Watching Music Train the Brain

Mar 23 2009 Published by under Neuroscience

First, I realize that I forgot to do my list of Pros and Cons for the last Journal Club option. So here they are:

  • In the media lately, and that sort of thing tends to get a good reception
  • The first author is the head of NIDA, which tends to carry some weight


  • n=10, for a human study, it's not a BAD n, but it's not that great either.
  • The study has been interpreted as possibly more meaningful than it really is.
  • Since the study does have problems, it will be harder to give a good presentation of.

And now, on to the second option. I'm not sure how many of you may remember my previous post on music and the brain, but there are lots of hypotheses out there on how music improves certain aspects of brain function. And now, for the first time, researchers actually showed that musical training can change the way the brain develops, not just in terms of eventual function, but in its very structure. Hyde, et al. "Musical training shapes structural brain development" Journal of Neuroscience, 2009.

There have been a pretty good number of studies which compare adult musicians and adult non-musicians for both function and structural differences. Many of these studies have found some very distinct changes, both in motor function (musicians tend to be faster and more accurate typists and perform better on fine motor skills tests than non-musicians), as well as structural differences. Musicians tends to have bigger brains areas devoted to sensorimotor skills and sensory integration. So you'd think that maybe being a musician can improve certain areas of the brain.
But there's a problem with these studies. In looking for "musicians", researchers tend to look at PROFESSIONAL musicians, who have clearly been practicing for years of their lives. Not only that, it's very possible that these musicians come from families with musical talent, or could have some sort of natural tendency toward music, self-selecting them into the musician pool and skewing the results. Not only that, these studies looked at adults, when the changes were already present. Were the changes the result of musical ability and learning? Or did musicians merely naturally already have these changes? Are the behavioral and brain structure results of musical training due to nuture of musical ability, or is there some nature involved?
To resolve these problems, the current study used children. The groups of children could be randomly divided into those who took music lessons and those who did not, and the researchers could look at their brains before and after training to see if changes had taken place that were not there before, allowing them to more concretely identify which changes could be occurring as a result of musical training.
So the authors took a group of 30 public school kids in 1st grade, and randomly assigned them to "music" or "non-music" groups, giving them a battery of tests beforehand to ensure the groups were evenly matched. Those in the music group were assigned to 30 minutes a week of private keyboard lessons, while those in the non-music group were not. 15 months later, the children were tested again on the same battery of tests, as well as subjected to something called a "brain deformation-based morphometry analysis", where they were given an MRI before and after training, and the MRIs were compared for differences in brain shape.
The behavioral results they found weren't surprising and have been documented before. The group assigned to private music lessons did have a higher socioeconomic status (which I think is kind of a flaw, maybe next study they could normalize for that and get funding to subsidize lower SES kids of lessons?), but at the beginning there were no other differences between the groups. After 15 months of music lessons, however, the music group showed significant improvements in fine motor skill tasks, performing much better than the control group. They also performed better on melody/rhythm tasks. The researchers did not find any evidence of far transfer (significant improvements in things like vocabulary), but that's not too surprising, as the music learning period was really pretty short, and far transfer seems to be a pretty difficult thing to get anyway. It's also very possible that the kids didn't practice enough. 🙂
They also found significant changes in brain morphology in children who had been learning an instrument. Both groups of children showed changes in brain size and development over the timing of the study, but the music group showed significant increases in size over the control group in brain areas like the precentral gyrus and the fourth and fifth segment of the corpus callosum, areas that are known to be connected with motor planning and execution. They also showed significant increases in the lateral part of Heschel's gyrus, an area association with auditory processing.
And now Sci's going to grab some of the figures. They do a great job of showing where in the brain they are talking about, as well as including some of the correlations.

Here you can see the increase in activity in size shown in the right precentral gyrus, which is functionall part of the primary motor cortex, which is responsible for planning and executing movement.

And HERE you can see changes in the volume of the right Heschl's gyrus, which is parimarily associated with auditory processing. For the rest of the figures, you gotta read the paper your ownself.
The coolest part was when the authors plotted each individual child's behavioral test scores against the changes found pre- and post- musical training. They were able to correlate the increase in fine motor skills in the left hand with comparable increases in the size of the right precentral gyrus (in terms of movement, the right side of the brain controls the left side of the body, so this makes sense). Not only that, the improvements in melody and rhythm testing could be correlated with changes in the auditory cortex, though melody performance did NOT correlate with changes in motor areas, and motor performance did not correlate with changes in auditory areas.
These findings are really pretty neat. They are the first to look at changes in both brain structure and behavior as a function of music training. Not only that, they are the first to show that this can happen in a REALLY short time. 15 months of music training really isn't a lot, and so it was impressive to show such dramatic changes in brain morphology over such little time. Of course, we can't know about the microstructure changes taking place, whether the growth represents more neurons, or more connections, or more glia, or all of the above, and though further work with high resolution imaging could help differentiate in changes between gray and white matter, it will probably be a while before we could look at changes in microstructure in humans.
The authors were very careful to point out that you can't take the results of the study TOO far. While the improvements after only 15 months of training were pretty impressive, only a natural inclination (or parental involvement) was going to keep the kids practicing. So while music lessons may change brain morphology, it's not the only thing that going to make your kid into the next Itzhak Perlman. There also has to be a drive to KEEP practicing, and of course a real love of music. Still, the study definitely shows that music lessons can improve fine motor movement, and the authors suggest that this knowledge could be used to help those naturally have problems with fine motor control, such as children with some developmental disorders, or even adults with neurobiological problems.
Finally, the pros and cons of this study:

  • An interesting subject, especially to me, and brain morphometry is an interesting technique
  • The study is awfully well controlled for, especially for being a human study. It's also extremely well-designed.
  • The findings are very interesting, and viscerally reinforce how behavior can change the very shape of your brain


  • The audience may not find the topic as interesting as Sci does.
  • The imaging part is rather difficult to explain


Gratuitous Izhak Perlman. I'll take any excuse. He's just fantastic!
Hyde, K., Lerch, J., Norton, A., Forgeard, M., Winner, E., Evans, A., & Schlaug, G. (2009). Musical Training Shapes Structural Brain Development Journal of Neuroscience, 29 (10), 3019-3025 DOI: 10.1523/JNEUROSCI.5118-08.2009

9 responses so far

  • leigh says:

    seriously, a con is explaining the imaging? come on sci, explaining things is what you do 🙂 i, for one, really like this paper.

  • JLK says:

    One of the faculty at my former MRU is a pretty major researcher in this area. I tried to find the lab page for him because I swear I remember some really cool stuff on it, but I couldn't scrounge it up for you.
    But if you're really interested in this area, I can give you his name so you can get his articles and what not.

  • Johan Richter says:

    "The group assigned to private music lessons did have a higher socioeconomic status (which I think is kind of a flaw, maybe next study they could normalize for that and get funding to subsidize lower SES kids of lessons?), but at the beginning there were no other differences between the groups."
    I don't understand. Do they mean that this just happened to be the case (since the assignment was random) or are you talking about the kids who took music lessons outside of the study?

  • Scicurious says:

    Johan: I'm sorry, I started that badly. They basically grabbed 15 kids who started taking private music lessons and 15 not, from the same public school. The parents had to pay for their own music lessons, so they tended to have a higher socioeconomic status. I personally think the better control would be for the researchers to pay for lessons for half the kids so they could control for the socioeconomic status of the kids from the outset. But oh well.

  • AK says:

    What comes to my mind immediately is the possibility of developing integrated training in music, puzzles (problem solving and tactics), keyboard and other manual skills, perhaps even hand/eye coordination. Such training might provide great improvement in ultimate adult skills. (By integrated, I mean that, perhaps, part of implementing solutions to puzzles includes specific musical performance.)

  • I think a better critique than being wishy-washy about their n is to ask what the power of the study is. You can say "10 isn't bad for a human study" but that's purely subjective unless you've set a beta, isn't it, sweet Sci?

  • David says:

    I know the title says "structural", but I wish they did some electrophysiology! And it would have been cool to see whether this result holds if you compare against controls that have done some sort of cognitively demanding task, rather than doing nothing. Like a second language (also auditory), or perhaps a sport, or drawing class. I'd be a horrible reviewer.

  • Maria says:

    I wonder if any studies have been done on the effects of singing on the brain - singing requires timing for example. I know there are studies of the musicality of 'motherese' for example and this seems to universally attract babies supporting language development as well as being so emotionally positive. There are a lot of problems with language development in the UK, parents don't seem to sing lullabies anymore - and sometimes don't seem to talk to their children much it might be interesting. Sorry, is this turning into a rant?!!

  • Barbara says:

    I recently came across your blog and have been reading along. I thought I would leave my first comment. I don't know what to say except that I have enjoyed reading. Nice blog. I will keep visiting this blog very often.

Leave a Reply