Please accept this repost today, originally posted on Jan 31, 2011.
Sci will admit I spent most the time "preparing" for this post by listening to LOTS of music.
This is your brain:
Is this your brain on Music?
Well, to be entirely honest...probably not. But music's still nice. Let's take a look at why.
So, let's start out with a little bit of a musical "high":
(ahhhhhh, that's the stuff)
Salimpoor, et al. "Anatomically distinct dopamine release during anticipation and experience of peak emotion to music" Nature Neuroscience, 2011.
Whenever I do outreach to kids in schools about drug research and drugs in the brain, we end up talking about "natural" highs. Pleasurable things like food and sex, and as I told them, rock and roll. I knew there had been previous studies out there showing that music activates pleasure and reward-related regions of the brain. This one is by no means the first, though it is the first to show a nice time correlation with specific brain area activation, and the first to really make certain that dopamine was the culprit. Nevertheless, this paper lit up the internet, because nothing sounds as good as the phrase "music gets you high!".
The idea is this: humans find a lot of things pleasurable. This is necessary, because if you don't find something pleasurable, you don't DO it, and things like eating and sex, those are a little necessary. When you need to feel pleasure in something you're doing, enter your mesolimbic dopamine system (I've written about dopamine pretty extensively before, you can check up on my previous work on it here). To sum up, your mesolimbic dopamine system starts with cell producing dopamine in an area called the ventral tegmental area, located deep in the center of your brain. These cells project, and send dopamine signals to, other areas, including an area called the ventral striatum, containing the nucleus accumbens.
(Dopamine system is in Blue, image is from NIDA)
The nucleus accumbens is mostly studied for the way dopamine signals within it change in response to drugs like cocaine or amphetamine. But dopamine in the ventral striatum is important for more than just drugs, it's also important for natural rewards like food and sex.
To see how much of an effect (and in what time the effect worked) music has on the brain, the authors of this study recruited people who responded strongly to music. Sci would have KILLED to be in this study, I get the music "chills" they were looking for, pick me!!
They took people who got "chills" when listening to music, and unlike other studies, they had them bring their music IN. Previous studies usually use strongly emotional music, but not music that specific participants prefer. I think this was a big step in the right direction for this study, musical tastes are very individual. The subjects brought in mostly classical, but also movie soundtracks, folk music, techno, and heavy metal. The lab that conducted the study actually cataloged all the music, and has it listed on their website!
They then studied the subjects using PET imaging and MRI while listening to either their favorite music, or music that was neutral (usually a selection that someone else brought in). PET, or positron emission tomography, uses a radioactive tracer (don't worry, it's very short lived) injected into the blood. These tracers can be specific for various things, but in this case they were interested in labeling dopamine, so they used a chemical called [11C]raclopride. Raclopride hits specific dopamine receptors in the brain (D2 receptors) where it is an antagonist. When you radiolabel raclopride and put it into the blood, it will collect in tissues like the brain, and start hitting D2 receptors, glowing all the while so the scanner can pick it up. Then, when you stimulate dopamine (in this case, using music), the dopamine will COMPETE with the raclopride for the D2 receptors. The competition will cause the raclopride to be displaced, and the amount of displacement correlates with the amount of dopamine you've got. It's a pretty neat way to study what's actually going on with dopamine in the brain.
And it's even better when you can combine it with something like MRI. MRI, or magnetic resonance imaging, is a technique which uses the brief magnetization of atoms in your body (oh yes, we can do that) to image your tissues in a VERY detailed manner. When you do this in the brain, you end up with a detailed image of the anatomy.
So now you have PET to determine how MUCH dopamine signaling you've got, and you've got MRI to determine exactly WHERE that signaling is taking place. The authors also took skin and heart rate and blood pressure measurements, to determine when the "chills" were happening as people listened to the music. They also asked the subjects whether they were experiencing "chills" and when, and how much they were enjoying themselves.
Time for pretty pictures.
You will notice on the right that all the correlation bars (between "neutral" and "chills" conditions) are going DOWN. That looks bad, but in fact that's RIGHT. Remember, you're looking for the raclopride binding to be DISPLACED by the dopamine, so you WANT it to go down. They got significant changes in the caudate and putament (part of the striatum, usually considered the dorsal part, meaning the top), and the nucleus accumbens, which is considered the ventral striatum. The changes here aligned nicely with the blood pressure and heartrate measures, as well as with when the participants said they got the "chills".
The authors then decided to go the extra research mile. They asked the participants to press a button when they got the chills. They they looked at that time point specifically in the MRI, as well as the 15 seconds leading up to it.
Take a look at panel C there. What you're looking at now is the change in signal (so going up is good, this time). You can see the big spike at the end with the nucleus accumbens correlates with the "chills", while the time leading up to it got a lot more signal from the caudate. Throughout the whole experiment, the number of "chills" the participants got was correlated with the activation specifically in the nucleus accumbens, which suggests that the intense pleasure we get from them is due to dopamine signals in this area.
The whole things shows that music can produce a nice increase in dopamine in the nucleus accumbens. Whether that's a "high", is another question. They couldn't really quantify what kind of a signal they were getting from the dopamine, other than that they got a significant change. Drugs like cocaine produce increases in dopamine in the nucleus accumbens of up to 300% of baseline, and drugs like meth can go even higher than that. Was intense pleasure when listening to music in the same category? Probably not. But that doesn't mean it doesn't feel good.
The whole study gives us a nice biological basis for our physical responses to music, but it also raises questions. WHY have be evolved such that music effects us this way? What is the function? Is it just enhancement of emotion? If so, how does that work? Is it familiarity with the music, knowing that a part you like is coming up? Does it have anything to do with language and the tones which we utilize in our voices for things like language?
We don't have these answers yet, but maybe someday we will. Until then, when you're listening to some great music and get the chills, you know what's happening.
Salimpoor VN, Benovoy M, Larcher K, Dagher A, & Zatorre RJ (2011). Anatomically distinct dopamine release during anticipation and experience of peak emotion to music. Nature neuroscience, 14 (2), 257-262 PMID: 21217764
I have to end this post by providing a bunch of Youtube links. These are all pieces of music that give Sci MAJOR chills, and possibly my Dopamine is going all wonky right now!
Let's start it off right:
(oooh, 1:21. Brrrrr.)
Or if that's not quite your speed:
(I tried to find the Benedictus from Schubert's Mass in G, but I couldn't find one Soprano soloist I liked, let alone a conductor)
And then there's this one:
And finally, this one:
And I'd like to open it up in the comments: what music has given YOU chills? It doesn't have to be classical (it is for me, but I know it ain't for everyone). Sci wants to hear!