What a chill down your back is doing

There is a particular moment in a piece of music when something gives. The voice goes up. The note holds longer than expected. The string section enters under the singer for the first time in the song. A particular chord lands. The hair on your arms stands up. There is a chill down your back, or across your scalp, or somewhere in the chest that has no specific anatomical name. The whole body briefly responds.

Different people get this in different places. Some people get it from Whitney Houston holding a note. Some people get it from a quiet horn entrance at the end of a Wagner act. Some people get it from a sample they have heard ten thousand times that still does it. The trigger is personal. The response is shared.

A team at McGill University took this response into the lab and made it visible.

The 2011 study

The work was led by Valorie Salimpoor in Robert Zatorre’s lab at the Montreal Neurological Institute and published in Nature Neuroscience in 2011. The paper is titled “Anatomically distinct dopamine release during anticipation and experience of peak emotion to music.” The title is unusually accurate. That is exactly what they found.

They asked subjects to bring in music that reliably gave them chills. Then they put the subjects through a combination of brain imaging techniques and tracked dopamine release in their reward circuits while they listened. Two findings came out.

The first finding was that yes, listening to chill-inducing music produced measurable dopamine release in the brain’s reward system. The same neurotransmitter that fires for food, sex, and cocaine fires for a violin entrance.

The second finding was more interesting. The dopamine release came in two distinct waves, in two anatomically different parts of the reward circuit. One wave fired in the caudate during the moments of anticipation, the few seconds before the peak. The other wave fired in the nucleus accumbens during the peak itself. Anticipation and experience were neurally separate.

What that means, briefly

Most popular writing on the reward system collapses anticipation and experience together. The Salimpoor finding showed they are distinct. The brain has one circuit for wanting and another circuit for getting, and music engages both of them on a schedule the music itself sets.

This is part of why music feels the way it feels. The chord is approaching. The chord arrives. The brain has been reading the music’s grammar the whole time, building up the prediction, and then collecting the resolution. Some of the pleasure is in the moment of resolution. Some of the pleasure is in the architecture that gets you there.

Zatorre and a number of follow-up researchers have proposed that this is the reason musical surprise works. A piece of music that resolves exactly as predicted is pleasant but flat. A piece that resolves in a slightly unexpected way produces a bigger dopamine response. The reward system is paying attention to its own predictions and responding to the gap between the prediction and the actual outcome.

This is the same machinery that makes humor work. The setup builds prediction. The punchline plays with that prediction. The laughter is the resolution of the gap.

The strange part

The strange part, worth sitting with, is that music is a fully abstract object with no nutritional, protective, or reproductive function. Evolutionarily speaking, the reward system was built to respond to things that matter for survival. A sequence of organized air pressure changes is not one of those things.

And yet. The same circuits that the brain uses to make sure you eat enough food to survive will, when a particular violinist plays a particular note in a particular way, light up as if something nutritionally important has just happened.

A 2001 paper by Anne Blood and Zatorre in PNAS showed something similar with PET imaging before the dopamine specificity was nailed down. A 2014 review by Matthew Sachs and colleagues in Frontiers in Psychology surveyed the broader phenomenology under the academic name frisson and confirmed that it appears across cultures, ages, and musical genres, though not everyone experiences it. About two thirds of people report getting chills from music at least sometimes.

What the research does not say is that this proves music is good for you, or that you should listen to more of it, or that the chills mean something deeper than themselves. The research mostly says: this is real, this is neural, this is measurable, this happens.

Why the strangeness matters

There is a habit, especially in writing aimed at making music feel important, of using findings like this as ammunition. Listen to more music! It’s like cocaine! Your brain loves it! This style misreads what the finding is doing.

The Salimpoor work is more interesting than that. It is showing that a fully cultural, fully learned, fully abstract object has the capacity to engage the deepest circuits in the brain. A piece of music written by a stranger you will never meet, performed by people who will never know you exist, can produce, in your specific body, a release of dopamine in two precisely timed waves.

That is the finding. The finding does not need to be turned into a recommendation. The finding is interesting on its own.

The next time the back of your neck registers a small chill at a particular moment in a song, that is your nucleus accumbens firing in response to a prediction your caudate made a second earlier. Both of those structures evolved for survival. Neither of them was supposed to care about a chord change.

Both of them do anyway.

If you’d like a few of these letters delivered each month, you can subscribe here.