Thursday, 20 October 2016

The Happy Song

Music psychology research at Goldsmiths, University of London has led to concrete results:  thefirst song scientifically tested to make babies happy.”  While of course nothing is guaranteed -- for instance if your child is miserably teething, it’s entirely possible that nothing will make him happy -- thousands of parents gave their input as to the sounds that make their babies happy, and then 56 children were tested on their responses to different versions of the song.

Babies cannot tell us in words whether they like a piece of music, but scientists were able to gauge their reactions based on heart rate, facial expressions, and the sounds and movements the babies made while listening to the music.  One of the responses that was not measured was brain activity, maybe because hooking a baby up for an EEG is a big effort, and maybe because there aren’t well-validated EEG signals of infant happiness to compare with. 

But on the same day that “The Happy Song” showed up in my news feed, I also saw this post:  The only way an algorithm can tell if you really like a song is by scanning your brain.” It reports on a small-scale mobile EEG system designed to assess how well people like a piece of music – hook it up to your computer and you would never need to hit “like” again.  It would be interesting to take a mini-EEG system like this, test babies’ responses to the Happy Song, correlate the EEG responses with the physiological and behavioural responses, and then have a neurological-based system to test what music the babies like.  But then again, babies’ smiles and laughter are pretty good indicators too.

You can listen to The Happy Song on Youtube (and below).  More details at

Tuesday, 1 March 2016

Finger movements and tension

Try this: hold up your hand with your fingers extended, and then turn it so you’re looking at it side-on, with all your fingers lined up. Now bend your middle finger, trying to keep the others standing straight up. What you probably see as you move your middle finger is that your ring finger also moves out of place, no matter how much your try to keep it still. Your index finger probably moves a little too. Don’t get too stressed if it’s hard to move just your middle finger: our fingers are designed to move together. The thumb moves quite independently, but when we move each of the long fingers, other fingers also try to move.

When flexing one finger, the other fingers move out of place too.

There are two main reasons that this is true: first, each finger is connected to adjacent fingers by skin webbing, tendons, and muscle fascia, so that when we move one finger, it pulls on its neighbours. The second and more relevant reason is that our muscle control of our fingers is not completely separate.

You might think that motor control of our fingers works like this: a certain area of our brain is responsible for each finger. When that part of the brain is active, a message gets sent to the muscle responsible for that finger, which causes that finger to move. In this simple scenario, there is a straight line connection between 1 brain area, 1 muscle, and 1 finger. Unfortunately, this is not how finger control works.

The neurons that send motor commands to our finger muscles are located in the hand area of the primary motor cortex. But there is not a separate area for each finger. Studies looking at which neurons control which finger movements have found that the neurons that control the separate fingers are all kind of jumbled together. Also, a lot of the neurons that were studied were found participate in the control of more than one finger.  

This idea of mapping neurons onto finger movements gets even more complicated when you look at the muscles that move our fingers. The muscles responsible for flexing (bending) and extending (straightening) our fingers are located in our forearms, and connect to the finger bones through tendons which go through our hands. But there are not separate sets of muscles for each finger.

In fact, there are only two muscles that control flexing the four long fingers: the flexor digitalis profundus and the flexor digitalis superficialis. Each of these muscles has four tendons, one connecting to each of the long fingers of the hand. The muscle has different sections (known as muscle bellies) that each primarily activate one finger, but they are not completely separate, and they also often contract together, since the neurons that activate the muscles don't act independently either.
Similarly, to extend the fingers, there is one main muscle:  the extensor digitorum communis, involved in extending all four long fingers. There are also two smaller muscles:  the extensor digitali minimi (extends the little finger) and the extensor indicis (extends the index finger).

If we don’t have separate muscles controlling each finger, how do we move them independently?   The most likely possibility is that we move our fingers individually due to the combined action of more than one muscle, each of which has different amounts of control over each finger.  This conclusion comes from the lab of Marc Schieber at the University of Rochester, who showed in 1995 that when monkeys move their fingers individually, more than one muscle is active, and the amount of muscle activity in each muscle is different for the movements of each finger. 

In other words, when we try to move one finger independently, we don’t just contract a muscle to move that finger, we also have to contract other muscles to prevent the other fingers from moving. For example, raising the ring finger while keeping the middle finger still is difficult, because these fingers not only share a common extensor, but they are also joined mechanically. In order to raise just the ring finger, you need to activate the flexor going to the middle finger (in order to stabilize it), and at the same time activate the common extensor. It’s also really handy to extend the little finger at the same time, since this will pull on the ring finger and help it extend. See how complicated that is? No wonder moving our fingers individually is difficult.

Also, no wonder we can build up a lot of tension in our hands while playing musical instruments. It’s very common for children to play with their hands very tense. By maintaining activity in both the flexor and extensor muscles of the fingers, they can stabilize their finger joints and prevent unwanted movement. This strategy of co-contraction does work, up to a certain point. The problems very quickly outweigh the advantages: with too much tension in the hand, moving each finger is more work, and therefore quick and fluent movements of the fingers are impossible.

Here is a video of a young boy whom I teach who clearly has excess tension in his hand.  Notice how his fifth finger is extended as he plays.


This child’s problems with tension get worse as he tries to increase the tempo of his pieces. He is absolutely unable to play fast passages. This is typical of students with tension issues. Worse, excess tension leads to fatigue, pain, and eventually injury. Obviously, this situation is something to be avoided.

What’s not obvious is the best way to help students who have too much tension. When I do an on-line search, most of the advice (as in this article) is to re-align body and hand position, and then play simple exercises while focusing on mental awareness of tension in the hand, wrist, and arm. 

The good news is that our control over individual fingers is not set in stone:  we can learn to use different muscle strategies to move our fingers independently (as suggested in Winges & Furuya, 2015; Semmler et al.,  2004). This is part of musical training:  all those Hanon exercises for the piano (and Moyse studies for the flute, etc.) are designed to help us learn to efficiently control our finger muscles so that we can move fluently between different notes.

If you’ve had success in minimizing hand tension (either for yourself or your students), I’d love to hear about it in the comments.


Schieber, M.H. (1995). Muscular production of individuated finger movements: the roles of extrinsic finger muscles. J. Neurosci. 15, 284–297.

Schieber, M.H. (2002). Motor cortex and the distributed anatomy of finger movements. Adv. Exp. Med. Biol. 508, 411–416.

Semmler, J.G., Sale, M.V., Meyer, F.G., and Nordstrom, M.A. (2004). Motor-unit coherence and its relation with synchrony are influenced by training. J. Neurophysiol. 92, 3320–3331.

Winges, S.A., and Furuya, S. (2015). Distinct digit kinematics by professional and amateur pianists. Neuroscience 284, 643–652.

Friday, 20 November 2015

Get moving: exercise improves motor skill learning

What if I told you there’s a way to improve your practicing efficiency and get healthier at the same time?  It’s true:  recent research has shown that moderate exercise right before learning a motor skill improves learning.

The study, published this month in the journal PLoS ONE, had three groups of people learning a motor skill task that involved using a force transducer to move a cursor through a kind of maze on the screen.  One group had to run on a treadmill for 30 minutes right before learning the task, the second group also ran on the treadmill but then got a one-hour break before doing the maze task, and the third group had a leisurely walk instead of exercising.

The researchers found that the people in the running group were the best learners, while the people who just walked or who ran and then rested learned at about the same rate.  The researchers actually looked at two different kinds of information about how the people learned:  the number of errors they made, and the speed at which they completed the task.  These two aspects of performing a motor skill are usually inversely related.  Think about playing a complicated piece of music:  if you play it slowly, you will make fewer mistakes.  If you play it quickly, you’ll make more mistakes.  This is known as the speed-accuracy trade-off, and it applies to pretty much everything we do.  So does exercising before learning alter the speed or the accuracy of the motor skill, or both?

The study found that the speed at which people moved the cursor on the screen was not affected by exercising.  Instead, people who exercised right before the learning the motor skill made fewer errors than those who didn’t exercise.  In other words, exercising increased accuracy but not speed of the motor skill.

Why does exercising help learning?  It’s been known for a while that, in general, exercise increases neuroplasticity – the capacity of the brain to change.  It’s thought that exercise increases the levels of hormones and growth factors that foster the chemical changes underlying learning.  But most of the previous studies have looked at the effect of exercise on declarative memory, the memory for facts and events, rather than motor learning.  Here we can see that exercise also boosts motor skill learning, the kind of learning that we do when we sit down to practice our musical instruments. 

So you might consider going for a run or a bike ride before your daily practice session.  We all know exercise is good for us, but it’s also good to know that it’ll help us learn our scales and pieces better.