a smaller percentage of fans grew to expect surprise in their music.  Or perhaps a better way to phrase this is that the bebop aficionados got better at predicting the changes of patterns and learned to expect the unexpected.

Jonah Lehrer wrote about this same phenomena (which I’ve blogged about before).  In discussing a study conducted using fMRI scanners Lehrer noted the researchers found that the brain activity associated with pleasure were lighting up well before the expected musical event.  It wasn’t so much the actual forecasted phrase that was found to be pleasurable, but instead enjoyment came from the anticipation.

However, it’s worth considering then what happens when we don’t get the confirmation we expect.  McNerney noted:

History has shown that this can get ugly. Some music performances defied expectations so dramatically that the audience resorted to rioting. Famous examples include performances of Béla Bartók’s The Miraculous Mandarin, Steve Reich’s Four Organs and Igor Stravinsky’s Rite of Spring. In each of these cases, the composers forced the audience to listen to exactly what they didn’t want to hear. It would be like if a Democrat was forced to watch Fox or a Republican went to a Glen Beck rally and heard him praise Obama. I am sure that both of these scenarios would provoke reactions similar to the ones that Stravinsky experienced.

So while we don’t want the completely unexpected in our music (or else it would be boring), we tend to require that our predictions get confirmed to a certain degree, albeit often from a roundabout or delayed manner.  Returning to Lehrer’s article, he notes how Beethoven’s String Quartet in C sharp minor, Op. 131 sets up a rhythmic and harmonic pattern in the 5th movement, only to merely hint at this until the very end of the movement.  The pleasure we get from hearing the anticipated pattern finally happen is made all the more profound for the delayed gratification we get from it.  But the expected resolution does come, or else we would be left feeling wanting.

What does this mean for not only our personal enjoyment and appreciation of music as well as the art that society at large accepts?  McNerney noted that seeing past our confirmation bias sometimes involved not just delaying our gratification until the end of the piece, such as in the Beethoven string quartet, but looking even further into the future, when our expectations and biases have changed due to enculturation.

Unfortunately, our audio and visual systems are programmed to look for art that we like, and to ignore art that we don’t like. And this is what makes artistic innovation so difficult. But when we turn off our confirmation bias, we realize that watching or listening to something that doesn’t fulfill our expectations can be ultimately rewarding. All groundbreaking artists are in on this well kept secret: they know that in the end, it is just as enjoyable to experience something that violates an expectation, which is why they replaced the expected with the unexpected. In other words, they are the ones who saw through their confirmation bias.

Like most musicians, I’ve discovered that the more I studied music the more my tastes changed.  The pieces and recordings I’ve grown to love above others are those pieces of music where it seems I always notice something different every time I listen to it.  This is perhaps why I find jazz to be most enjoyable to me, every new performance ends up different.  I’m now biased towards having my expectations shaken up, rather than easily confirmed.

So the moral of the story is that becoming more aware of our biases and expectations doesn’t just make good sense for evaluating the way we teach and practice music, but also in how we create and consume music as an art.  Science and art can be two sides of the same coin and aren’t really the separate entities we frequently make them out to be.  Instead, they are essentially similar methods to exploring the our human nature and the world we live in that compliment and enhance each other.

Lyle Sanford

Over time I’ve increasingly begun to think that the dynamic you’re talking about with, ” the more I studied music the more my tastes changed”, is central to being a creative musician. It can seem really simplistic to say, but realizations – such as there is no one right way to play something, or that no piece is the best of a genre, or that no genre is better than others, (and that we can have different takes on all that at different times in our lives) are what open us to deepening our ability to play better and listen better. It’s like that Dylan line, “he not busy being born is busy dying” (or for musicians, getting burnt out).

Your point about art and science being two sides of the same coin is also a very good one. My idea is that over time the neuroscience is going to lessen that dualism.

By the way, in a previous post on embouchures you said something like science may never settle the issue – but my thought is that it might. Back in the 60’s and 70’s the idea that science could prove out the usefulness of music as a therapeutic modality seemed far fetched – but it’s happening. If they can come up with some way of imaging the way the musculature is working, and then tie that into brain imaging, there will be a lot more hard data to use in talking about embouchure.

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.